id int64 0 190k | prompt stringlengths 21 13.4M | docstring stringlengths 1 12k ⌀ |
|---|---|---|
179,734 | from dataclasses import asdict
from datetime import datetime
from logging import getLogger
from typing import Optional
from urllib.parse import urlencode
import justext
import requests
from django.conf import settings
from django.contrib.auth.decorators import login_required
from django.http import HttpResponseBadRequest
from django.shortcuts import render
from django.views.decorators.http import require_http_methods
from justext.core import html_to_dom, ParagraphMaker, classify_paragraphs, revise_paragraph_classification, \
LENGTH_LOW_DEFAULT, STOPWORDS_LOW_DEFAULT, MAX_LINK_DENSITY_DEFAULT, NO_HEADINGS_DEFAULT, LENGTH_HIGH_DEFAULT, \
STOPWORDS_HIGH_DEFAULT, MAX_HEADING_DISTANCE_DEFAULT, DEFAULT_ENCODING, DEFAULT_ENC_ERRORS, preprocessor
from requests.exceptions import RequestException
from mwmbl.models import Curation
from mwmbl.search_setup import ranker, index_path
from mwmbl.settings import NUM_EXTRACT_CHARS
from mwmbl.tinysearchengine.indexer import Document, DocumentState, TinyIndex
from mwmbl.tinysearchengine.rank import fix_document_state
from mwmbl.tokenizer import tokenize
from mwmbl.utils import add_term_infos
def _get_results_and_activity(request):
query = request.GET.get("q")
if query:
# There may be extra results in the request that we need to add in
# format is ?enhanced=google&title=title1&url=url1&extract=extract1&title=title2&url=url2&extract=extract2
# source = request.GET.get("enhanced", "unknown")
titles = request.GET.getlist(f"title")
urls = request.GET.getlist(f"url")
extracts = request.GET.getlist(f"extract")
term = " ".join(tokenize(query))
# For now, we only support the Google source
additional_results = [
Document(title=title, url=url, extract=extract, score=100.0 * 2 ** -i, term=term, state=DocumentState.FROM_GOOGLE)
for i, (title, url, extract) in enumerate(zip(titles, urls, extracts))
]
results = ranker.search(query, additional_results=additional_results)
activity = None
else:
results = None
activity = Curation.objects.order_by("-timestamp")[:500]
return activity, query, results
def home_fragment(request):
activity, query, results = _get_results_and_activity(request)
response = render(request, "home.html", {
"results": results,
"query": query,
"activity": activity,
})
# Encode the new query string
if query:
new_query_string = urlencode({"q": query}, doseq=True)
new_url = "/?" + new_query_string
else:
new_url = "/"
response["HX-Replace-Url"] = new_url
return response | null |
179,735 | from dataclasses import asdict
from datetime import datetime
from logging import getLogger
from typing import Optional
from urllib.parse import urlencode
import justext
import requests
from django.conf import settings
from django.contrib.auth.decorators import login_required
from django.http import HttpResponseBadRequest
from django.shortcuts import render
from django.views.decorators.http import require_http_methods
from justext.core import html_to_dom, ParagraphMaker, classify_paragraphs, revise_paragraph_classification, \
LENGTH_LOW_DEFAULT, STOPWORDS_LOW_DEFAULT, MAX_LINK_DENSITY_DEFAULT, NO_HEADINGS_DEFAULT, LENGTH_HIGH_DEFAULT, \
STOPWORDS_HIGH_DEFAULT, MAX_HEADING_DISTANCE_DEFAULT, DEFAULT_ENCODING, DEFAULT_ENC_ERRORS, preprocessor
from requests.exceptions import RequestException
from mwmbl.models import Curation
from mwmbl.search_setup import ranker, index_path
from mwmbl.settings import NUM_EXTRACT_CHARS
from mwmbl.tinysearchengine.indexer import Document, DocumentState, TinyIndex
from mwmbl.tinysearchengine.rank import fix_document_state
from mwmbl.tokenizer import tokenize
from mwmbl.utils import add_term_infos
def justext_with_dom(html_text, stoplist, length_low=LENGTH_LOW_DEFAULT,
length_high=LENGTH_HIGH_DEFAULT, stopwords_low=STOPWORDS_LOW_DEFAULT,
stopwords_high=STOPWORDS_HIGH_DEFAULT, max_link_density=MAX_LINK_DENSITY_DEFAULT,
max_heading_distance=MAX_HEADING_DISTANCE_DEFAULT, no_headings=NO_HEADINGS_DEFAULT,
encoding=None, default_encoding=DEFAULT_ENCODING,
enc_errors=DEFAULT_ENC_ERRORS):
"""
Converts an HTML page into a list of classified paragraphs. Each paragraph
is represented as instance of class ˙˙justext.paragraph.Paragraph˙˙.
"""
dom = html_to_dom(html_text, default_encoding, encoding, enc_errors)
titles = dom.xpath("//title")
title = titles[0].text if len(titles) > 0 else None
dom = preprocessor(dom)
paragraphs = ParagraphMaker.make_paragraphs(dom)
classify_paragraphs(paragraphs, stoplist, length_low, length_high,
stopwords_low, stopwords_high, max_link_density, no_headings)
revise_paragraph_classification(paragraphs, max_heading_distance)
return paragraphs, title
def _get_curation(request, query, documents, reranked_documents):
curation_id = request.POST.get("curation_id")
reranked_document_dicts = [asdict(d) for d in reranked_documents]
if curation_id is not None:
curation = Curation.objects.get(id=curation_id)
curation.new_results = reranked_document_dicts
curation.num_changes += 1
curation.save()
else:
user = request.user
if not user.is_authenticated:
user = None
with TinyIndex(Document, index_path, 'r') as indexer:
tokens = tokenize(query)
term = " ".join(tokens)
original_index_results = [doc for doc in indexer.retrieve(term) if doc.term == term]
curation = Curation(
user=user,
timestamp=datetime.utcnow(),
query=query,
original_index_results=[asdict(d) for d in original_index_results],
original_results=[asdict(d) for d in documents.values()],
new_results=reranked_document_dicts,
num_changes=1,
)
curation.save()
return curation
def _insert_document(documents, approved_document):
reranked_documents = []
inserted_approved = False
for document in documents.values():
if document.url == approved_document.url:
continue
if (document.state is None or document.state < DocumentState.ORGANIC_APPROVED) and not inserted_approved:
reranked_documents.append(approved_document)
inserted_approved = True
reranked_documents.append(document)
if not inserted_approved:
reranked_documents.append(approved_document)
return reranked_documents
def _get_documents(request, term: str):
urls = request.POST.getlist("url")
titles = request.POST.getlist("title")
extracts = request.POST.getlist("extract")
states = request.POST.getlist("state")
scores = request.POST.getlist("score")
assert len(urls) == len(titles) == len(extracts) == len(states) == len(scores)
documents = {}
for url, title, extract, state, score in zip(urls, titles, extracts, states, scores):
try:
state_enum = DocumentState(int(state))
except ValueError:
state_enum = None
documents[url] = Document(
title=title, url=url, extract=extract, score=float(score), term=term, state=state_enum)
return documents
def _save_to_index(query: str, new_results: list[Document]):
with TinyIndex(Document, index_path, 'w') as indexer:
term = " ".join(tokenize(query))
documents = [
Document(
title=result.title,
url=result.url,
extract=result.extract,
score=MAX_CURATED_SCORE - i,
term=term,
state=result.state,
)
for i, result in enumerate(new_results)
if result.state is not None and result.state >= DocumentState.ORGANIC_APPROVED
]
page_index = indexer.get_key_page_index(term)
existing_documents_no_terms = indexer.get_page(page_index)
existing_documents = add_term_infos(existing_documents_no_terms, indexer, page_index)
new_urls = {doc.url for doc in documents}
other_documents = [doc for doc in existing_documents if doc.url not in new_urls]
logger.info(f"Found {len(other_documents)} other documents for term {term} at page {page_index} "
f"with terms { {doc.term for doc in other_documents} }")
# Update state for other documents
states = {doc.url: doc.state for doc in new_results}
for doc in other_documents:
doc.state = states.get(doc.url, doc.state)
all_documents = documents + other_documents
logger.info(f"Storing {len(all_documents)} documents at page {page_index}")
indexer.store_in_page(page_index, all_documents)
return {"curation": "ok"}
NUM_EXTRACT_CHARS = 155
class DocumentState(IntEnum):
"""
The state of the document in the index. A value of None indicates an organic search result.
"""
DELETED = -1
FROM_USER = 2
FROM_GOOGLE = 3
ORGANIC_APPROVED = 7
FROM_USER_APPROVED = 8
FROM_GOOGLE_APPROVED = 9
class Document:
title: str
url: str
extract: str
score: float
term: Optional[str] = None
state: Optional[int] = None
def __init__(self, title, url, extract, score, term=None, state=None):
# Sometimes the title or extract may be None, probably because of user generated content
# It's not allowed to be None though, or things will break
self.title = title if title is not None else ''
self.url = url
self.extract = extract if extract is not None else ''
self.score = score
self.term = term
self.state = state
def tokenize(input_text):
cleaned_text = clean_unicode(input_text)
tokens = cleaned_text.lower().split()
if input_text.endswith('…'):
# Discard the last two tokens since there will likely be a word cut in two
tokens = tokens[:-2]
return tokens
def add_url(request):
new_url = request.POST["new_url"]
query = request.POST["query"]
try:
response = requests.get(new_url, timeout=5)
except RequestException:
return HttpResponseBadRequest("Could not fetch URL")
paragraphs, title = justext_with_dom(response.content, justext.get_stoplist("English"))
good_paragraphs = [p for p in paragraphs if p.class_type == 'good']
extract = ' '.join([p.text for p in good_paragraphs])
if len(extract) > NUM_EXTRACT_CHARS:
extract = extract[:NUM_EXTRACT_CHARS - 1] + '…'
term = " ".join(tokenize(query))
result = Document(title=title, url=new_url, extract=extract, score=0.0, term=term, state=DocumentState.FROM_USER_APPROVED)
documents = _get_documents(request, term)
reranked_documents = _insert_document(documents, result)
curation = _get_curation(request, query, documents, reranked_documents)
_save_to_index(query, reranked_documents)
return render(request, "home.html", {
"results": reranked_documents,
"query": query,
"activity": None,
"curation": curation,
}) | null |
179,736 | from dataclasses import asdict
from datetime import datetime
from logging import getLogger
from typing import Optional
from urllib.parse import urlencode
import justext
import requests
from django.conf import settings
from django.contrib.auth.decorators import login_required
from django.http import HttpResponseBadRequest
from django.shortcuts import render
from django.views.decorators.http import require_http_methods
from justext.core import html_to_dom, ParagraphMaker, classify_paragraphs, revise_paragraph_classification, \
LENGTH_LOW_DEFAULT, STOPWORDS_LOW_DEFAULT, MAX_LINK_DENSITY_DEFAULT, NO_HEADINGS_DEFAULT, LENGTH_HIGH_DEFAULT, \
STOPWORDS_HIGH_DEFAULT, MAX_HEADING_DISTANCE_DEFAULT, DEFAULT_ENCODING, DEFAULT_ENC_ERRORS, preprocessor
from requests.exceptions import RequestException
from mwmbl.models import Curation
from mwmbl.search_setup import ranker, index_path
from mwmbl.settings import NUM_EXTRACT_CHARS
from mwmbl.tinysearchengine.indexer import Document, DocumentState, TinyIndex
from mwmbl.tinysearchengine.rank import fix_document_state
from mwmbl.tokenizer import tokenize
from mwmbl.utils import add_term_infos
def switch_state(state: Optional[DocumentState]) -> Optional[DocumentState]:
if state is None:
return DocumentState.ORGANIC_APPROVED
if state == DocumentState.FROM_GOOGLE:
return DocumentState.FROM_GOOGLE_APPROVED
if state == DocumentState.FROM_USER:
return DocumentState.FROM_USER_APPROVED
if state == DocumentState.FROM_GOOGLE_APPROVED:
return DocumentState.FROM_GOOGLE
if state == DocumentState.FROM_USER_APPROVED:
return DocumentState.FROM_USER
if state == DocumentState.ORGANIC_APPROVED:
return None
raise ValueError(f"Unexpected state {repr(state)}")
def _get_curation(request, query, documents, reranked_documents):
curation_id = request.POST.get("curation_id")
reranked_document_dicts = [asdict(d) for d in reranked_documents]
if curation_id is not None:
curation = Curation.objects.get(id=curation_id)
curation.new_results = reranked_document_dicts
curation.num_changes += 1
curation.save()
else:
user = request.user
if not user.is_authenticated:
user = None
with TinyIndex(Document, index_path, 'r') as indexer:
tokens = tokenize(query)
term = " ".join(tokens)
original_index_results = [doc for doc in indexer.retrieve(term) if doc.term == term]
curation = Curation(
user=user,
timestamp=datetime.utcnow(),
query=query,
original_index_results=[asdict(d) for d in original_index_results],
original_results=[asdict(d) for d in documents.values()],
new_results=reranked_document_dicts,
num_changes=1,
)
curation.save()
return curation
def _insert_document(documents, approved_document):
reranked_documents = []
inserted_approved = False
for document in documents.values():
if document.url == approved_document.url:
continue
if (document.state is None or document.state < DocumentState.ORGANIC_APPROVED) and not inserted_approved:
reranked_documents.append(approved_document)
inserted_approved = True
reranked_documents.append(document)
if not inserted_approved:
reranked_documents.append(approved_document)
return reranked_documents
def _get_documents(request, term: str):
urls = request.POST.getlist("url")
titles = request.POST.getlist("title")
extracts = request.POST.getlist("extract")
states = request.POST.getlist("state")
scores = request.POST.getlist("score")
assert len(urls) == len(titles) == len(extracts) == len(states) == len(scores)
documents = {}
for url, title, extract, state, score in zip(urls, titles, extracts, states, scores):
try:
state_enum = DocumentState(int(state))
except ValueError:
state_enum = None
documents[url] = Document(
title=title, url=url, extract=extract, score=float(score), term=term, state=state_enum)
return documents
def _save_to_index(query: str, new_results: list[Document]):
with TinyIndex(Document, index_path, 'w') as indexer:
term = " ".join(tokenize(query))
documents = [
Document(
title=result.title,
url=result.url,
extract=result.extract,
score=MAX_CURATED_SCORE - i,
term=term,
state=result.state,
)
for i, result in enumerate(new_results)
if result.state is not None and result.state >= DocumentState.ORGANIC_APPROVED
]
page_index = indexer.get_key_page_index(term)
existing_documents_no_terms = indexer.get_page(page_index)
existing_documents = add_term_infos(existing_documents_no_terms, indexer, page_index)
new_urls = {doc.url for doc in documents}
other_documents = [doc for doc in existing_documents if doc.url not in new_urls]
logger.info(f"Found {len(other_documents)} other documents for term {term} at page {page_index} "
f"with terms { {doc.term for doc in other_documents} }")
# Update state for other documents
states = {doc.url: doc.state for doc in new_results}
for doc in other_documents:
doc.state = states.get(doc.url, doc.state)
all_documents = documents + other_documents
logger.info(f"Storing {len(all_documents)} documents at page {page_index}")
indexer.store_in_page(page_index, all_documents)
return {"curation": "ok"}
class Document:
title: str
url: str
extract: str
score: float
term: Optional[str] = None
state: Optional[int] = None
def __init__(self, title, url, extract, score, term=None, state=None):
# Sometimes the title or extract may be None, probably because of user generated content
# It's not allowed to be None though, or things will break
self.title = title if title is not None else ''
self.url = url
self.extract = extract if extract is not None else ''
self.score = score
self.term = term
self.state = state
def tokenize(input_text):
cleaned_text = clean_unicode(input_text)
tokens = cleaned_text.lower().split()
if input_text.endswith('…'):
# Discard the last two tokens since there will likely be a word cut in two
tokens = tokens[:-2]
return tokens
def approve(request):
approve_url = request.POST.get("approve_url")
query = request.POST.get("query")
term = " ".join(tokenize(query))
documents = _get_documents(request, term)
# The approved Document should be pushed below the last Document with status > 0
# If there are no such documents, push it to the top
document_to_approve = documents[approve_url]
approved_document = Document(
title=document_to_approve.title,
url=document_to_approve.url,
extract=document_to_approve.extract,
score=document_to_approve.score,
term=document_to_approve.term,
state=switch_state(document_to_approve.state),
)
reranked_documents = _insert_document(documents, approved_document)
curation = _get_curation(request, query, documents, reranked_documents)
_save_to_index(query, reranked_documents)
response = render(request, "home.html", {
"results": reranked_documents,
"query": query,
"activity": None,
"curation": curation,
})
return response | null |
179,737 | from dataclasses import asdict
from datetime import datetime
from logging import getLogger
from typing import Optional
from urllib.parse import urlencode
import justext
import requests
from django.conf import settings
from django.contrib.auth.decorators import login_required
from django.http import HttpResponseBadRequest
from django.shortcuts import render
from django.views.decorators.http import require_http_methods
from justext.core import html_to_dom, ParagraphMaker, classify_paragraphs, revise_paragraph_classification, \
LENGTH_LOW_DEFAULT, STOPWORDS_LOW_DEFAULT, MAX_LINK_DENSITY_DEFAULT, NO_HEADINGS_DEFAULT, LENGTH_HIGH_DEFAULT, \
STOPWORDS_HIGH_DEFAULT, MAX_HEADING_DISTANCE_DEFAULT, DEFAULT_ENCODING, DEFAULT_ENC_ERRORS, preprocessor
from requests.exceptions import RequestException
from mwmbl.models import Curation
from mwmbl.search_setup import ranker, index_path
from mwmbl.settings import NUM_EXTRACT_CHARS
from mwmbl.tinysearchengine.indexer import Document, DocumentState, TinyIndex
from mwmbl.tinysearchengine.rank import fix_document_state
from mwmbl.tokenizer import tokenize
from mwmbl.utils import add_term_infos
class Curation(models.Model):
user = models.ForeignKey(MwmblUser, on_delete=models.CASCADE, null=True)
timestamp = models.DateTimeField()
query = models.CharField(max_length=300)
original_index_results = models.JSONField(default=list)
original_results = models.JSONField()
new_results = models.JSONField()
num_changes = models.IntegerField(default=0)
index_path = Path(settings.DATA_PATH) / settings.INDEX_NAME
class Document:
title: str
url: str
extract: str
score: float
term: Optional[str] = None
state: Optional[int] = None
def __init__(self, title, url, extract, score, term=None, state=None):
# Sometimes the title or extract may be None, probably because of user generated content
# It's not allowed to be None though, or things will break
self.title = title if title is not None else ''
self.url = url
self.extract = extract if extract is not None else ''
self.score = score
self.term = term
self.state = state
class TinyIndex(Generic[T]):
def __init__(self, item_factory: Callable[..., T], index_path, mode='r'):
if mode not in {'r', 'w'}:
raise ValueError(f"Mode should be one of 'r' or 'w', got {mode}")
with open(index_path, 'rb') as index_file:
metadata_page = index_file.read(METADATA_SIZE)
metadata_bytes = metadata_page.rstrip(b'\x00')
metadata = TinyIndexMetadata.from_bytes(metadata_bytes)
if metadata.item_factory != item_factory.__name__:
raise ValueError(f"Metadata item factory '{metadata.item_factory}' in the index "
f"does not match the passed item factory: '{item_factory.__name__}'")
self.item_factory = item_factory
self.index_path = index_path
self.mode = mode
self.num_pages = metadata.num_pages
self.page_size = metadata.page_size
logger.info(f"Loaded index with {self.num_pages} pages and {self.page_size} page size")
self.index_file = None
self.mmap = None
def __enter__(self):
self.index_file = open(self.index_path, 'r+b')
prot = PROT_READ if self.mode == 'r' else PROT_READ | PROT_WRITE
self.mmap = mmap(self.index_file.fileno(), 0, prot=prot)
return self
def __exit__(self, exc_type, exc_val, exc_tb):
self.mmap.close()
self.index_file.close()
def retrieve(self, key: str) -> List[T]:
index = self.get_key_page_index(key)
logger.debug(f"Retrieving index {index}")
page = self.get_page(index)
return [item for item in page if item.term is None or item.term == key]
def get_key_page_index(self, key) -> int:
key_hash = mmh3.hash(key, signed=False)
return key_hash % self.num_pages
def get_page(self, i) -> list[T]:
"""
Get the page at index i, decompress and deserialise it using JSON
"""
results = self._get_page_tuples(i)
return [self.item_factory(*item) for item in results]
def _get_page_tuples(self, i):
page_data = self.mmap[i * self.page_size + METADATA_SIZE:(i + 1) * self.page_size + METADATA_SIZE]
decompressor = ZstdDecompressor()
try:
decompressed_data = decompressor.decompress(page_data)
except ZstdError as e:
logger.exception(f"Error decompressing page {i}: {e}")
return []
return json.loads(decompressed_data.decode('utf8'))
def store_in_page(self, page_index: int, values: list[T]):
value_tuples = [astuple(value) for value in values]
self._write_page(value_tuples, page_index)
def _write_page(self, data, i: int):
"""
Serialise the data using JSON, compress it and store it at index i.
If the data is too big, it will store the first items in the list and discard the rest.
"""
if self.mode != 'w':
raise UnsupportedOperation("The file is open in read mode, you cannot write")
page_data = _get_page_data(self.page_size, data)
logger.debug(f"Got page data of length {len(page_data)}")
self.mmap[i * self.page_size + METADATA_SIZE:(i+1) * self.page_size + METADATA_SIZE] = page_data
def create(item_factory: Callable[..., T], index_path: str, num_pages: int, page_size: int):
if os.path.isfile(index_path):
raise FileExistsError(f"Index file '{index_path}' already exists")
metadata = TinyIndexMetadata(VERSION, page_size, num_pages, item_factory.__name__)
metadata_bytes = metadata.to_bytes()
metadata_padded = _pad_to_page_size(metadata_bytes, METADATA_SIZE)
page_bytes = _get_page_data(page_size, [])
with open(index_path, 'wb') as index_file:
index_file.write(metadata_padded)
for i in range(num_pages):
index_file.write(page_bytes)
return TinyIndex(item_factory, index_path=index_path)
def fix_document_state(result: Document):
try:
fixed_state = DocumentState(result.state)
except ValueError:
fixed_state = None
fixed_document = Document(result.title, result.url, result.extract, result.score, result.term, fixed_state)
return fixed_document
def tokenize(input_text):
cleaned_text = clean_unicode(input_text)
tokens = cleaned_text.lower().split()
if input_text.endswith('…'):
# Discard the last two tokens since there will likely be a word cut in two
tokens = tokens[:-2]
return tokens
def revert_current_curation(request):
curation_id = request.POST.get("curation_id")
curation = Curation.objects.get(id=curation_id)
query = curation.query
with TinyIndex(Document, index_path, 'w') as indexer:
term = " ".join(tokenize(query))
documents = [Document(**doc) for doc in curation.original_index_results]
page_index = indexer.get_key_page_index(term)
existing_documents = indexer.get_page(page_index)
other_term_documents = [doc for doc in existing_documents if doc.term != term]
# Replace all existing documents for the term with the original documents
all_documents = documents + other_term_documents
indexer.store_in_page(page_index, all_documents)
# Delete the curation
curation.delete()
original_documents_unfixed = [Document(**doc) for doc in curation.original_results]
original_documents = [fix_document_state(doc) for doc in original_documents_unfixed]
return render(request, "home.html", {
"results": original_documents,
"query": query,
"activity": None,
"curation": None,
}) | null |
179,738 | from dataclasses import asdict
from datetime import datetime
from logging import getLogger
from typing import Optional
from urllib.parse import urlencode
import justext
import requests
from django.conf import settings
from django.contrib.auth.decorators import login_required
from django.http import HttpResponseBadRequest
from django.shortcuts import render
from django.views.decorators.http import require_http_methods
from justext.core import html_to_dom, ParagraphMaker, classify_paragraphs, revise_paragraph_classification, \
LENGTH_LOW_DEFAULT, STOPWORDS_LOW_DEFAULT, MAX_LINK_DENSITY_DEFAULT, NO_HEADINGS_DEFAULT, LENGTH_HIGH_DEFAULT, \
STOPWORDS_HIGH_DEFAULT, MAX_HEADING_DISTANCE_DEFAULT, DEFAULT_ENCODING, DEFAULT_ENC_ERRORS, preprocessor
from requests.exceptions import RequestException
from mwmbl.models import Curation
from mwmbl.search_setup import ranker, index_path
from mwmbl.settings import NUM_EXTRACT_CHARS
from mwmbl.tinysearchengine.indexer import Document, DocumentState, TinyIndex
from mwmbl.tinysearchengine.rank import fix_document_state
from mwmbl.tokenizer import tokenize
from mwmbl.utils import add_term_infos
class DocumentState(IntEnum):
def _get_document_state(validated: bool, source: str) -> Optional[DocumentState]:
if validated:
if source.lower() == "user":
return DocumentState.FROM_USER_APPROVED
elif source.lower() == "google":
return DocumentState.FROM_GOOGLE_APPROVED
else:
return DocumentState.ORGANIC_APPROVED
elif source.lower() == "user":
return DocumentState.FROM_USER
elif source.lower() == "google":
return DocumentState.FROM_GOOGLE
else:
return None | null |
179,739 | import os
import pickle
from datetime import datetime
from pathlib import Path
from queue import Queue
from mwmbl.indexer import record_urls_in_database
def run_update_urls_on_fixed_batches():
with open(Path(os.environ["HOME"]) / "data" / "mwmbl" / "hashed-batches.pickle", "rb") as file:
batches = pickle.load(file)
# print("Batches", batches[:3])
queue = Queue()
start = datetime.now()
record_urls_in_database(batches, queue)
total_time = (datetime.now() - start).total_seconds()
print("Total time:", total_time) | null |
179,740 | from collections import defaultdict
from random import Random
from zstandard import ZstdCompressor, ZstdDecompressor
random = Random(1)
def test_compress_decompress():
error_type_counts = defaultdict(int)
for i in range(10000):
compressor = ZstdCompressor()
decompressor = ZstdDecompressor()
random_string = (''.join(random.choice('abcdefghijklmnopqrstuvwxyz') for _ in range(4096))).encode('utf8')
compressed_data = compressor.compress(random_string)
# Corrupt the data
compressed_data_array = bytearray(compressed_data)
# Corrupt some random bytes
for _ in range(10):
compressed_data_array[random.randint(0, len(compressed_data_array) - 1)] = random.randint(0, 255)
corrupted_data = bytes(compressed_data_array)
try:
decompressed_data = decompressor.decompress(corrupted_data)
except Exception as e:
error_type_counts[f"{type(e)}: {str(e)}"] += 1
continue
error_type_counts["none"] += 1
print("Counts")
for error_type, count in error_type_counts.items():
print(f"{error_type}: {count}") | null |
179,741 | import glob
import gzip
import json
from collections import defaultdict, Counter
from urllib.parse import urlparse
from mwmbl.crawler import HashedBatch
from mwmbl.indexer import CRAWL_GLOB, MWMBL_DATA_DIR
def get_urls():
for path in glob.glob(CRAWL_GLOB):
data = json.load(gzip.open(path))
batch = HashedBatch.parse_obj(data)
user = batch.user_id_hash
for item in batch.items:
if item.content is not None:
for url in item.content.links:
yield user, url
def analyse_urls(urls):
url_set = defaultdict(list)
domains = set()
for user, url in urls:
url_set[url].append(user)
parsed_url = urlparse(url)
path = parsed_url.path.strip('/')
if path == '':
domains.add(parsed_url.netloc)
count = sum(len(x) for x in url_set.values())
print("Root pages crawled", sorted(domains))
find_worst_pages(url_set)
print(f"Got {len(url_set)} URLs and {len(domains)} root pages from {count} items")
url_list_size = len(json.dumps(list(url_set.keys())))
print("Length of all URLs", url_list_size)
def run():
urls = get_urls()
analyse_urls(urls) | null |
179,742 | import multiprocessing
import os
import string
from multiprocessing.pool import ThreadPool
from pathlib import Path
from random import Random
from mwmbl.tinysearchengine.indexer import TinyIndex, Document
random = Random(2)
def random_string():
randomly_generated_string = "".join([random.choice(string.ascii_letters) for _ in range(500)])
return randomly_generated_string | null |
179,743 | import multiprocessing
import os
import string
from multiprocessing.pool import ThreadPool
from pathlib import Path
from random import Random
from mwmbl.tinysearchengine.indexer import TinyIndex, Document
random = Random(2)
indexer = TinyIndex(Document, str(INDEX_PATH), 'r')
indexer.__enter__()
def read_page(page_index: int):
def index_reading_writing_multithreading():
# with multiprocessing.Pool(processes=10) as pool:
with ThreadPool(processes=10) as pool:
print("Start")
pool.map_async(read_page, [random.randint(0, indexer.num_pages) for i in range(1000000)])
pool.close()
pool.join()
print("End") | null |
179,744 | import json
from mwmbl.indexer import TOP_DOMAINS_JSON_PATH
from mwmbl.hn_top_domains_filtered import DOMAINS
DOMAINS = {'blog.samaltman.com': 0.9906157038365982,
'paulgraham.com': 0.9900792629020208,
'blog.rust-lang.org': 0.9890328852259848,
'blog.ycombinator.com': 0.9883892917770221,
'ciechanow.ski': 0.9880439204029129,
'blog.mozilla.org': 0.9862851554386893,
'stripe.com': 0.9861322763760931,
'danluu.com': 0.9861047192293313,
'www.kalzumeus.com': 0.9850478020093707,
'signal.org': 0.9850307268861681,
'antirez.com': 0.9848258724546234,
'www.mozilla.org': 0.9848038469599045,
'jacquesmattheij.com': 0.9844261150401603,
'www.chris-granger.com': 0.9842666557846631,
'www.apple.com': 0.9836460584625046,
'letsencrypt.org': 0.9835295202763794,
'www.teslamotors.com': 0.9831965330638592,
'hacks.mozilla.org': 0.982862363147374,
'lore.kernel.org': 0.9822733167137851,
'frame.work': 0.9821010345456243,
'www.righto.com': 0.9819349251846876,
'www.troyhunt.com': 0.9814966052145125,
'www.sec.gov': 0.9814725718254595,
'www.tesla.com': 0.9810577172203161,
'keybase.io': 0.9808967126687541,
'www.blender.org': 0.9807600197929631,
'www.spacex.com': 0.9807226511637516,
'www.susanjfowler.com': 0.9805638280990112,
'www.eff.org': 0.9805327186535832,
'blog.golang.org': 0.9800638229086632,
'github.com': 0.9793809676855824,
'www.bloomberg.com': 0.9791747058863844,
'matrix.org': 0.9788189361330681,
'openai.com': 0.9781666403563757,
'www.nytimes.com': 0.9780992933501785,
'www.sublimetext.com': 0.9779900989360693,
'mtlynch.io': 0.9777886722980439,
'www.reuters.com': 0.9776899139875952,
'blog.pinboard.in': 0.9776638411070188,
'whispersystems.org': 0.9776203868069737,
'fabiensanglard.net': 0.9774982104911971,
'drewdevault.com': 0.9774354534940557,
'blog.interviewing.io': 0.9772072688889477,
'github.blog': 0.9770459926321585,
'calpaterson.com': 0.9768950731213937,
'rachelbythebay.com': 0.9768404166128064,
'bellard.org': 0.9767955350707707,
'twitter.com': 0.976728011118412,
'training.kalzumeus.com': 0.9766770275278582,
'www.npr.org': 0.976513349382135,
'jvns.ca': 0.9765007166886884,
'www.daemonology.net': 0.9763913310944178,
'ifixit.org': 0.9762565420103206,
'www.bbc.com': 0.9761524214719092,
'karpathy.github.io': 0.9759182558450648,
'tonsky.me': 0.9759061232371565,
'www.buzzfeednews.com': 0.975834047407925,
'deepmind.com': 0.9757513978763775,
'www.postgresql.org': 0.9756297562580343,
'krebsonsecurity.com': 0.9755023613643372,
'overreacted.io': 0.9754149079228597,
'stratechery.com': 0.9754126620517041,
'brave.com': 0.9752551626581946,
'lwn.net': 0.9751973038824019,
'about.gitlab.com': 0.9751871932295729,
'www.stephendiehl.com': 0.9751552345204195,
'techcrunch.com': 0.9751406961249492,
'raganwald.posterous.com': 0.9750289679638235,
'blog.discordapp.com': 0.9750045998850335,
'bugs.chromium.org': 0.9749443016112311,
'www.marco.org': 0.9749044929470128,
'words.steveklabnik.com': 0.9748959016378708,
'www.backblaze.com': 0.9744710342430156,
'robert.ocallahan.org': 0.9744576196125118,
'blog.cryptographyengineering.com': 0.9744312947315787,
'citizenlab.ca': 0.9742406063151361,
'www.propublica.org': 0.9742125433042081,
'blog.benjojo.co.uk': 0.9741593610657648,
'justine.lol': 0.9740984398105378,
'lists.zx2c4.com': 0.9739788216838905,
'googleblog.blogspot.com': 0.9739234991924623,
'www.supremecourt.gov': 0.9738392227472342,
'idlewords.com': 0.9737410870758311,
'randomascii.wordpress.com': 0.97368784410819,
'jepsen.io': 0.9736187109986643,
'zachholman.com': 0.9735522719786461,
'githubengineering.com': 0.9734850428577687,
'tinyprojects.dev': 0.9733692809444601,
'www.quantamagazine.org': 0.9731098630624817,
'aws.amazon.com': 0.9730261011769831,
'www.washingtonpost.com': 0.9727186917658874,
'www.tbray.org': 0.9727073933211833,
'sqlite.org': 0.9725225769916973,
'sivers.org': 0.9725186799112472,
'inkscape.org': 0.9725022016010344,
'www.deepsouthventures.com': 0.972453385720152,
'blog.openai.com': 0.9723220955611721,
'courses.csail.mit.edu': 0.9722700145319921,
'www.theguardian.com': 0.9722441587612369,
'dcurt.is': 0.9721843131015032,
'blog.atom.io': 0.9720527700753158,
'www.newyorker.com': 0.9720054386788728,
'lists.gnu.org': 0.9719999849041815,
'www.ccc.de': 0.9719484596362211,
'googleprojectzero.blogspot.com': 0.9719076672640107,
'blog.cloudflare.com': 0.9718848285343656,
'jeffhuang.com': 0.9718720207664465,
'duckduckgo.com': 0.9718309347264379,
'www.sqlite.org': 0.9717513581610722,
'marc.info': 0.9716364945361756,
'www.anandtech.com': 0.9715269935352944,
'setosa.io': 0.9714750377943311,
'gogameguru.com': 0.9714728059591474,
'foundation.mozilla.org': 0.9714509129021823,
'patrickcollison.com': 0.9713258071934422,
'daringfireball.net': 0.9711795273015573,
'www.codinghorror.com': 0.971168931371747,
'raphlinus.github.io': 0.9711321390363085,
'blog.chriszacharias.com': 0.9711291613244216,
'facebook.github.io': 0.9711262368674779,
'code.facebook.com': 0.9711146723366064,
'blog.priceonomics.com': 0.9710225579333793,
'petapixel.com': 0.9709592479735079,
'aphyr.com': 0.9709163111780117,
'blog.jupyter.org': 0.970807531784939,
'www.vice.com': 0.9707422391855615,
'josephg.com': 0.9707173697331065,
'groups.google.com': 0.9707047476201209,
'www.candyjapan.com': 0.9706864280847017,
'www.bbc.co.uk': 0.9706752756956991,
'www.raspberrypi.org': 0.9706652392429661,
'paulbuchheit.blogspot.com': 0.9706297170684384,
'blog.triplebyte.com': 0.9706184118205334,
'web.eecs.utk.edu': 0.9705843602166302,
'www.foundationdb.org': 0.970531828365254,
'factorio.com': 0.9705079220882863,
'brew.sh': 0.9702877485101176,
'arstechnica.com': 0.9702828920003089,
'jlongster.com': 0.970220590502143,
'blog.dustinkirkland.com': 0.9702178011253874,
'daniel.haxx.se': 0.9701944690304003,
'rethinkdb.com': 0.9701899521411085,
'blog.linode.com': 0.9700603805951686,
'andrewkelley.me': 0.9699936860995682,
'theintercept.com': 0.9698660114939796,
'morepypy.blogspot.com': 0.9698368234837993,
'lkml.org': 0.9698185224636935,
'www.cbc.ca': 0.9697944363631238,
'www.justinobeirne.com': 0.9697393718499845,
'www.pine64.org': 0.9696643433350861,
'www.thirtythreeforty.net': 0.9696053743544627,
'www.aclu.org': 0.9696051245989374,
'robertheaton.com': 0.9695217351343498,
'www.justice.gov': 0.96946427848646,
'startupclass.samaltman.com': 0.9694186083625935,
'www.atlasobscura.com': 0.9693830118062187,
'blog.archive.org': 0.9692692877978696,
'wozniak.ca': 0.9692591742384942,
'shkspr.mobi': 0.9690681057241932,
'wikileaks.org': 0.9688043568238527,
'lucumr.pocoo.org': 0.9687437893479095,
'www.defmacro.org': 0.9685745716212145,
'cloudplatform.googleblog.com': 0.9685595849759793,
'www.wsj.com': 0.9685329922823139,
'www.gatesnotes.com': 0.9685175797648193,
'lapcatsoftware.com': 0.9684021881186297,
'www.bunniestudios.com': 0.9683871579382967,
'puri.sm': 0.9681483253888264,
'madned.substack.com': 0.9680061059600782,
'www.openculture.com': 0.9679136786699476,
'google.github.io': 0.9679127919179034,
'spreadprivacy.com': 0.9678904674640757,
'37signals.com': 0.9677339661772352,
'thedaywefightback.org': 0.96772024509243,
'asia.nikkei.com': 0.9676721486940052,
'blog.pragmaticengineer.com': 0.9676540840884938,
'brandur.org': 0.9675999497340669,
'worrydream.com': 0.9675574587791994,
'www.gimp.org': 0.9675475521140265,
'www.gabrielweinberg.com': 0.9675439099356123,
'tyler.io': 0.9675272832002738,
'www.theatlantic.com': 0.9674601161371825,
'deno.land': 0.9674479589983649,
'www.economist.com': 0.9674323307648195,
'sheddingbikes.com': 0.9672694753671274,
'pud.com': 0.9671820049226395,
'lavabit.com': 0.9671393498011488,
'blog.bradfieldcs.com': 0.9671289002274366,
'www.smithsonianmag.com': 0.9670852442662822,
'elm-lang.org': 0.9670684273054104,
'themacro.com': 0.9670447125421023,
'security.googleblog.com': 0.9669238650508819,
'www.voxelquest.com': 0.9669230619143929,
'www.gnu.org': 0.966899856247157,
'mail.mozilla.org': 0.9668731777544588,
'www.jeffgeerling.com': 0.9668579896794777,
'stallman.org': 0.9668266183513317,
'edwardsnowden.substack.com': 0.9666584767798317,
'www.latimes.com': 0.9666263838504232,
'blog.torproject.org': 0.9666181158060496,
'bugzilla.mozilla.org': 0.966563230800294,
'developer.apple.com': 0.9665153046671373,
'www.nature.com': 0.9665034870295038,
'blog.docker.com': 0.9664589097197918,
'www.cockroachlabs.com': 0.9664546158920483,
'earthly.dev': 0.9664394437158831,
'nullprogram.com': 0.9664053052413792,
'code.visualstudio.com': 0.9664051613640476,
'teachyourselfcs.com': 0.96638716524213,
'immersivemath.com': 0.9663756953881175,
'fly.io': 0.9663690278982495,
'blog.system76.com': 0.966297694451117,
'torrentfreak.com': 0.9662949409422105,
'swift.org': 0.9662794829073483,
'plausible.io': 0.9661981619745931,
'blog.timescale.com': 0.9660867756775726,
'elixir-lang.org': 0.9660464292856342,
'einaregilsson.com': 0.9660162058065089,
'blog.servo.org': 0.9658921829300279,
'tsaoutofourpants.wordpress.com': 0.9658298664203504,
'www.brendangregg.com': 0.9657781891262812,
'blog.jessfraz.com': 0.965647862426781,
'www.gwern.net': 0.9656005597335781,
'techblog.netflix.com': 0.9654696251760763,
'www.videolan.org': 0.9654507268439513,
'blogs.windows.com': 0.9653807535564003,
'arp242.net': 0.965353872976349,
'mjg59.dreamwidth.org': 0.9653138636026742,
'www.stavros.io': 0.9651634551351548,
'www.expeditedssl.com': 0.9651401658177345,
'matt.might.net': 0.9651059400349459,
'flif.info': 0.9650927452914915,
'apenwarr.ca': 0.9650906673961581,
'googleonlinesecurity.blogspot.com': 0.9650850292429668,
'priceonomics.com': 0.9650537614148901,
'www.humbledmba.com': 0.9649972831093808,
'www.unrealengine.com': 0.964977222983089,
'people.xiph.org': 0.9648830246858617,
'betterexplained.com': 0.9648763473852692,
'al3x.net': 0.9647293502855223,
'notch.tumblr.com': 0.9646905626593433,
'deno.com': 0.9646806401289668,
'www.wired.com': 0.9646357841756316,
'git.kernel.org': 0.9645834348149406,
'www.djangoproject.com': 0.9645683143391122,
'www.ruby-lang.org': 0.964565023525263,
'www.catonmat.net': 0.9645567860107777,
'blog.plover.com': 0.9645429951505131,
'ocw.mit.edu': 0.9644857262463964,
'steveblank.com': 0.9644558764763906,
'blog.haschek.at': 0.9643961053958355,
'old.reddit.com': 0.9643838762037145,
'blog.jgc.org': 0.9643196349091965,
'utcc.utoronto.ca': 0.9643144411744943,
'blog.getbootstrap.com': 0.964223508260427,
'protonmail.com': 0.9641434352105049,
'stackoverflow.blog': 0.964131467903275,
'www.feynmanlectures.caltech.edu': 0.9640844147082678,
'www.nationalgeographic.com': 0.9640002724660541,
'thume.ca': 0.9639172934049174,
'www.ftc.gov': 0.9638662117372075,
'blog.google': 0.9638278246372475,
'talks.golang.org': 0.9638081833676005,
'identity.mozilla.com': 0.9637913542538376,
'www.antipope.org': 0.9637681454362576,
'blog.coinbase.com': 0.9637227588096539,
'freedom.press': 0.9637055279129609,
'research.googleblog.com': 0.9636957954323193,
'sneak.berlin': 0.9636104943866459,
'ferd.ca': 0.9636081542021665,
'system76.com': 0.9636004595237442,
'jsomers.net': 0.9635822620240366,
'nautil.us': 0.9635511482835416,
'blog.fogus.me': 0.963547608254157,
'probablydance.com': 0.9635177117293328,
'blog.stephenwolfram.com': 0.9633882383623975,
'blog.42floors.com': 0.9633188247961246,
'www.thoughtcrime.org': 0.9631942686290309,
'www.winehq.org': 0.9631344924353825,
'www.benkuhn.net': 0.9631304906372505,
'status.cloud.google.com': 0.9630969501387503,
'dev.stephendiehl.com': 0.9630485789318193,
'www.thedrive.com': 0.9629761123330604,
'explainshell.com': 0.9629455061466635,
'krita.org': 0.9629305982536617,
'www.paulgraham.com': 0.96284858742987,
'www.google.com': 0.9626443067571239,
'blogs.msdn.microsoft.com': 0.9626200437043355,
'www.techdirt.com': 0.9625942195899927,
'homakov.blogspot.com': 0.9625907627104796,
'gist.github.com': 0.9625295427025308,
'lists.freebsd.org': 0.9624953857245481,
'www.axios.com': 0.9624629027024122,
'triplebyte.com': 0.9623309816301445,
'blog.filippo.io': 0.9621095196217451,
'mattstoller.substack.com': 0.9620940895652949,
'weblog.rubyonrails.org': 0.9620546173159494,
'viewsourcecode.org': 0.9620510951208904,
'gcc.gnu.org': 0.9619503003058159,
'tjcx.me': 0.9619235394931922,
'www.scottaaronson.com': 0.9618863749512754,
'www.mattkeeter.com': 0.9618540812590697,
'cloud.google.com': 0.9618480240586683,
'edweissman.com': 0.9616816249147265,
'www.citylab.com': 0.961667739783419,
'ziglang.org': 0.9616247712590222,
'ffmpeg.org': 0.9615920017638592,
'www.tablix.org': 0.9615841584158416,
'opensource.googleblog.com': 0.9615491317254445,
'www.lettersofnote.com': 0.9615449446825298,
'blog.pushbullet.com': 0.9614414831416516,
'www.jacquesmattheij.com': 0.9613892326733957,
'zwischenzugs.com': 0.9613828940987429,
'www.reddit.com': 0.9613787026817021,
'blogs.valvesoftware.com': 0.9613624503908588,
'markosaric.com': 0.9613621062764696,
'os.phil-opp.com': 0.9613360087194533,
'blogs.dropbox.com': 0.961291183608417,
'docs.google.com': 0.9612568870731552,
'juliareda.eu': 0.9612194124785559,
'uspirg.org': 0.9611978470349601,
'fasterthanli.me': 0.961115384552988,
'ipfs.io': 0.9610605476989421,
'queue.acm.org': 0.9610281321922659,
'www.evanmiller.org': 0.9609644541353691,
'ejohn.org': 0.9609035373663662,
'releases.ubuntu.com': 0.9608845798797817,
'support.google.com': 0.9608695426684296,
'lcamtuf.coredump.cx': 0.9608422980804076,
'mango.pdf.zone': 0.9607785943599466,
'godotengine.org': 0.9607731456425479,
'justgetflux.com': 0.960753420730463,
'bradfrost.com': 0.9607430819229773,
'nim-lang.org': 0.9607343905204542,
'simonwillison.net': 0.9606872824809413,
'www.blog.google': 0.96065581868309,
'www.sciencemag.org': 0.9605214304889523,
'byorgey.wordpress.com': 0.9604978104194218,
'www.groklaw.net': 0.9604516891162848,
'geon.github.io': 0.9603711026188184,
'lithub.com': 0.9603590822727315,
'fiber.google.com': 0.9602031412872665,
'dave.cheney.net': 0.9601949227730773,
'www.scientificamerican.com': 0.9601944955422137,
'beepb00p.xyz': 0.9601931949641657,
'blog.fastmail.com': 0.9601797872297155,
'seclists.org': 0.9601735085911461,
'www.hillelwayne.com': 0.9600285295350695,
'blogs.scientificamerican.com': 0.9599933529455126,
'www.filfre.net': 0.9599628009632043,
'www.oculus.com': 0.959958903358201,
'graydon2.dreamwidth.org': 0.9599484289082291,
'www.jonathanturner.org': 0.9599080310891523,
'jameshfisher.com': 0.9598899186695382,
'julialang.org': 0.9598652009213735,
'blogs.harvard.edu': 0.9598532297709538,
'academictorrents.com': 0.9598305702264962,
'staltz.com': 0.9598086566619791,
'www.openbsd.org': 0.9597593869835838,
'psyche.co': 0.9597358328264424,
'blog.rongarret.info': 0.9597212936777375,
'www.redblobgames.com': 0.9597207704543808,
'dtrace.org': 0.9596738647309077,
'boingboing.net': 0.9596494107644848,
'www.schneier.com': 0.9595794118397106,
'support.mozilla.org': 0.9595011501832328,
'arxiv.org': 0.959473232072393,
'twobithistory.org': 0.9594688834229274,
'tttthis.com': 0.9594411193380538,
'www.windows93.net': 0.9594238637098613,
'med.stanford.edu': 0.9594108760218417,
'ascii.textfiles.com': 0.959256899162529,
'spectrum.ieee.org': 0.9591657750656283,
'www.aaronsw.com': 0.959152603211572,
'blog.chromium.org': 0.9590740081078435,
'daltoncaldwell.com': 0.9588971457836526,
'jeffe.cs.illinois.edu': 0.9588078449509829,
'hookrace.net': 0.9587652481709775,
'lessig.tumblr.com': 0.9587508585934718,
'blog.garrytan.com': 0.958672955025337,
'leahneukirchen.org': 0.9586676171256273,
'journal.stuffwithstuff.com': 0.9586488857012637,
'status.slack.com': 0.9586248004018981,
'www.hanselman.com': 0.958563907251422,
'www.theverge.com': 0.9585530026858043,
'prog21.dadgum.com': 0.9584999221959226,
'www.craigmurray.org.uk': 0.9584843626546714,
'motherboard.vice.com': 0.9584647236016686,
'awesomekling.github.io': 0.9584478246123319,
'acko.net': 0.9584223182205939,
'www.cnbc.com': 0.9583609005406114,
'misc-stuff.terraaeon.com': 0.9582853790359502,
'blog.robertelder.org': 0.9581077255960667,
'www.patrick-breyer.de': 0.958008658008658,
'mainichi.jp': 0.9579948206173503,
'notepad-plus-plus.org': 0.9579831679547762,
'dolphin-emu.org': 0.957980480121585,
'www.adriancourreges.com': 0.9577752697052809,
'www.anishathalye.com': 0.9577244298214679,
'martin.kleppmann.com': 0.9576931939838363,
'developers.google.com': 0.9576890556647433,
'code.google.com': 0.9576138371902363,
'support.apple.com': 0.9575695045037443,
'buy.louisck.net': 0.9575368691358187,
'sandstorm.io': 0.9575210697630335,
'blog.fogcreek.com': 0.9575063092775307,
'panic.com': 0.9575015340393305,
'svelte.dev': 0.957467967080876,
'justinkan.com': 0.9574047166110535,
'www.ianbicking.org': 0.9573945069975465,
'firstlook.org': 0.9573366845702125,
'foundersatwork.posthaven.com': 0.9572959138256529,
'mail.python.org': 0.9572072777215898,
'acoup.blog': 0.9571594417654118,
'jacobian.org': 0.9571381159110411,
'hakibenita.com': 0.9568994480645705,
'mothership.sg': 0.9568299369913541,
'www.vitavonni.de': 0.9568005963591861,
'palant.de': 0.9567983422438127,
'benhoyt.com': 0.9567603051598327,
'david.li': 0.9567049710339931,
'web.stanford.edu': 0.9566872891659536,
'a16z.com': 0.9565808474917423,
'www.battleforthenet.com': 0.9565732116764964,
'uncrunched.com': 0.9565198150211067,
'www.debian.org': 0.9564674281841146,
'nullsweep.com': 0.9564159156215992,
'wiki.postgresql.org': 0.9563944999840933,
'www.snellman.net': 0.9563712102508779,
'www.craigkerstiens.com': 0.9563664501753555,
'wikimediafoundation.org': 0.9563345661162456,
'eater.net': 0.9563202100645842,
'microsoft.github.io': 0.9563042502322844,
'redditblog.com': 0.9562631224182452,
'www.aaronkharris.com': 0.9561555239477143,
'blogs.microsoft.com': 0.9560820272756707,
'qz.com': 0.955949644460612,
'tailscale.com': 0.9559251595739959,
'codewords.recurse.com': 0.9558536488585196,
'bost.ocks.org': 0.9557437711557101,
'gmailblog.blogspot.com': 0.9557079840056077,
'norvig.com': 0.9556798812425891,
'pluralistic.net': 0.9556745944240528,
'www.simplethread.com': 0.9556668873107272,
'nltimes.nl': 0.9555494259023241,
'newsroom.intel.com': 0.9554793469031995,
'news.mit.edu': 0.9554200420895869,
'jitsi.org': 0.9554166636194328,
'www.cs.cmu.edu': 0.955323422003546,
'article.gmane.org': 0.9553116742480078,
'joearms.github.io': 0.9552623762850195,
'mzucker.github.io': 0.9552342582549446,
'qntm.org': 0.9552256231956235,
'iterm2.com': 0.955216811415582,
'blog.github.com': 0.9552099253702252,
'engineering.shopify.com': 0.9551782438728066,
'geoff.greer.fm': 0.9551520599196336,
'reactjs.org': 0.9551193524754155,
'www.citusdata.com': 0.9551065831945164,
'blog.dropbox.com': 0.9550988360514683,
'pastebin.com': 0.9550820324377278,
'www.fast.ai': 0.9550621528031076,
'blog.twitter.com': 0.9550007700333996,
'marco.org': 0.9549900576536702,
'begriffs.com': 0.9549733491222114,
'lkml.iu.edu': 0.9549524012989699,
'eclecticlight.co': 0.9549360423325072,
'v8project.blogspot.com': 0.9549013081514219,
'sam.zeloof.xyz': 0.9548519610162581,
'www.facebook.com': 0.954738504671582,
'cfenollosa.com': 0.954730634374589,
'stevelosh.com': 0.9547082859524895,
'mosh.mit.edu': 0.9546865955088134,
'archive.org': 0.954683235332995,
'marctenbosch.com': 0.9546562578949468,
'www.dw.com': 0.954642614537768,
'thehustle.co': 0.9546290284301826,
'phabricator.wikimedia.org': 0.9546112568993993,
'twitter.github.com': 0.9545898933522629,
'brendaneich.com': 0.9545746174697838,
'42floors.com': 0.9545517561468556,
'blog.alexmaccaw.com': 0.9545454626799591,
'martinfowler.com': 0.9545379389229811,
'freedom-to-tinker.com': 0.9544719860088577,
'www.slate.com': 0.9544534258667388,
'www.iphoneincanada.ca': 0.9544449857023617,
'bugs.debian.org': 0.9543851860127158,
'ir.tesla.com': 0.9543767275771423,
'blog.asmartbear.com': 0.9543610682085355,
'greenwald.substack.com': 0.9543135677806216,
'gameprogrammingpatterns.com': 0.9543089003699947,
'blogs.msdn.com': 0.9543003666109717,
'www.meteor.com': 0.9542891479234723,
'www.nasa.gov': 0.9542472672213591,
'petitions.whitehouse.gov': 0.9542172936354238,
'segment.com': 0.9542063188154968,
'lexi-lambda.github.io': 0.9541906249333629,
'www.goodnewsnetwork.org': 0.9541839952530666,
'www.crypto101.io': 0.954146742893107,
'hintjens.com': 0.9541384067709378,
'changelog.complete.org': 0.9541382845549411,
'tech.mit.edu': 0.9541192032072701,
'webkit.org': 0.9540750203597949,
'blog.emacsen.net': 0.9540459808061132,
'instagram-engineering.tumblr.com': 0.9540392857685359,
'boringtechnology.club': 0.9540365522550216,
'swombat.com': 0.9540149579111236,
'berthub.eu': 0.9539835716288688,
'erikbern.com': 0.9539560598971636,
'www.theregister.com': 0.9538737853031798,
'blog.reddit.com': 0.9538429471349307,
'learnpythonthehardway.org': 0.9537983780986163,
'dynomight.net': 0.953604369679408,
'danshipper.com': 0.9535908420696627,
'status.aws.amazon.com': 0.9535221631672808,
'www.kennethreitz.org': 0.9535129582723397,
'www.jefftk.com': 0.9534058135420997,
'idiallo.com': 0.9533596892670664,
'developer.mozilla.org': 0.9532377447812391,
'discourse.mozilla.org': 0.9531776168886109,
'www.vox.com': 0.953171184497987,
'blog.burntsushi.net': 0.9531628061816667,
'www.roguelazer.com': 0.9531341112752366,
'beej.us': 0.9531338211286129,
'www.destroyallsoftware.com': 0.9530554229381429,
'eev.ee': 0.9530405280939158,
'konklone.com': 0.9530176239519549,
'www.twilio.com': 0.9530110571482041,
'www.outsideonline.com': 0.9530080955571504,
'quuxplusone.github.io': 0.9528999786913215,
'www.chrisstucchio.com': 0.9528639692666151,
'themarkup.org': 0.9528465949993555,
'www.nobelprize.org': 0.9528281612185788,
'samcurry.net': 0.9528119859873818,
'phoboslab.org': 0.9527996714263891,
'media.ccc.de': 0.9527965238893377,
'tom.preston-werner.com': 0.9527727795158389,
'iism.org': 0.9527632645423888,
'blog.joinmastodon.org': 0.9527602941375386,
'www.accc.gov.au': 0.9527263269590833,
'www.ncbi.nlm.nih.gov': 0.9527198508675186,
'ianstormtaylor.com': 0.952712496504424,
'www.imperialviolet.org': 0.9526934873020364,
'ridiculousfish.com': 0.9526563705263446,
'developers.googleblog.com': 0.9526246790332386,
'backreaction.blogspot.com': 0.9525612446013791,
'www.theparisreview.org': 0.9525261097376244,
'blog.jquery.com': 0.9524629435660441,
'www.michaeleisen.org': 0.9524499448559609,
'news.berkeley.edu': 0.9524491052294287,
'travisdowns.github.io': 0.9523196619971284,
'blog.nuclino.com': 0.9523119161198071,
'news.stanford.edu': 0.9522754316114468,
'www.universityofcalifornia.edu': 0.9522373906378331,
'gravitational.com': 0.9522355620834344,
'zserge.com': 0.9521909899856118,
'observablehq.com': 0.9521499301321013,
'mathoverflow.net': 0.9521475886704486,
'www.joyent.com': 0.9520628241406525,
'blog.kevmod.com': 0.9520548060547861,
'blog.heroku.com': 0.9520355160215065,
'www.strongtowns.org': 0.952027603849832,
'nextjs.org': 0.9520143111821525,
'www.icij.org': 0.9519656123992113,
'www.ifixit.com': 0.9519637451973145,
'www.windytan.com': 0.9518643147043334,
'www.servethehome.com': 0.9518529509789444,
'grumpy.website': 0.9517679506064377,
'bitwarden.com': 0.9517316136273904,
'gitlab.com': 0.9515782210258252,
'distill.pub': 0.9515137257464145,
'craftinginterpreters.com': 0.9514779102710061,
'lord.io': 0.9514586580765187,
'www.eso.org': 0.9514502937076248,
'www.codersnotes.com': 0.9514270198661366,
'thereader.mitpress.mit.edu': 0.9513838603840151,
'blog.regehr.org': 0.9513649990290555,
'googleresearch.blogspot.com': 0.9513477926783215,
'highscalability.com': 0.9513429766551695,
'toddwschneider.com': 0.9513421190975475,
'slatestarcodex.com': 0.9511855687467806,
'www.cloudflarestatus.com': 0.9511829060902408,
'www.cnx-software.com': 0.9511778260754474,
'www.statnews.com': 0.9511048711614947,
'youtube-eng.blogspot.com': 0.9510868926645011,
'news.yale.edu': 0.9510347346201274,
'reneweconomy.com.au': 0.9509936100187831,
'www.jwz.org': 0.950973914048226,
'ncase.me': 0.9509706871520447,
'www.metzdowd.com': 0.9509697689890211,
'edu.mkrecny.com': 0.9508994617990429,
'medium.com': 0.9508522487424811,
'mastermind.atavist.com': 0.9508126598566774,
'www.laphamsquarterly.org': 0.9507738399701613,
'telegram.org': 0.9507726262466255,
'therecord.media': 0.9507691852400171,
'www.salon.com': 0.9507685396993281,
'www.howacarworks.com': 0.9507437871644115,
'web.mit.edu': 0.9507345330973588,
'www.zdnet.com': 0.9507029832966206,
'paulstamatiou.com': 0.9506812073068672,
'eloquentjavascript.net': 0.9506744929567004,
'nathanmarz.com': 0.9506394532348129,
'en.wikipedia.org': 0.9506273469081818,
'explained.ai': 0.9506171121716123,
'golang.org': 0.9505833588437288,
'99percentinvisible.org': 0.9505598730904006,
'restoreprivacy.com': 0.9505353262581112,
'ycombinator.com': 0.9505016273288341,
'www.joshwcomeau.com': 0.950477087355932,
'easydns.com': 0.9504253361110822,
'lists.w3.org': 0.9504081521659986,
'nvidianews.nvidia.com': 0.9504026864806336,
'cyberlaw.stanford.edu': 0.9503746350188895,
'www.redhat.com': 0.9503564767146361,
'blog.racket-lang.org': 0.9503359424455089,
'www.ctrl.blog': 0.9503095809400977,
'bjk5.com': 0.9502965197985591,
'refactoringui.com': 0.9502851812628592,
'reactos.org': 0.9502772626866873,
'matthias-endler.de': 0.9502126168099793,
'blog.usejournal.com': 0.9501925685645481,
'www.deprocrastination.co': 0.9501876199581508,
'blog.llvm.org': 0.9501759187745217,
'blog.keras.io': 0.9501685115979439,
'markodenic.com': 0.9501517264182752,
'tirania.org': 0.9501284983515588,
'www.copetti.org': 0.9500923230891157,
'www.interviewcake.com': 0.9500659008199689,
'newatlas.com': 0.9500230591731798,
'blog.okcupid.com': 0.9499974432299237,
'smallstep.com': 0.949970101995039,
'act.eff.org': 0.9499506284870781,
'www.extremetech.com': 0.9498989577556406,
'kottke.org': 0.9498599050938643,
'aeon.co': 0.9498533547972355,
'blog.printf.net': 0.949841911956293,
'www.johndcook.com': 0.9497830834287447,
'community.letsencrypt.org': 0.9497776958809281,
'www.hashicorp.com': 0.9497516160457014,
'thewaltdisneycompany.com': 0.9495508540625482,
'www.haiku-os.org': 0.9495281464383918,
'scotthelme.co.uk': 0.949509887884973,
'www.gamasutra.com': 0.9494506172378354,
'cdixon.org': 0.9494179897832785,
'azeria-labs.com': 0.9493868918991253,
'stevehanov.ca': 0.949386070425381,
'calebporzio.com': 0.9493736951983298,
'blog.erlang.org': 0.9493443327606497,
'blog.backblaze.com': 0.9493279590675145,
'nymag.com': 0.9493193709909513,
'lists.debian.org': 0.9492598455856424,
'blog.mailgun.com': 0.9492541788226916,
'www.microsoft.com': 0.9492199020476939,
'tokio.rs': 0.9492084252828289,
'woodgears.ca': 0.949182590244603,
'www.zotero.org': 0.9491798215496959,
'zeit.co': 0.9491636525275446,
'tech.dropbox.com': 0.9491502994308182,
'pages.cs.wisc.edu': 0.9491309277173726,
'engineering.instagram.com': 0.9491288216402667,
'www.realworldtech.com': 0.9491223130451418,
'www.mattcutts.com': 0.9490815088845603,
'www.thecut.com': 0.9489293872710988,
'wiki.mozilla.org': 0.9489007109681905,
'theunderstatement.com': 0.9488791842136701,
'donmelton.com': 0.9488523200756159,
'learncodethehardway.org': 0.9487806920147928,
'element.io': 0.9487707078398451,
'thehill.com': 0.948742537996995,
'wattenberger.com': 0.9487142290281894,
'raganwald.com': 0.94870117403287,
'nytimes.com': 0.9487002268786773,
'www.danshapiro.com': 0.9486602162771469,
'amasci.com': 0.9484567715181271,
'steve-yegge.blogspot.com': 0.9484502644034212,
'blog.instapaper.com': 0.9483924654208046,
'www.pathsensitive.com': 0.9483799539691706,
'sachagreif.com': 0.9482981085006165,
'daedtech.com': 0.948282297833387,
'www.phoenixframework.org': 0.948180038678699,
'avc.com': 0.9481175720932656,
'jasmcole.com': 0.9480827978886457,
'lemire.me': 0.9480710409382858,
'ycombinator.posterous.com': 0.9480447746714599,
'crystal-lang.org': 0.9480289688481155,
'www.flightradar24.com': 0.9480106923973552,
'www.lofibucket.com': 0.9479892761394102,
'www.agner.org': 0.9479708121651846,
'www.mercurynews.com': 0.947909748405247,
'peteris.rocks': 0.947883344813371,
'www.haneycodes.net': 0.9478734625701677,
'blog.royalsloth.eu': 0.9478558802047412,
'jeremykun.com': 0.9478499398633942,
'netmarketshare.com': 0.9478165755279099,
'streaming.media.ccc.de': 0.9477448566079527,
'scattered-thoughts.net': 0.9477334018972962,
'theoutline.com': 0.9476703233752298,
'www.tedunangst.com': 0.9476236926963729,
'siberiantimes.com': 0.9476162638549342,
'avherald.com': 0.947493333568265,
'hackaday.com': 0.9474690019349598,
'www.hackerfactor.com': 0.9474384569226819,
'doist.com': 0.9473521795738291,
'ai.googleblog.com': 0.9473483560369622,
'macwright.com': 0.9473244768842966,
'www.cake.co': 0.9473057089030983,
'blog.ptsecurity.com': 0.9472729024375095,
'harpers.org': 0.9472208781730334,
'research.swtch.com': 0.9471779558478995,
'landing.google.com': 0.9471768097472296,
'stackoverflow.com': 0.9471708004823144,
'www.khanacademy.org': 0.9470872439554527,
'www.malwaretech.com': 0.9470832155302483,
'pippinbarr.github.io': 0.9470597529838243,
'www.chronicle.com': 0.9470439916157474,
'www.michaelnielsen.org': 0.9470172835353264,
'yahoo.tumblr.com': 0.9470082148918799,
'threadreaderapp.com': 0.9470043666496185,
'lcamtuf.blogspot.com': 0.9469939359995508,
'www.lowtechmagazine.com': 0.9469818679809128,
'thorstenball.com': 0.9469695919624959,
'copy.sh': 0.946952731836535,
'datagubbe.se': 0.9468822401332838,
'thehftguy.com': 0.9468555549213109,
'www.amazon.com': 0.9468350213972545,
'fishshell.com': 0.9468195302562247,
'lists.torproject.org': 0.9467852152182991,
'www.wbur.org': 0.9467718490726582,
'blog.inkdrop.info': 0.9467189844358732,
'retool.com': 0.9467024959584281,
'e360.yale.edu': 0.9466990573296785,
'magcius.github.io': 0.9466830296864568,
'www.aboutamazon.com': 0.9466566508253851,
'howonlee.github.io': 0.9466352935788864,
'www.randsinrepose.com': 0.9466262113364665,
'www.inkandswitch.com': 0.9466108105187082,
'bhorowitz.com': 0.9466022453520991,
'keenen.xyz': 0.9465855123017971,
'www.forbes.com': 0.9463974275879253,
'www.zachtronics.com': 0.9463917451738472,
'engineering.fb.com': 0.9463604397235058,
'taibbi.substack.com': 0.946310654934769,
'blog.elementary.io': 0.9462655485137151,
'blog.waymo.com': 0.9461989396383733,
'www.factorio.com': 0.9461844645059527,
'ghostbin.com': 0.9461717385490003,
'blog.cubeofm.com': 0.9461649703706937,
'privacyinternational.org': 0.9461573725205245,
'mondaynote.com': 0.9461425333335384,
'chrome.blogspot.com': 0.9461275447307622,
'www.swissinfo.ch': 0.9460542777095045,
'fs.blog': 0.9460214401678857,
'rosenzweig.io': 0.9459592348864171,
'thehackerblog.com': 0.9459179773558213,
'palant.info': 0.9459138777363664,
'course.fast.ai': 0.9459131445844281,
'www.python.org': 0.9459056139143512,
'www.samsung.com': 0.9459051315675879,
'www.niemanlab.org': 0.9458995158159859,
'perens.com': 0.9458876688022819,
'www-cs-students.stanford.edu': 0.9458541576360181,
'www.matuzo.at': 0.9458150903467117,
'yosefk.com': 0.9457825651296167,
'samsaffron.com': 0.9457319055612109,
'www.cs.usfca.edu': 0.9457009533921606,
'www.the-tls.co.uk': 0.9456885441587526,
'erlang.org': 0.9456437311862155,
'home.cern': 0.945571703124721,
'www.joelonsoftware.com': 0.945567200214926,
'scottlocklin.wordpress.com': 0.9455154840830812,
'blog.zoom.us': 0.9454631310396542,
'clojure.org': 0.9453997954968069,
'nest.com': 0.9453765377092626,
'pi-hole.net': 0.9453523308356795,
'erik-engheim.medium.com': 0.9453523308356795,
'projecteuler.net': 0.9453380897239767,
# 'web.archive.org': 0.945336089251807,
'coreos.com': 0.945323515243916,
'slack.engineering': 0.9453141410084411,
'jenkins.io': 0.9452507411562461,
'libcello.org': 0.9452471895570808,
'www.ietf.org': 0.9452216766608553,
'vivaldi.com': 0.9452169051506946,
'blog.acolyer.org': 0.9452136927944143,
'opensource.apple.com': 0.945156165859438,
'www.devever.net': 0.9451481294849068,
'judiciary.house.gov': 0.9451436080311264,
'cstack.github.io': 0.9451357466063348,
'neocities.org': 0.945112946847744,
'www.vanityfair.com': 0.9451005472578717,
'heidiroizen.tumblr.com': 0.945076890496307,
'www.newsweek.com': 0.9450607070750251,
'fosdem.org': 0.9450470405160809,
'lists.llvm.org': 0.9450101340465437,
'swannodette.github.io': 0.9450060465106352,
'www.forbrukerradet.no': 0.944980556159889,
'www.alexkras.com': 0.944800412259009,
'yehudakatz.com': 0.9447946105109732,
'www.politico.eu': 0.9447851995448193,
'research.microsoft.com': 0.944725678463696,
'matklad.github.io': 0.9447162628284546,
'www.fieggen.com': 0.9446507411542784,
'fermatslibrary.com': 0.9446378294996843,
'llvm.org': 0.9446335834779859,
'www.codeofhonor.com': 0.9446318312955025,
'www.sandimetz.com': 0.9445834519299251,
'beautifulracket.com': 0.944575515895676,
'solar.lowtechmagazine.com': 0.9445367689612141,
'www.rollingstone.com': 0.9444958767732993,
'www.theglobeandmail.com': 0.9444675154775829,
'crypto.stanford.edu': 0.9444289674480927,
'www.ucsf.edu': 0.9443956140487278,
'mailman.stanford.edu': 0.9443461149709853,
'games.greggman.com': 0.9442484049862852,
'robrhinehart.com': 0.9442240528350656,
'evertpot.com': 0.9441892143713023,
'ryancarson.com': 0.944157143243096,
'www.oculusvr.com': 0.944136637962384,
'news.microsoft.com': 0.9441300159133728,
'reasonablypolymorphic.com': 0.9440992853388425,
'www.nybooks.com': 0.944043771273854,
'blog.malwarebytes.com': 0.9440378261380166,
'sourcehut.org': 0.9440365818314045,
'antonz.org': 0.9439630271519353,
'weavesilk.com': 0.9439624519736566,
'jakevdp.github.io': 0.9439416004808824,
'www.githubstatus.com': 0.9439219010573966,
'thenextweb.com': 0.9438984254861107,
'squareup.com': 0.9438684150241483,
'www.businessinsider.com': 0.9438381542606629,
'www.teamten.com': 0.9438162837927613,
'minimaxir.com': 0.94381424365665,
'www.sparkfun.com': 0.9437861060710178,
'www.seattletimes.com': 0.9437565154559231,
'int10h.org': 0.943749694541571,
'cpbotha.net': 0.9437489993962991,
'cslibrary.stanford.edu': 0.9437134939561775,
'mosaicscience.com': 0.943712263624496,
'www.cjr.org': 0.9436996745182413,
'lareviewofbooks.org': 0.9436930841143853,
'www.theregister.co.uk': 0.9436788261536825,
'www.phoronix.com': 0.9436747415359684,
'blog.discord.com': 0.9436725616861676,
'doc.rust-lang.org': 0.9436705513749973,
'onezero.medium.com': 0.9436010567522964,
'fuchsia.googlesource.com': 0.9435751703627357,
'www.tomanthony.co.uk': 0.9435582795515635,
'blog.ethereum.org': 0.943554185163966,
'rhaas.blogspot.com': 0.9435179337782055,
'www.historytoday.com': 0.9434724777695044,
'veekaybee.github.io': 0.9434651701356596,
'www.michaelburge.us': 0.9433630281363066,
'www.hackerschool.com': 0.9433243309324416,
'507movements.com': 0.9433217341059063,
'www.bmj.com': 0.9433027312455573,
'blogs.wsj.com': 0.9432673977773712,
'leginfo.legislature.ca.gov': 0.9432572372586395,
'people.csail.mit.edu': 0.9430404662547742,
'nibblestew.blogspot.com': 0.9429934909491315,
'backchannel.com': 0.942970399762578,
'www.openssl.org': 0.9429668863010962,
'www.openwall.com': 0.9429455499515123,
'www.petapixel.com': 0.9428687830757905,
'jeff-vogel.blogspot.com': 0.9427633325613003,
'roslyn.codeplex.com': 0.9427563891114413,
'forums.dropbox.com': 0.9427243466398204,
'answers.microsoft.com': 0.9426879968589384,
'blog.jazzychad.net': 0.9426788682750873,
'www.atrium.co': 0.9426550434185653,
'www.ycombinator.com': 0.9425591138591901,
'statmodeling.stat.columbia.edu': 0.9425542863968303,
'cacm.acm.org': 0.9425392273023008,
'firstround.com': 0.9424999923251012,
'blog.erratasec.com': 0.9423971981109275,
'colah.github.io': 0.942392249847614,
'online.wsj.com': 0.9423603972359367,
'blog.paralleluniverse.co': 0.9423524400607514,
'boz.com': 0.9422809938913371,
'dmitry.gr': 0.942260189831035,
'www.rethinkdb.com': 0.9422304687024133,
'www.linuxjournal.com': 0.9421550256435562,
'www.bottomupcs.com': 0.9421232771919728,
'commoncog.com': 0.9420976962954309,
'status.github.com': 0.9419938569642178,
'www.drmaciver.com': 0.9419852322184868,
'imgur.com': 0.9419771536808862,
'privacylog.blogspot.com': 0.9418867600921542,
'lawsofux.com': 0.9418843340175082,
'missing.csail.mit.edu': 0.9418514250128648,
'mitpress.mit.edu': 0.9418492695998034,
'blog.wesleyac.com': 0.9418306497508919,
'sanctum.geek.nz': 0.9418252921101978,
'rarehistoricalphotos.com': 0.9418058693265607,
'blog.patreon.com': 0.9416385088594889,
'stephaniehurlburt.com': 0.9416294852135288,
'www.ssh.com': 0.9416225784673645,
'sahillavingia.com': 0.9415810451498678,
'justinmeiners.github.io': 0.9415524612629268,
'www.bleepingcomputer.com': 0.9415420860759846,
'thejh.net': 0.941539543862244,
'codingnest.com': 0.941522096507017,
'www.motorauthority.com': 0.9415012015037275,
'kevinlynagh.com': 0.9414829971920764,
'www.home-assistant.io': 0.941436958446535,
'spritesmods.com': 0.9413587273840005,
'www.esa.int': 0.9412959233934257,
'obsproject.com': 0.9412832929782082,
'www.1843magazine.com': 0.9412677053641687,
'therealkatie.net': 0.9412648438023827,
'christine.website': 0.9412231716105919,
'aadrake.com': 0.941209497556143,
'zapier.com': 0.9411980033904498,
'diegobasch.com': 0.9411801630225842,
'lichess.org': 0.9411613867773386,
'www.collectorsweekly.com': 0.9411605509869836,
'www.spiegel.de': 0.9411529028868376,
'kev.inburke.com': 0.9410868672076137,
'matt.sh': 0.9410639090369376,
'turkeyblocks.org': 0.9410266902131287,
'translate.google.com': 0.9410059430133028,
'well.blogs.nytimes.com': 0.9410008856098844,
'vitalik.ca': 0.9409975246700096,
'www.independent.co.uk': 0.9409567204051296,
'www.themarshallproject.org': 0.9409445330421784,
'www.usenix.org': 0.9409235997678326,
'blog.docker.io': 0.9408971471442287,
'kristoff.it': 0.9408956155328897,
'radio.garden': 0.9408954856512393,
'tholman.com': 0.9408838927997484,
'gregoryszorc.com': 0.9408696765619068,
'phys.org': 0.9408448137648736,
'www.prisma.io': 0.9408393039096592,
'aprilzero.com': 0.9408291999272044,
'blog.parse.com': 0.9408215317556151,
'borncity.com': 0.9407974930873888,
'theconversation.com': 0.9407949039397295,
'www.cl.cam.ac.uk': 0.9407721538543924,
'grumpygamer.com': 0.9407621103834799,
'ml.berkeley.edu': 0.9407575811882065,
'gizmodo.com': 0.9406627597554199,
'rsms.me': 0.9406458076838878,
'bitquabit.com': 0.9406214706399899,
'www.avc.com': 0.9406135247252191,
'iamnotaprogrammer.com': 0.940590780285089,
'chrishateswriting.com': 0.940590780285089,
'newsroom.fb.com': 0.9405869291621605,
'saagarjha.com': 0.9405291077737054,
'supabase.io': 0.9404661433287882,
'me.veekun.com': 0.9404573019098256,
'250bpm.com': 0.9404292658187561,
'blog.gregbrockman.com': 0.9403842788606359,
'www.mono-project.com': 0.9403717514626122,
'www.minimallyminimal.com': 0.9403321431642234,
'9to5mac.com': 0.9402902474697965,
'char.gd': 0.9402697487347172,
'plumshell.com': 0.9402496652921678,
'nixos.org': 0.9402384786300783,
'underjord.io': 0.9402384786300781,
'landoflisp.com': 0.940191169313915,
'www.tnhh.net': 0.9401820210600681,
'rapgenius.com': 0.9401741065194024,
'blog.floydhub.com': 0.9401390082080993,
'aws.typepad.com': 0.9401029943851995,
'leafletjs.com': 0.9400456080747939,
'fonts.google.com': 0.9400428820587936,
'matthewearl.github.io': 0.9400368182082475,
'www.kqed.org': 0.9400004457307097,
'timharford.com': 0.9399850780559476,
'blog.innerht.ml': 0.9399767635106161,
'blog.mecheye.net': 0.9399503933564736,
'about.sourcegraph.com': 0.9399124106162126,
'www.wi-fi.org': 0.9398176506266921,
'www.4chan.org': 0.9397638081665932,
'www.thedailybeast.com': 0.9397280815769313,
'www.ipp.mpg.de': 0.9397137553292056,
'pythonsweetness.tumblr.com': 0.9397041924803161,
'arstechnica.co.uk': 0.9396691911012381,
'news.railstutorial.org': 0.9395578781950391,
'terrytao.wordpress.com': 0.9395312291825673,
'www.inputmag.com': 0.9394997769231928,
'labs.detectify.com': 0.9394881401388651,
'undeadly.org': 0.9394661374968044,
'tech.channable.com': 0.939456208198243,
'mobile.nytimes.com': 0.9394237175327378,
'falkvinge.net': 0.9393964166317929,
'www.omniref.com': 0.9393438893288835,
'notes.unwieldy.net': 0.9392345223233198,
'blog.newsblur.com': 0.9391842678491475,
'aturon.github.io': 0.9391810647723153,
'mattwarren.org': 0.9391805858225434,
'www.fastcompany.com': 0.939180509863255,
'reason.com': 0.9391096246428814,
'ramiro.org': 0.9390819120838915,
'www.nsa.gov': 0.9390773999685927,
'www.cs.cornell.edu': 0.9390402992969881,
'www.japantimes.co.jp': 0.9389848040411876,
'www.twitlonger.com': 0.9389760828355987,
'nginx.com': 0.9389702817402761,
'www.humblebundle.com': 0.9389598199049259,
'publicdomainreview.org': 0.9389382795628163,
'blog.nuclearsecrecy.com': 0.9388477073828405,
'pubmed.ncbi.nlm.nih.gov': 0.9388458315114857,
'qotoqot.com': 0.9388402176228176,
'pytorch.org': 0.93884021461205,
'jalammar.github.io': 0.9387918236299037,
'www.ubuntu.com': 0.9387702819164052,
'docs.python.org': 0.9387127698679522,
'google-opensource.blogspot.com': 0.9386938623019518,
'www.muckrock.com': 0.9386705784507212,
'dropbox.tech': 0.9386587283618043,
'www.jasondavies.com': 0.938652854203873,
'kite.com': 0.9386407930634363,
'media.netflix.com': 0.9386280658984226,
'www.indiehackers.com': 0.9386153918054686,
'blog.detectify.com': 0.9385636423663651,
'sixcolors.com': 0.9385558509083574,
'blog.qt.io': 0.9385531275962717,
'www.quirksmode.org': 0.938550073671435,
'community.namecheap.com': 0.9384479268346089,
'thecodist.com': 0.9383878278372055,
'www.hongkongfp.com': 0.9383791221566731,
'www.stephenfry.com': 0.9383772523821312,
'coffeescript.org': 0.9383760425896429,
'major.io': 0.9383578478495119,
'typesetinthefuture.com': 0.9383193833277421,
'www.nayuki.io': 0.9383068163327775,
'www.hscott.net': 0.9382713522578224,
'freebsdfoundation.blogspot.com': 0.9382576426660746,
'eng.uber.com': 0.93823010979631,
'www.cis.upenn.edu': 0.9382279776034698,
'open.nytimes.com': 0.9382271017946164,
'pyvideo.org': 0.9381709441441788,
'bugs.launchpad.net': 0.9381396092893388,
'ai.google': 0.9381254410971094,
'michael.stapelberg.ch': 0.9380985374502866,
'v8.dev': 0.9380973536567017,
'digiday.com': 0.9379742396004388,
'commandcenter.blogspot.com': 0.9379620301227771,
'tools.ietf.org': 0.9379500242941622,
'thewalrus.ca': 0.9378381775859617,
'discussions.apple.com': 0.9377827205296552,
'ignorethecode.net': 0.937717324851746,
'thread.gmane.org': 0.9376901941531085,
'sellercentral.amazon.com': 0.9376857021855104,
'www.sciencealert.com': 0.9376774589797282,
'slackhq.com': 0.937668726802223,
'blog.datawrapper.de': 0.9376182617912265,
'seriouspony.com': 0.9376182617912265,
'musicforprogramming.net': 0.9376182617912265,
'meowni.ca': 0.9375988271796821,
'www.redox-os.org': 0.9375879483491127,
'matt-welsh.blogspot.com': 0.937504055669681,
'minnie.tuhs.org': 0.9374529581205222,
'rondam.blogspot.com': 0.9374449372004028,
'warpspire.com': 0.9374088502294052,
'locusmag.com': 0.9373920717203438,
'www.bloombergview.com': 0.9373542498215184,
'www.damninteresting.com': 0.9373300477725874,
'joeyh.name': 0.9373215784915493,
'www.newstatesman.com': 0.9373076981929569,
'www.freebsd.org': 0.9372715325620732,
'andregarzia.com': 0.9372609577852429,
'caddyserver.com': 0.9372293681509105,
'kukuruku.co': 0.9372291284737027,
'www.jbkempf.com': 0.9372233784657422,
'pharo.org': 0.9371878563371911,
'netblocks.org': 0.9371746434498315,
'chromium.googlesource.com': 0.9371719999296655,
'phrack.org': 0.9371521351298083,
'www.codelitt.com': 0.9371442586175533,
'www.baekdal.com': 0.9371414507989151,
'www.ianvisits.co.uk': 0.9371147927406074,
'www.megaprocessor.com': 0.937090322173857,
'www.yosefk.com': 0.9370290915888351,
'www.mail-archive.com': 0.936986491855952,
'abe-winter.github.io': 0.9369860158978973,
'www.privateinternetaccess.com': 0.9369851095051764,
'nginx.org': 0.9369470816071664,
'www.intel.com': 0.9369378525632501,
'www.meltingasphalt.com': 0.9369168990825746,
'juliacomputing.com': 0.9368655840942366,
'rein.pk': 0.936849859435914,
'store.steampowered.com': 0.9368264728223703,
'www.osnews.com': 0.9368126058230493,
'ourworldindata.org': 0.9367969636489132,
'blogs.law.harvard.edu': 0.93674988182828,
'www-cs-faculty.stanford.edu': 0.9367219385122602,
'blog.joda.org': 0.9367157779314322,
'blog.stephsmith.io': 0.9366996246387541,
'www.linusakesson.net': 0.9366993761963279,
'blog.stackoverflow.com': 0.9366671617187088,
'learn.shayhowe.com': 0.9366612381135532,
'www.chrisfenton.com': 0.9366359092132852,
'alexdanco.com': 0.9366140932238274,
'bert.org': 0.9365314575663983,
'www.buyforlifeproducts.com': 0.9365183246073299,
'slate.com': 0.936478665114466,
'adventofcode.com': 0.9364182659002771,
'reuters.com': 0.9364022987065517,
'wordsandbuttons.online': 0.9364000242256861,
'blog.trailofbits.com': 0.9363648191551225,
'wwws.whitehouse.gov': 0.9362358726756497,
'www.saminiir.com': 0.9362027194085935,
'buttondown.email': 0.9361904266453563,
'www.polygon.com': 0.9361786245135907,
'ryanwaggoner.com': 0.9361726319454605,
'media.defense.gov': 0.9361328144489018,
'meta.wikimedia.org': 0.9361317597043629,
'journals.plos.org': 0.9360648852842034,
'users.ece.utexas.edu': 0.936037069978365,
'www.wired.co.uk': 0.9360115157816417,
'code.fb.com': 0.9359956830531816,
'blog.apnic.net': 0.9359552179738138,
'www.craftinginterpreters.com': 0.9359520862311609,
'blog.agilebits.com': 0.9359127041563384,
'ar.al': 0.935800217022854,
'tomdale.net': 0.9357863410079088,
'doublepulsar.com': 0.9357738329674509,
'www.keyvalues.com': 0.9357728930152032,
'android-developers.googleblog.com': 0.9357673687378363,
'www.w3.org': 0.9357529019651287,
'googlecloudplatform.blogspot.com': 0.9357511057780838,
'gravityandlevity.wordpress.com': 0.9357231633007936,
'zoekeating.tumblr.com': 0.9357009188423131,
'www.startupschool.org': 0.935616713926276,
'manishearth.github.io': 0.9355839880400915,
'android.googlesource.com': 0.9355592628654659,
'thegradient.pub': 0.9355328767434344,
'orgmode.org': 0.9355234169604703,
'www.wsws.org': 0.9355201201831335,
'citizenlab.org': 0.9355108358152496,
'blog.notdot.net': 0.9355042170349265,
'eduardo.intermeta.com.br': 0.9354939284774484,
'joreteg.com': 0.9354927179012053,
'issues.apache.org': 0.9354858969424931,
'www.hopkinsmedicine.org': 0.9354732723093431,
'stopa.io': 0.9354640998209997,
'rigsomelight.com': 0.9354524106623595,
'docs.djangoproject.com': 0.9354103729663484,
'americanaffairsjournal.org': 0.9353862576701931,
'forums.aws.amazon.com': 0.9353840941845297,
'jsfiddle.net': 0.9353603756027571,
'www.metalevel.at': 0.9353562467278554,
'hexatomium.github.io': 0.9353495280479229,
'andreasgal.com': 0.9353197730789111,
'c9x.me': 0.9352853909343191,
'danieltenner.com': 0.9352765782885712,
'danielcompton.net': 0.9352733622613952,
'netflixtechblog.com': 0.9352444459107085,
'www.vidarholen.net': 0.9352427951268865,
'themargins.substack.com': 0.9351811093645357,
'eli.thegreenplace.net': 0.9351789153416379,
'cybernews.com': 0.9351782885205434,
'blogs.smithsonianmag.com': 0.9351577172122721,
'rambo.codes': 0.9351531025036611,
'hackermonthly.com': 0.9351416375508997,
'www.sifive.com': 0.9351248366429326,
'plus.maths.org': 0.9351218966133573,
'incoherency.co.uk': 0.9351026897171553,
'www.erlang.org': 0.9350952451519492,
'circleci.com': 0.9350778094676155,
'catern.com': 0.9350750533895246,
'www.penny-arcade.com': 0.935065899223226,
'howistart.org': 0.935059575032339,
'www.middleeasteye.net': 0.9350521639498252,
'unlikekinds.com': 0.9350373828027304,
'www.videogameschronicle.com': 0.9350222208625469,
'www.roadandtrack.com': 0.9350012079006418,
'hyperallergic.com': 0.9349862159226987,
'blog.level3.com': 0.9349610044277377,
'www.os2museum.com': 0.934946992149697,
'soranews24.com': 0.9349080141492474,
'patshaughnessy.net': 0.9348951953108801,
'bill.harding.blog': 0.9348850338635601,
'blog.wikimedia.org': 0.9348790533709054,
'blog.plan99.net': 0.9348288512702018,
'www.gq.com': 0.9348208606560191,
'www.occrp.org': 0.9347939860561864,
'chronicle.com': 0.9347573157010687,
'mailarchive.ietf.org': 0.9347044422954542,
'devonzuegel.com': 0.9346592212655933,
'www.ben-evans.com': 0.9346544305688902,
'www.mediumdifficulty.com': 0.9345920431557654,
'www.fsf.org': 0.9345910501167647,
'learn.sparkfun.com': 0.9345230614646949,
'www.cs.princeton.edu': 0.934519791347039,
'dlang.org': 0.9345196775723191,
'www.templeos.org': 0.9345190524769148,
'code.blender.org': 0.9345112418088564,
'www.singhkays.com': 0.9344753142105167,
'www.repubblica.it': 0.9344751927302707,
'www.aerofs.com': 0.9344708443467806,
'www.lrb.co.uk': 0.9344236297924859,
'eng.lyft.com': 0.9344184442002427,
'jakearchibald.com': 0.9343998635925788,
'blog.youtube': 0.9343948680470999,
'richg42.blogspot.com': 0.9343774893527464,
'np.reddit.com': 0.9343550039538729,
'blog.easydns.org': 0.9343498007033786,
'caseyhandmer.wordpress.com': 0.9343353091084834,
'xiph.org': 0.9343228401668484,
'www.newscientist.com': 0.9343192570672714,
'www.nicolasbize.com': 0.9343100575008362,
'lineageos.org': 0.9342978511086933,
'briancallahan.net': 0.9342859603892616,
'blog.ltse.com': 0.9342402900090527,
'www.bryanbraun.com': 0.9341978241526933,
'blog.thea.codes': 0.9341759010883351,
'laphamsquarterly.org': 0.9341164747404341,
'code.mixpanel.com': 0.9340744683300655,
'fi.google.com': 0.9340569727587175,
'web.law.duke.edu': 0.9340401302760873,
'www.technologyreview.com': 0.93403386212399,
'iamtrask.github.io': 0.9340084148579416,
'news.ncsu.edu': 0.9340051509637882,
'www.acsh.org': 0.9339402739014211,
'repl.it': 0.9339084167158871,
'roscidus.com': 0.9339046191083922,
'blog.ghost.org': 0.9338973081214549,
'www.icann.org': 0.9338815927929048,
'digest.bps.org.uk': 0.9338601133658677,
'www.nj.com': 0.9338599954559099,
'www.90percentofeverything.com': 0.9338469537566946,
'lars.ingebrigtsen.no': 0.9338290703630162,
'harmful.cat-v.org': 0.933806937838211,
'www.ecma-international.org': 0.9337395167070882,
'engineering.riotgames.com': 0.9337282192512538,
'www.lexaloffle.com': 0.9337178769327638,
'www.noaa.gov': 0.9337158126136492,
'en.greatfire.org': 0.933688333932158,
'www.techcrunch.com': 0.9336781278522835,
'research.fb.com': 0.9336749976938089,
'karl-voit.at': 0.9336349938839054,
'fair.org': 0.9336252434606269,
'www.askamathematician.com': 0.9336136547935397,
'unixism.net': 0.9335805282449469,
'www.rockpapershotgun.com': 0.9335525383720096,
'developers.redhat.com': 0.9335410437326863,
'www.realdanlyons.com': 0.9335250444350301,
'play0ad.com': 0.9334866458886125,
'jwz.livejournal.com': 0.933468431827225,
'tedium.co': 0.9334408912081676,
'bitcoin.org': 0.9333591671165428,
'loup-vaillant.fr': 0.933343126525129,
'getpolarized.io': 0.9333167489840644,
'thewire.in': 0.9333119721322196,
'melmagazine.com': 0.9331968311771961,
'sambleckley.com': 0.9331955922865014,
'www.inc.com': 0.9331571334144225,
'david.heinemeierhansson.com': 0.9331531237247338,
'secondbreakfast.co': 0.9331357922232552,
'www.asahi.com': 0.933135633505878,
'devblogs.microsoft.com': 0.9331213750483387,
'www.nintendo.co.jp': 0.9331114844804387,
'annoying.technology': 0.9330709402145108,
'www.kiro7.com': 0.9330656946067081,
'bits.debian.org': 0.933059049771486,
'www.markshuttleworth.com': 0.9330005151597153,
'anders.unix.se': 0.9329923290063773,
'blogs.sciencemag.org': 0.9329560797100107,
'blog.bitbucket.org': 0.9329522711475849,
'blog.kubernetes.io': 0.9329204663072305,
'devdocs.io': 0.932893331029596,
'justinjackson.ca': 0.9327982068928226,
'corte.si': 0.9327949868127001,
'kotaku.com': 0.9327897653993101,
'cdn.ca9.uscourts.gov': 0.9327301316906936,
'forum.dlang.org': 0.9326997809060901,
'spacecraft.ssl.umd.edu': 0.9326765708381659,
'www.sciencedirect.com': 0.9326581830227065,
'neuralnetworksanddeeplearning.com': 0.9326459352221996,
'www.politico.com': 0.9326420745727587,
'notamonadtutorial.com': 0.9326372939715099,
'news.umich.edu': 0.9325964202141381,
'duck.co': 0.9325821384299628,
'lg.io': 0.9325797024877297,
'vorpus.org': 0.9325422497012895,
'objective-see.com': 0.9324963202009166,
'www.erlang-solutions.com': 0.9324852099306608,
'alexgaynor.net': 0.9324583238847478,
'sub.garrytan.com': 0.9324345058249506,
'www.dropbox.com': 0.9324087214510501,
'xkcd.com': 0.932407776611293,
'blog.codinghorror.com': 0.9323885522938681,
'wingolog.org': 0.9323392944843699,
'www.pipelinedb.com': 0.9322802131775214,
'iquantny.tumblr.com': 0.932275204317245,
'hbr.org': 0.9322550416701433,
'www.washington.edu': 0.9321755936890619,
'blog.nelhage.com': 0.9321436277497521,
'tonyarcieri.com': 0.932131262902639,
'www.geoffreylitt.com': 0.9321152273514123,
'viniciusvacanti.com': 0.9320235111465629,
'www.magicleap.com': 0.9319299806186524,
'www.igvita.com': 0.9318778756566067,
'chrisdone.com': 0.9318776390307953,
'stackshare.io': 0.9318592007853939,
'latacora.singles': 0.9318477865160556,
'www.dell.com': 0.9318413237056871,
'djhworld.github.io': 0.9317862165963432,
'www.infoq.com': 0.9317836684869253,
'papersplease.org': 0.931751319080821,
'thepiratebay.org': 0.9316799771304791,
'calnewport.com': 0.9316572134183647,
'krausefx.com': 0.9316559307678887,
'blog.circleci.com': 0.9316255482754808,
'blog.cognitect.com': 0.9315784532891852,
'www.epicgames.com': 0.9315429629794932,
'scopeblog.stanford.edu': 0.9315377292068189,
'www.masswerk.at': 0.9314679327889084,
'0x65.dev': 0.9314351318900396,
'www.engadget.com': 0.9314186236786943,
'dangerousminds.net': 0.93133503311817,
'www.businessweek.com': 0.9313339649208416,
'meta.slashdot.org': 0.9313051774463162,
'john-millikin.com': 0.9312674134931619,
'www.raptitude.com': 0.9312570267721416,
'www.mersenne.org': 0.9312544294826364,
'blog.reverberate.org': 0.931247035488847,
'alexisohanian.com': 0.9311960984991072,
'electrek.co': 0.9311696023394275,
'www.jeanhsu.com': 0.9311256778507919,
'svelte.technology': 0.9311186925197594,
'programmers.stackexchange.com': 0.9311091436248506,
'discuss.ocaml.org': 0.9311072488914112,
'nedbatchelder.com': 0.9310954334270662,
'www.haskellforall.com': 0.931051756515657,
'retrocomputing.stackexchange.com': 0.9310144532795315,
'catonmat.net': 0.9309769810532454,
'vole.wtf': 0.9309344769161616,
'www.scribd.com': 0.9309198232076064,
'yurichev.com': 0.9309143601885577,
'joel.is': 0.9309094060954536,
'yager.io': 0.9308009815388928,
'boilingsteam.com': 0.9307989777220242,
'gamasutra.com': 0.9307558317808844,
'www.bigmessowires.com': 0.9307114429921958,
'fail0verflow.com': 0.930710252209961,
'news.harvard.edu': 0.9306854591781405,
'beagleboard.org': 0.9306733479517327,
'www.floridatoday.com': 0.9306710239151541,
'www.popehat.com': 0.9306350599333171,
'blog.tutorspree.com': 0.9306202167888316,
'daeken.com': 0.9305668550552411,
'techstartups.com': 0.9305594478826466,
'www.wikitribune.com': 0.930515310124751,
'www.chess.com': 0.9304928063593388,
'newsoffice.mit.edu': 0.9304895970361445,
'wired.com': 0.9304827864643992,
'sealedabstract.com': 0.9304819216359799,
'0x46.net': 0.9304806845007771,
'unsplash.com': 0.9304637690166982,
'my.opera.com': 0.930450954749869,
'nerds.airbnb.com': 0.930435238682839,
'longreads.com': 0.9304235614345844,
'www.dispatch.com': 0.9303745139221324,
'www.fda.gov': 0.9303439075870598,
'lists.centos.org': 0.9303434013425956,
'adgrok.com': 0.9303388809532392,
'thinkingthrough.substack.com': 0.9303337259943262,
'mcfunley.com': 0.9303298670383908,
'beginners.re': 0.9302874468948247,
'khan.github.io': 0.9302341812757136,
'joshworth.com': 0.9302242175504744,
'lamport.azurewebsites.net': 0.9302036598082833,
'calv.info': 0.9301915155599901,
'jdk.java.net': 0.9301739684537851,
'keepassxc.org': 0.9301626315930274,
'thepiratebay.se': 0.9301506979843667,
'metatalk.metafilter.com': 0.9301352170901589,
'gregable.com': 0.9301219028430462,
'newrepublic.com': 0.9301135242301953,
'cryptome.org': 0.9301123136352331,
'www.smashcompany.com': 0.9301070862923108,
'mixergy.com': 0.9301017907321477,
'blog.bolt.io': 0.9300384028733197,
'lesswrong.com': 0.9299875078740503,
'www.recode.net': 0.9299803603149057,
'baremetrics.com': 0.9299780878801276,
'idea.popcount.org': 0.92997751068095,
'blog.nodejs.org': 0.9299713179632104,
'googlecode.blogspot.com': 0.9299671011900217,
'dustincurtis.com': 0.929962639691764,
'capnproto.org': 0.9299597749554257,
'samy.pl': 0.9299580709168783,
'fivethirtyeight.com': 0.9299521782157305,
'babeljs.io': 0.9299138678024598,
'venturebeat.com': 0.9299090359763643,
'bartoszmilewski.com': 0.9298782796519226,
'interactivepython.org': 0.9298650433817537,
'dbmsmusings.blogspot.com': 0.9298460930808272,
'feross.org': 0.9298143975383142,
'www.nateberkopec.com': 0.9297897921141977,
'www.bnnbloomberg.ca': 0.9297228959508879,
'sourcegraph.com': 0.9296937397028914,
'www.visualstudio.com': 0.9296781764315996,
'www.dampfkraft.com': 0.9296648658804626,
'sockpuppet.org': 0.9296582232650001,
'vuejs.org': 0.9296438971766601,
'mozillagfx.wordpress.com': 0.929620568345932,
'www.tomshardware.com': 0.9295421933042515,
'go.googlesource.com': 0.929539952433035,
'every.to': 0.9295113508234609,
'community.signalusers.org': 0.9295104306484039,
'features.propublica.org': 0.9294651901563945,
'trigger.io': 0.9294592741801949,
'blog.wolfram.com': 0.9294394447255451,
'science.nasa.gov': 0.9294364753474607,
'www.globalcitizen.org': 0.9294355607283274,
'brennan.io': 0.9294094367716018,
'www.princeton.edu': 0.9293972895467508,
'www.typescriptlang.org': 0.9293477418572129,
'blog.hboeck.de': 0.9293102195703717,
'www.kickstarter.com': 0.9293062867407036,
'blog.siftscience.com': 0.9293015387633619,
'blog.jetbrains.com': 0.9292407139248418,
'mag.newsweek.com': 0.9291970802919708,
'libcom.org': 0.9291864757522856,
'pudding.cool': 0.9291654244837007,
'www.geek.com': 0.9291554612391715,
'zedshaw.com': 0.9291438333180465,
'mailman.nginx.org': 0.9291339657438804,
'www.techspot.com': 0.9291295247017188,
'www.popularmechanics.com': 0.9291147919612897,
'www.nbcnews.com': 0.9290863046432193,
'acesounderglass.com': 0.929082768087054,
'stitcher.io': 0.9290511337189973,
'www.texasmonthly.com': 0.92902453382746,
'mattmaroon.com': 0.9290233630671553,
'www.gabrielgambetta.com': 0.9290195296996202,
'm.dw.com': 0.9290040778772607,
'www.mit.edu': 0.9289941184727176,
'tratt.net': 0.9289824042121754,
'machinelearning.apple.com': 0.928981486544923,
'blog.posterous.com': 0.9289604924288977,
'oxide.computer': 0.9289445169614982,
'trackchanges.postlight.com': 0.928879705271154,
'codahale.com': 0.9288692041231931,
'emberjs.com': 0.9288495718673251,
'applehelpwriter.com': 0.9287812041116006,
'shaunlebron.github.io': 0.9286838540725274,
'open.buffer.com': 0.9286596873950792,
'www.expatsoftware.com': 0.9286078722400491,
'edgedb.com': 0.928573715100738,
'etherpad.com': 0.9285692478417551,
'muratbuffalo.blogspot.com': 0.9285656789226785,
'fgiesen.wordpress.com': 0.9285630732312407,
'www.firebase.com': 0.9285546878582209,
'blog.thestartuptoolkit.com': 0.9285487410065263,
'cloudirregular.substack.com': 0.9285463604002121,
'flynn.io': 0.9285393525237848,
'www.eevblog.com': 0.9285287752743465,
'www.balena.io': 0.9284888543895996,
'rt.com': 0.9284735976773327,
'www.bell-labs.com': 0.9284678162678174,
'www.persuasion.community': 0.9283945188643842,
'eugeneyan.com': 0.9283782094499546,
'80000hours.org': 0.9283725313298165,
'www.zerotier.com': 0.9283630467998711,
'residentcontrarian.substack.com': 0.9283528817911175,
'httptoolkit.tech': 0.9283475144274776,
'www.ekioh.com': 0.9283450473725626,
'www.cia.gov': 0.9283207863801844,
'ryanleecarson.tumblr.com': 0.9283074648928308,
'nickcraver.com': 0.9282875328539527,
'syonyk.blogspot.com': 0.9282602083788998,
'www.simonsfoundation.org': 0.9282557678633827,
'www.artsy.net': 0.9281354567126978,
'tryretool.com': 0.9281338223307144,
'www.bloombergquint.com': 0.928121311983667,
'johncarlosbaez.wordpress.com': 0.9281129934414499,
'vas3k.com': 0.9280973443374919,
'100r.co': 0.9280948851000741,
'blogs.adobe.com': 0.928088150929166,
'www.motherjones.com': 0.9280843595076367,
'www.nasdaq.com': 0.9280594095034941,
'scarybeastsecurity.blogspot.com': 0.9280588046711343,
'bitcointalk.org': 0.9280388108667238,
'hackershelf.com': 0.9280341162118197,
'news.nationalgeographic.com': 0.9279960640653149,
'myactivity.google.com': 0.9279454537052139,
'blog.nightly.mozilla.org': 0.9279411885747856,
'slashcrypto.org': 0.9279346210995543,
'www.phoboslab.org': 0.927902020777888,
'tosdr.org': 0.9278899901888288,
'www.fiercebiotech.com': 0.9278829786533511,
'pomax.github.io': 0.9278608982168398,
'www.haskell.org': 0.9278608403233632,
'tribunemag.co.uk': 0.927831772588806,
'alexeymk.com': 0.9278193647918856,
'www.cs.vu.nl': 0.9277449579009552,
'news.cnet.com': 0.927730426058379,
'www.caltech.edu': 0.9277282839323142,
'www.azarask.in': 0.9276539782249613,
'gigaom.com': 0.9276147331753969,
'pdos.csail.mit.edu': 0.9276091373570645,
'www.virtualbox.org': 0.9275378961204167,
'increment.com': 0.9275213097235798,
'blog.janestreet.com': 0.9274833694141223,
'www.0x65.dev': 0.9274613287943144,
'textslashplain.com': 0.9274390770834591,
'www.planetary.org': 0.927429526861741,
'allthingsd.com': 0.9274270826887906,
'clojure.com': 0.9274173307178728,
'git.postgresql.org': 0.9274089095546911,
'www.howtogeek.com': 0.927406503902944,
'vr.google.com': 0.9273952095808383,
'blog.kapeli.com': 0.9273878300515732,
'quickdraw.withgoogle.com': 0.9273611719298221,
'erkaman.github.io': 0.9273454992466085,
'discussions.agilebits.com': 0.9273075851865854,
'fauna.com': 0.9272945460775945,
'popehat.com': 0.9272692801582506,
'blog.sandstorm.io': 0.9272550707158864,
'www.openvim.com': 0.9272549340276682,
'www.panic.com': 0.9272306781608166,
'waymo.com': 0.9271995570219662,
'blog.eladgil.com': 0.9271995137279543,
'uk.reuters.com': 0.9271613384786328,
'www.openstreetmap.org': 0.9270974234890762,
'projekte.sueddeutsche.de': 0.9270885496912139,
'whytheluckystiff.net': 0.9270441393847589,
'the-paper-trail.org': 0.9269567323760859,
'medium.dave-bailey.com': 0.9269393622635275,
'www.kadavy.net': 0.9269319992417328,
'symbo1ics.com': 0.9269313741433315,
'erikrood.com': 0.926847335423336,
'text.npr.org': 0.9268424682157039,
'www.gnome.org': 0.9268089290588881,
'www.arp242.net': 0.9267978785624127,
'nvlabs.github.io': 0.9267181712194288,
'craigkerstiens.com': 0.9267131513833771,
'www.evanjones.ca': 0.9266986776879761,
'johnsalvatier.org': 0.926681783824641,
'notes.eatonphil.com': 0.9266798244317584,
'cityobservatory.org': 0.9266470517985763,
'www.autocar.co.uk': 0.9266248470761401,
'store.google.com': 0.9265910046118487,
'blog.alinelerner.com': 0.9265889786185729,
'a.singlediv.com': 0.9265847984290865,
'www.city-journal.org': 0.926568347566864,
'blog.bugsnag.com': 0.9265163617009136,
'blog.steveklabnik.com': 0.9265012336320916,
'www.ox.ac.uk': 0.9264790742628204,
'www.jgc.org': 0.92646604158067,
'maraoz.com': 0.9264393162283295,
'www.somethingsimilar.com': 0.9264338569528214,
'electronupdate.blogspot.com': 0.9264036418816388,
'productforums.google.com': 0.9264016849144228,
'macro.ycombinator.com': 0.9263799756741565,
'damienkatz.net': 0.9263697789667,
'www.moserware.com': 0.9263363409676512,
'nikic.github.io': 0.9263232996382066,
'www.elastic.co': 0.9263050573116817,
'www.cs.utexas.edu': 0.9263034743689128,
'www.bhorowitz.com': 0.9262984080290757,
'tenthousandmeters.com': 0.9262841230837943,
'about.fb.com': 0.9262484440409288,
'www.fmwconcepts.com': 0.9262450791824973,
'recraigslist.com': 0.9262425641046296,
'www.codingjohnson.com': 0.9262281542120302,
'www.aftenposten.no': 0.9262270297289266,
'www-math.mit.edu': 0.9261881534042337,
'blog.1password.com': 0.9261268579416305,
'neugierig.org': 0.9261248188625917,
'01.org': 0.9261246392123041,
'jalopnik.com': 0.9261214539229355,
'www.mnot.net': 0.9261212790512424,
'jmoiron.net': 0.9260837453092551,
'bjoern.brembs.net': 0.9260528919272174,
'en.community.dell.com': 0.9260523403380299,
'yieldthought.com': 0.9260498267973828,
'news.gallup.com': 0.9260444722659747,
'www.quora.com': 0.9260005301179841,
'www.boingboing.net': 0.9259933447507237,
'plusbryan.com': 0.9259774485563826,
'webcache.googleusercontent.com': 0.9259579793050731,
'rameznaam.com': 0.9259525603398252,
'dev.windows.com': 0.9259474200369756,
'www.cnn.com': 0.9259450168290682,
'ij.org': 0.9259017125754639,
'hackingdistributed.com': 0.9258935189335176,
'aosabook.org': 0.9258707865599977,
'webtransparency.cs.princeton.edu': 0.9258302497498668,
'nikcub.appspot.com': 0.9257984690273875,
'www.fcc.gov': 0.9257766215669876,
'multithreaded.stitchfix.com': 0.9256971609820355,
'www.ybrikman.com': 0.9256541091635649,
'www.neh.gov': 0.925636711625259,
'abc.xyz': 0.9256328360448464,
'thecorrespondent.com': 0.9256254843843928,
'www.techempower.com': 0.9256081793071045,
'letsencrypt.status.io': 0.9255938806311487,
'grail.cs.washington.edu': 0.925560977684432,
'syncthing.net': 0.9255454895835606,
'torontolife.com': 0.9255329261936903,
'www.notion.so': 0.9255238074027249,
'www.empiricalzeal.com': 0.925521405860939,
'www.brown.edu': 0.925521405860939,
'dustri.org': 0.9255123842761099,
'restofworld.org': 0.9254846646984972,
'meta.stackoverflow.com': 0.9254699602796586,
'www.mikeash.com': 0.9254652180552938,
'emacsair.me': 0.9254649535069708,
'fossil-scm.org': 0.9254641195091267,
'drive.google.com': 0.9254499381167396,
'rauchg.com': 0.9254066104836953,
'www.oregonlive.com': 0.9254001048984968,
'madebyevan.com': 0.9253946331608134,
'bitbucket.org': 0.925357050068107,
'justinvincent.com': 0.9253388274119266,
'shape-of-code.coding-guidelines.com': 0.9253230426310416,
'www.journals.uchicago.edu': 0.9252909862687759,
'www.hodinkee.com': 0.925286129490443,
'blog.getprismatic.com': 0.9252627177795818,
'android-developers.blogspot.com': 0.9252451042228654,
'jeremyaboyd.com': 0.9252351792387165,
'mozilla.github.io': 0.9252103780415603,
'www.pcgamer.com': 0.9251974859644215,
'tutanota.com': 0.9251840115849959,
'review.chicagobooth.edu': 0.9251560984825087,
'programmingisterrible.com': 0.9251358332003701,
'zainamro.com': 0.9251300763149733,
'www.locusmag.com': 0.9251291781651696,
'waxy.org': 0.9250938835073874,
'bentilly.blogspot.com': 0.9250916663441209,
'hisham.hm': 0.925069607765931,
'www.foundersatwork.com': 0.9250661963070413,
'neil.computer': 0.9250487557545186,
'youmightnotneedjquery.com': 0.9250308951241109,
'sparktoro.com': 0.9250076259146361,
'supporters.eff.org': 0.9250027991144959,
'www.youtube.com': 0.9250024575604987,
'www.django-rest-framework.org': 0.9249975958434261,
'fuse.wikichip.org': 0.9249946532155069,
'alexanderell.is': 0.9249832648395125,
'www.jetbrains.com': 0.9249775253793479,
'rootsofprogress.org': 0.9249775157472907,
'blog.jonlu.ca': 0.9249720526798552,
'grafana.com': 0.9249363529284085,
'joinpeertube.org': 0.92492611085508,
'trac.torproject.org': 0.92492568982512,
'www.kapwing.com': 0.9249078774898674,
'jgthms.com': 0.9249053279219519,
'www.lighterra.com': 0.9248749337230386,
'www.bzarg.com': 0.9248749337230386,
'itep.org': 0.924864446165763,
'xuanji.appspot.com': 0.924862527414928,
'dood.al': 0.9248541691533322,
'seriot.ch': 0.924849625532851,
'go-review.googlesource.com': 0.9248263431754091,
'benjamincongdon.me': 0.924814411910629,
'coderwall.com': 0.9247619851237905,
'www.betabeat.com': 0.9247390428437294,
'cryto.net': 0.924723051999456,
'blogs.bl.uk': 0.9247163678317087,
'datatracker.ietf.org': 0.924674483012171,
'www.b-list.org': 0.9246677557779804,
'apple.stackexchange.com': 0.924658393133553,
'www.mozilla.com': 0.9246560951725633,
'www.mailpile.is': 0.9246491366569076,
'see.stanford.edu': 0.9246092475112012,
'www.chiark.greenend.org.uk': 0.9246053483860566,
'groups.csail.mit.edu': 0.9245841594300752,
'www.arcfn.com': 0.9245434848298174,
'urbit.org': 0.9245310249759199,
'www.docker.com': 0.9245172226554722,
'eprint.iacr.org': 0.9245057684882811,
'dl.acm.org': 0.9244842820223327,
'blog.sessionstack.com': 0.9244816364620212,
'www.breck-mckye.com': 0.9244526176174389,
'www.aosabook.org': 0.9244287224439842,
'prospect.org': 0.9243862527888093,
'micro-editor.github.io': 0.9243497443248759,
'banksimple.com': 0.9243293845078053,
'practicaltypography.com': 0.9243063840022637,
'blogmaverick.com': 0.9242840626877411,
'instagram-engineering.com': 0.9242395677129099,
'theartofmachinery.com': 0.9242293335123167,
'www.cpushack.com': 0.9241635632263332,
'guzey.com': 0.9241634452596086,
'morrick.me': 0.9241492448464671,
'chanind.github.io': 0.9241272337118139,
'blog.emojipedia.org': 0.9241070688292589,
'arjunsreedharan.org': 0.9240265801481968,
'superuser.com': 0.9240150880034484,
'zulko.github.io': 0.9240082509797493,
'blog.fosketts.net': 0.923996721917042,
'darkpatterns.org': 0.9239687269035174,
'www.ulisp.com': 0.9239409936091142,
'herbsutter.com': 0.9238650642442959,
'research.google.com': 0.9238529070048137,
'www.darkcoding.net': 0.9238452017940412,
'latacora.micro.blog': 0.9237958672976048,
'www.drdobbs.com': 0.9237812616938089,
'www.hollywoodreporter.com': 0.9237805856737081,
'www.nginx.com': 0.9237795637254299,
'thestartupfoundry.com': 0.9237690021040942,
'blog.fastmail.fm': 0.923729095971477,
'wit.ai': 0.9237161044120704,
'www.devttys0.com': 0.923715140193862,
'www.qubes-os.org': 0.9237100158767054,
'www.robinsloan.com': 0.9236869500281771,
'timetobleed.com': 0.9236836576824262,
'blog.jitbit.com': 0.9236712552599629,
'gowers.wordpress.com': 0.9236675842265081,
'www.pnas.org': 0.9236623882869138,
'mrale.ph': 0.9236621608796006,
'play.freeciv.org': 0.9235744210893093,
'www.blackhat.com': 0.9235710982881401,
'laforge.gnumonks.org': 0.9234795250933899,
'marker.medium.com': 0.9234758009863355,
'www.sfchronicle.com': 0.9234493747928338,
'tutorial.math.lamar.edu': 0.9234097897078604,
'www.holovaty.com': 0.9233603823514369,
'www.cringely.com': 0.9232863734060375,
'www.folklore.org': 0.9232666029542594,
'apnews.com': 0.9232599354911131,
'devblogs.nvidia.com': 0.9231813399979636,
'www.linuxvoice.com': 0.9231576608319696,
'papers.ssrn.com': 0.9231181748415175,
'brianschrader.com': 0.9231068216939985,
'www.kreci.net': 0.923106349521949,
'www.nngroup.com': 0.9230948271379453,
'funcall.blogspot.com': 0.9230679265670205,
'www.jasonshen.com': 0.9230677977899461,
'lists.freedesktop.org': 0.9230222479984594,
'nodejs.org': 0.92301963271793,
'www.graalvm.org': 0.9230158730158731,
'blog.mozilla.com': 0.923006927671048,
'www.dailydot.com': 0.9230068718762484,
'lbrandy.com': 0.9229742803918614,
'erratasec.blogspot.com': 0.9229733794911794,
'www.chemistryworld.com': 0.9229661370816306,
'www.solipsys.co.uk': 0.9229220876704523,
'layerci.com': 0.9229159114633821,
'www.stellar.org': 0.9229142185891028,
'etherscan.io': 0.9228765015817684,
'www.macrumors.com': 0.922864948439789,
'vim-adventures.com': 0.9228321400159109,
'mailinator.blogspot.com': 0.9228075138073343,
'www.macleans.ca': 0.9227372105459439,
'blog.appcanary.com': 0.9227212708074342,
'www.stallman.org': 0.9227212708074342,
'briancarper.net': 0.9227017702961716,
'joehewitt.com': 0.9226894917334,
'sfconservancy.org': 0.9226892388952699,
'mariovittone.com': 0.9226475279106858,
'500hats.com': 0.9226369106547709,
'blog.tailwindcss.com': 0.9226012568165967,
'dilbert.com': 0.9226004268083031,
'mewo2.com': 0.9225891352707621,
'blog.getpebble.com': 0.9225467649630955,
'preshing.com': 0.922540762544208,
'www.medievalists.net': 0.9225218713450589,
'daily.jstor.org': 0.9225038762181162,
'www.lynalden.com': 0.922503215089193,
'hindenburgresearch.com': 0.9225025964121197,
'joindiaspora.com': 0.9224924773526838,
'chadfowler.com': 0.922481090652384,
'git.sr.ht': 0.9224628516537063,
'twitter.github.io': 0.9224447323362635,
'www.rackspace.com': 0.92242477632422,
'www.science.org': 0.9224216244734532,
'www.esquire.com': 0.9224172218178369,
'augustl.com': 0.9224131314072078,
'news.cornell.edu': 0.9223803694889144,
'paulrouget.com': 0.9223675768999002,
'joelgrus.com': 0.9222919914652804,
'electronjs.org': 0.9222709958305552,
'williamedwardscoder.tumblr.com': 0.9222429833639271,
'fastmail.blog': 0.9222426931980966,
'trixter.oldskool.org': 0.9222240643067603,
'palladiummag.com': 0.9221161388490351,
'www.dwmkerr.com': 0.9220928814524835,
'www.blueorigin.com': 0.9220497550322969,
'blogs.perl.org': 0.922041527915435,
'peter.bourgon.org': 0.9220345737360726,
'boats.gitlab.io': 0.9220270248890161,
'news.usc.edu': 0.9220177070210377,
'www.media.uzh.ch': 0.9220147439869917,
'rubini.us': 0.9220102711604998,
'ognjen.io': 0.9219863609200528,
'www.dragonflybsd.org': 0.9219689458962362,
'time.com': 0.9219350046913536,
'www.vagrantup.com': 0.9219259642840062,
'reclaimthenet.org': 0.921903148362967,
'blogs.lse.ac.uk': 0.921869679768062,
'threader.app': 0.9218497168940121,
'pgdash.io': 0.9217440948884412,
'www.namecheap.com': 0.9217409250741742,
'www.mtgox.com': 0.9217171114471724,
'blog.filepicker.io': 0.9217169081929824,
'mullvad.net': 0.9216909513671212,
'm.signalvnoise.com': 0.9216188131338918,
'gamerant.com': 0.9215933437650176,
'www.computerhistory.org': 0.9215810214981082,
'www.bennettnotes.com': 0.9215678399791286,
'codegolf.stackexchange.com': 0.9215396697671643,
'www.amnesty.org': 0.9214886045349302,
'quip.com': 0.9214846662604561,
'www.balsamiq.com': 0.9214784555047167,
'www.conwaylife.com': 0.9214510848919855,
'quoteinvestigator.com': 0.9214277867192284,
'www.stephanboyer.com': 0.9214274950425619,
'www.ibm.com': 0.9214034975155869,
'kkovacs.eu': 0.9213904944480831,
'www.cs.virginia.edu': 0.9213138750457075,
'blog.heapanalytics.com': 0.9212598460858725,
'www.khronos.org': 0.9212580247777246,
'infotrope.net': 0.9212298613344002,
'r6.ca': 0.9212165287852896,
'elemental.medium.com': 0.9211754093730133,
'standardnotes.org': 0.9211590022701598,
'www.thanassis.space': 0.9211469562862422,
'www.essential.com': 0.9211382113821138,
'blog.checkpoint.com': 0.9210707601354348,
'firebase.googleblog.com': 0.9210648800435821,
'theamericanscholar.org': 0.9210499166083035,
'mxb.dev': 0.9210070283654466,
'arrow.apache.org': 0.9209592797673791,
'moderncrypto.org': 0.9209488760965047,
'edition.cnn.com': 0.9209449218392389,
'penny-arcade.com': 0.9208903657692026,
'www.gresearch.co.uk': 0.9208877326699497,
'news.rapgenius.com': 0.9208723565849319,
'duartes.org': 0.9208704095563033,
'access.redhat.com': 0.9208698327943827,
'dealbook.nytimes.com': 0.9208660000281566,
'www.robinwieruch.de': 0.9208515295663545,
'blog.viktomas.com': 0.9208241650543216,
'www.thenation.com': 0.9208156841051072,
'evilmartians.com': 0.9207891062516377,
'www.unixsheikh.com': 0.9207398153686466,
'david.weebly.com': 0.9207376999076698,
'lanyrd.com': 0.9207234118522407,
'www.slowboring.com': 0.9206921783803773,
'unix.stackexchange.com': 0.9206801573666252,
'seldo.com': 0.9206416671015177,
'www.elischiff.com': 0.920635133939363,
'www.arrl.org': 0.9206234339791557,
'www.spacetelescope.org': 0.9205965394020418,
'writing.markchristian.org': 0.9205569205569206,
'elaineou.com': 0.9205345242508725,
'blog.lastpass.com': 0.9205320290125674,
'netzpolitik.org': 0.9205194544865914,
'ghuntley.com': 0.9205047884758858,
'knightcolumbia.org': 0.9205029147967208,
'www.comparitech.com': 0.9204765637809545,
'scanlime.org': 0.9204328320221652,
'www.sfgate.com': 0.920422056660529,
'eklitzke.org': 0.9203820542379546,
'www.scala-lang.org': 0.9203673159816037,
'people.eecs.berkeley.edu': 0.9203504139456036,
'bits.blogs.nytimes.com': 0.9203453783143634,
'floooh.github.io': 0.9203399579895484,
'obscuritory.com': 0.9203310173661203,
'habitatchronicles.com': 0.9203279953622738,
'www.giftrocket.com': 0.9203035040887337,
'www.debugbear.com': 0.9202964328943275,
'makezine.com': 0.920278418856363,
'www.slideshare.net': 0.920271947086287,
'sandlab.cs.uchicago.edu': 0.9202302631578948,
'timkadlec.com': 0.9202082523474097,
'momentjs.com': 0.9201811740344046,
'www.euractiv.com': 0.9201333156342562,
'www.daniellemorrill.com': 0.9201158212493131,
'writings.stephenwolfram.com': 0.9200560602651936,
'www.gkogan.co': 0.9200121929029601,
'needwant.com': 0.9199907538489269,
'www.medialens.org': 0.9199880352359148,
'blog.wepay.com': 0.9199344757712936,
'maps.google.com': 0.9199182118968083,
'www.dustmop.io': 0.9199009083402148,
'www.comp.nus.edu.sg': 0.9198995227263196,
'wiki.ubuntu.com': 0.9198686814026275,
'remotebase.io': 0.9198347107438016,
'www.37signals.com': 0.9198326817403155,
'alumni.media.mit.edu': 0.9198312121350687,
'www.smashingmagazine.com': 0.9197923257499647,
'kylehalladay.com': 0.9197075896809243,
'fortune.com': 0.9196957257034758,
'fosterelli.co': 0.9196146019753699,
'www.braveclojure.com': 0.9196013394110684,
'anvaka.github.io': 0.919572990689544,
'nomasters.io': 0.9195688225538972,
'reasonstobecheerful.world': 0.9195521506144941,
'thonyc.wordpress.com': 0.9195493663014705,
'x264dev.multimedia.cx': 0.9195307216192476,
'www.androidpolice.com': 0.9194976713702179,
'www.ribbonfarm.com': 0.919494859687766,
'workplace.stackexchange.com': 0.9194815293298146,
'jcs.org': 0.9194811525194942,
'www.codusoperandi.com': 0.9194731766129004,
'behind-the-enemy-lines.blogspot.com': 0.9194452174142858,
'peternixey.com': 0.9194135955492282,
'www.theplayerstribune.com': 0.9194126818562519,
'basho.com': 0.9194014734604079,
'www.curiousgnu.com': 0.9193682460515379,
'm.imgur.com': 0.9193616517608175,
'gafferongames.com': 0.9193033055945021,
'www.idontplaydarts.com': 0.9192963504549827,
'www.forrestthewoods.com': 0.9192741785228133,
'earthquake.usgs.gov': 0.9192641098431819,
'blog.bitpay.com': 0.9192625409487661,
'www.brookings.edu': 0.9192507068377601,
'18f.gsa.gov': 0.9192284715908775,
'www.minecraftforum.net': 0.9192112937034953,
'www.lyft.com': 0.9191683431386107,
'hhvm.com': 0.9191453265158583,
'blog.wolfire.com': 0.9191450229699666,
'www.usatoday.com': 0.9191347594941367,
'nbviewer.ipython.org': 0.9191131949298217,
'reddit.com': 0.9190800965203674,
'www.eenewseurope.com': 0.9190688625693296,
'www.smh.com.au': 0.9190104812043022,
'www.figma.com': 0.9189780967661552,
'www.yubico.com': 0.9189671558387075,
'larrysanger.org': 0.9189574986097159,
'www.cs.berkeley.edu': 0.9189369697206728,
'www.qualys.com': 0.9188953942561601,
'www.charlieharrington.com': 0.9188769529436779,
'casual-effects.com': 0.91886846135451,
'alexkrupp.typepad.com': 0.9188063903443455,
'blog.macromates.com': 0.9187968191722226,
'beza1e1.tuxen.de': 0.9187934225268114,
'www.marketwatch.com': 0.9187035877780302,
'dragan.rocks': 0.9186972641760346,
'scottmeyers.blogspot.com': 0.9186922599750114,
'brson.github.io': 0.9186844191523297,
'kevq.uk': 0.9186712741984764,
'tuhdo.github.io': 0.9186408903885764,
'blog.ploeh.dk': 0.918640800603575,
'www.spoon-tamago.com': 0.918580162432909,
'www.maizure.org': 0.9185797770662466,
'californiaglobe.com': 0.9185558354324097,
'yarchive.net': 0.9185157064745997,
'helpx.adobe.com': 0.918505551895513,
'sourceforge.net': 0.9184879275921072,
'gail.com': 0.9184873949579831,
'snikt.net': 0.9184873949579831,
'blog.plaid.com': 0.9184773866383011,
'ebb.org': 0.9184174897915893,
'insomnia.rest': 0.9184173422143894,
'www.biorxiv.org': 0.9184172443974019,
'cmdrtaco.net': 0.91840628090816,
'tom7.org': 0.9183916897269802,
'taler.net': 0.9183916897269802,
'docs.rs': 0.9183858344184437,
'www.ioccc.org': 0.9183852068279685,
'sites.google.com': 0.9183471189460477,
'domainnamewire.com': 0.918330600390645,
'dafoster.net': 0.9183054021034863,
'spreadsheets.google.com': 0.9182933889184417,
'www.abc.net.au': 0.9182274259544088,
'www.boringcompany.com': 0.918212478920742,
'www.bankofengland.co.uk': 0.9181974227639322,
'blog.ntpsec.org': 0.9181951887885417,
'huyenchip.com': 0.9181777870982438,
'pganalyze.com': 0.9181448902734191,
'startupschool.org': 0.9181360100179967,
'bl.ocks.org': 0.9180840436489085,
'codeamigo.dev': 0.9180588577925552,
'www.lesswrong.com': 0.9179747514962872,
'blog.x.company': 0.9179518350159086,
'inews.co.uk': 0.9179395572482176,
'www.talyarkoni.org': 0.9179286455845682,
'amazon.com': 0.917918074063271,
'pijul.org': 0.9179029713542893,
'www.thegatesnotes.com': 0.917896287388944,
'blogs.reuters.com': 0.9178784513377969,
'bartongeorge.io': 0.9178137271960294,
'arslan.io': 0.9178051685037625,
'bytecodealliance.org': 0.917754880834866,
'journal.paul.querna.org': 0.9177358613483362,
'jasonlbaptiste.com': 0.9177192783299879,
'www.sharelatex.com': 0.9176170787075834,
'www.jessesquires.com': 0.9176086328709556,
'iquilezles.org': 0.9175867575102823,
'yeokhengmeng.com': 0.9175790835471582,
'software.rajivprab.com': 0.9175657081772484,
'news.utexas.edu': 0.917544773761157,
'www.fbi.gov': 0.9175354761400789,
'sirupsen.com': 0.9175322740193886,
'recode.net': 0.9175305114277775,
'www.studyfinds.org': 0.9175005975481785,
'rdegges.com': 0.9174949072553962,
'cemerick.com': 0.9174490483219848,
'cr.openjdk.java.net': 0.9174209932539029,
'mapbox.com': 0.9174119653310046,
'www.lihaoyi.com': 0.917356761960145,
'www.ams.org': 0.9173555467413949,
'golem.ph.utexas.edu': 0.9173444265784106,
'www.oilshell.org': 0.9173296046923102,
'www.darktable.org': 0.9173295566709162,
'www.madinamerica.com': 0.9173266085137967,
'blog.scrapinghub.com': 0.9173116907321415,
'www.sheshbabu.com': 0.9172725950029555,
'nickdrozd.github.io': 0.9172699096194682,
'diveintomark.org': 0.9172600663351143,
'modernfarmer.com': 0.9172547499030632,
'news.ucsc.edu': 0.9172085056879004,
'blogs.hbr.org': 0.9171843582580473,
'opinionator.blogs.nytimes.com': 0.9171826095986618,
'www.rtl-sdr.com': 0.9171698455400066,
'byuu.org': 0.9171665984622892,
'bugzilla.gnome.org': 0.9171648163962425,
'www.sfwa.org': 0.9171648163962425,
'www.spinellis.gr': 0.9171319512518645,
'0pointer.net': 0.9171049662814222,
'eff.org': 0.917104278613718,
'learnxinyminutes.com': 0.9170978082207938,
'9to5google.com': 0.917097241243007,
'www.cs.columbia.edu': 0.9170952792848854,
'www.fourmilab.ch': 0.9170935424089711,
'hynek.me': 0.9170881237147662,
'www.sarahmei.com': 0.9170796812591611,
'blog.piekniewski.info': 0.9170369069517826,
'cb.vu': 0.9169587328599824,
'math.stackexchange.com': 0.916937205478264,
'news.bloomberglaw.com': 0.9169353599665077,
'blog.darknedgy.net': 0.9169312054506662,
'telegra.ph': 0.9168505250691076,
'sniperinmahwah.wordpress.com': 0.9168254460707557,
'lists.swift.org': 0.9167939569922435,
'adit.io': 0.9167855816882253,
'steamcommunity.com': 0.9167584498934136,
'www.buzzfeed.com': 0.9167426632611669,
'www.privacyinternational.org': 0.9167293554781794,
'www.thurrott.com': 0.9167249467906359,
'www.cbsnews.com': 0.9167194231229859,
'socket.io': 0.9166902296520829,
'maryrosecook.com': 0.9166809726360262,
'tomayko.com': 0.9166652528198918,
'if50.substack.com': 0.9166328442343356,
'phenomena.nationalgeographic.com': 0.9166322627474873,
'www.netmeister.org': 0.9166258322005095,
'transition.fcc.gov': 0.9165981621015247,
'playground.tensorflow.org': 0.9165966237016168,
'www.macworld.com': 0.916575278474562,
'www.mahdiyusuf.com': 0.9165626020992179,
'nplusonemag.com': 0.9165533544728419,
'astralcodexten.substack.com': 0.9165406144271586,
'rjlipton.wordpress.com': 0.9164585931916936,
'kunle.app': 0.9164517870748075,
'cheatsheetseries.owasp.org': 0.9164513350559862,
'blogs.janestreet.com': 0.916405195738336,
'www.chromium.org': 0.9163955502793474,
'aht.seriouseats.com': 0.9163820584014595,
'arcadeblogger.com': 0.9162907826609931,
'lock.cmpxchg8b.com': 0.9162763903656581,
'codewithoutrules.com': 0.9162515056697728,
'blog.roboflow.ai': 0.9162348877374784,
'jolla.com': 0.9162125492359456,
'philipwalton.github.io': 0.9162125492359456,
'www.textfiles.com': 0.9161483132429212,
'landshark.io': 0.9160843328162562,
'www.syfy.com': 0.9160823277886534,
'www.a1k0n.net': 0.9160533176276322,
'thirtybees.com': 0.916017316017316,
'wp.josh.com': 0.9159844747688354,
'community.rapid7.com': 0.9159729296552184,
'variety.com': 0.9159645406845424,
'www.petekeen.net': 0.9159458498387062,
'pragprog.com': 0.9158994769642369,
'addons.mozilla.org': 0.9158845258968069,
'blog.abevoelker.com': 0.9158703069575684,
'uxdesign.cc': 0.915853392301296,
'tomblomfield.com': 0.91585125248835,
'www.platformer.news': 0.9158262569189228,
'airbnb.design': 0.9158145365560538,
'www.php.net': 0.9157901084713228,
'blakemasters.tumblr.com': 0.915774364872077,
'forum.bitcoin.org': 0.9157050312564461,
'victorzhou.com': 0.9156996812390578,
'www.newcriterion.com': 0.9156866313823816,
'blog.sagemath.com': 0.9156636337212226,
'security-center.intel.com': 0.9156401908652926,
'secret.club': 0.9156139932226358,
'codebutler.com': 0.9156072900794726,
'gfx-rs.github.io': 0.9155911821045295,
'open.blogs.nytimes.com': 0.9155276050596644,
'lambda-the-ultimate.org': 0.9155112419708785,
'www.universetoday.com': 0.9155071362428421,
'vc.gg': 0.9155052264808362,
'www.bostonglobe.com': 0.9154711894014951,
'slides.com': 0.915438865746552,
'openid.net': 0.9154218308800682,
'sizovs.net': 0.9154120001317235,
'minimalmac.com': 0.9154065991271528,
'pandodaily.com': 0.9153369387847924,
'www.ft.com': 0.9153363776443395,
'money.cnn.com': 0.9153197950917878,
'blogs.technet.com': 0.9152916278406317,
'tynan.com': 0.9152784227275973,
'antipolygraph.org': 0.9152601338015678,
'www.bing.com': 0.9152555063347714,
'dev.lemmy.ml': 0.9152501219202281,
'ai.facebook.com': 0.9151989015144473,
'www.spectator.co.uk': 0.9151910833402672,
'bollu.github.io': 0.915178606993226,
'cognitivemedium.com': 0.9151678416414023,
'www.nih.gov': 0.9151430151901143,
'www.cyanogenmod.org': 0.9151094600863039,
'codepen.io': 0.9150991029481277,
'www.media.mit.edu': 0.9150898590392416,
'ianthehenry.com': 0.9150718483883374,
'sloanreview.mit.edu': 0.9150432572484367,
'speakerdeck.com': 0.9150398484887767,
'en.lichess.org': 0.9150022514278866,
'www.datagenetics.com': 0.9150012436762652,
'people.kernel.org': 0.9149910445768656,
'ma.tt': 0.9149514110201404,
'people.freebsd.org': 0.9149417112190068,
'measuringmeasures.com': 0.9149271022386467,
'www.catb.org': 0.9149270125747468,
'defn.io': 0.9148816198946629,
'www.fastly.com': 0.914873343624634,
'www.eetimes.com': 0.9148540001117531,
'www.newrepublic.com': 0.9146889391211901,
'www.chicagomaroon.com': 0.9146877748460862,
'blog.openbloc.fr': 0.9146877748460862,
'blog.moertel.com': 0.9146704350942687,
'nathanbarry.com': 0.9146551473361769,
'bombich.com': 0.914654069788301,
'mojang.com': 0.9146416740750689,
'www.pgbovine.net': 0.914639262393428,
'www.bicycling.com': 0.9146271639157105,
'www.zemanta.com': 0.9146271639157105,
'www.perell.com': 0.914609307465025,
'mjtsai.com': 0.9145962250738603,
'www.softwarebyrob.com': 0.9145691937602046,
'fusion.net': 0.914564049263131,
'blogs.forbes.com': 0.914542164424382,
'beta.observablehq.com': 0.9145307006213452,
'www.zedshaw.com': 0.9145061280161968,
'apps.npr.org': 0.9144928315823566,
'lists.gnupg.org': 0.9144675565289525,
'gleam.run': 0.9144324984811291,
'www.thestar.com': 0.9143957132161558,
'www.militarytimes.com': 0.9143764543930346,
'blog.memsql.com': 0.9143635601136089,
'cr.yp.to': 0.9143445920251045,
'thepointsguy.com': 0.9143303279494647,
'dinosaurpictures.org': 0.9143080414361231,
'quillette.com': 0.9142948614738509,
'ticki.github.io': 0.914276003611421,
'neil.fraser.name': 0.9142481924113248,
'www3.nhk.or.jp': 0.9142479994563674,
'www.muppetlabs.com': 0.9142412765407236,
'blogs.gnome.org': 0.9142360167665817,
'readonlymemory.vg': 0.9142351900972591,
'news.osu.edu': 0.9142251289181204,
'www.cgl.uwaterloo.ca': 0.9142062865582451,
'www.fossil-scm.org': 0.9142014076198545,
'hub.jhu.edu': 0.9141963295065768,
'raw.github.com': 0.9141853922096216,
'www.sebastianmarshall.com': 0.9141316785617728,
'physicstoday.scitation.org': 0.9141308856828069,
'www.edge.org': 0.9140771607780698,
'blog.bugreplay.com': 0.914047971434993,
'www.medrxiv.org': 0.9139991128038881,
'permalink.gmane.org': 0.9139842285208853,
'www.notechmagazine.com': 0.9139648152317646,
'blog.ipfs.io': 0.9139402591654684,
'ninjasandrobots.com': 0.9139402119313542,
'blog.johnnovak.net': 0.9139144009903326,
'tldr.sh': 0.9138931911912991,
'internals.rust-lang.org': 0.9138888662706784,
'formidable.com': 0.9138769104260843,
'thenounproject.com': 0.9138746646560476,
'blog.earbits.com': 0.9138048995152518,
'gamehistory.org': 0.9137927637033101,
'www.influxdata.com': 0.913761562076886,
'm.washingtonpost.com': 0.9137051021098685,
'pythonspeed.com': 0.9137019640744012,
'observer.com': 0.9136945429590685,
'mastodon.social': 0.9136829141356647,
'whatever.scalzi.com': 0.9136722362354969,
'rietta.com': 0.9136487147921646,
'cushionapp.com': 0.9136267087023814,
'www.cantorsparadise.com': 0.9135854914860095,
'www.intelligence.senate.gov': 0.9135819523129652,
'blog.nullspace.io': 0.9134635980748237,
'keleshev.com': 0.9134557014620313,
'www.freethink.com': 0.913438665510595,
'singularityhub.com': 0.9134225553478318,
'www.greenteapress.com': 0.9134066096671126,
'hazlitt.net': 0.9134026214327062,
'www.zainrizvi.io': 0.913399178678123,
'www.welcometothejungle.co': 0.9133545417162321,
'joeduffyblog.com': 0.9133528454968176,
'tbbuck.com': 0.913350699282715,
'blog.fawny.org': 0.9133445360688772,
'cyclingtips.com': 0.9133347976741121,
'www.f-secure.com': 0.913329909915946,
'www.tweag.io': 0.9133006601895646,
'edri.org': 0.9132989780194611,
'brooker.co.za': 0.913269877818758,
'learnyouahaskell.com': 0.9132658585002156,
'dmitrysoshnikov.com': 0.9132544931220842,
'frankgroeneveld.nl': 0.9132379248658319,
'inconvergent.net': 0.9132264356820329,
'kennethreitz.org': 0.913224092229413,
'editorconfig.org': 0.9131943780375904,
'atom.io': 0.9131691418622392,
'mattgemmell.com': 0.9131626341114434,
'www.fastcoexist.com': 0.9131427628529503,
'atap.google.com': 0.9130249616995455,
'esbuild.github.io': 0.9130044843049328,
'blog.koehntopp.info': 0.9129670374057173,
'europa.eu': 0.9129663426547358,
'htmx.org': 0.912955421768738,
'www.markozivanovic.com': 0.912917338189758,
'number-none.com': 0.9129109262881329,
'sequoia-pgp.org': 0.9128836629372516,
'tanelpoder.com': 0.9128836629372516,
'www.nvidia.com': 0.9128613184383916,
'www.prospectmagazine.co.uk': 0.9128286921650719,
'bouk.co': 0.9128166963356691,
'www.espn.com': 0.9127399070748246,
'www.courtlistener.com': 0.912716064588026,
'lav.io': 0.9126813150141094,
'landave.io': 0.9126647018297601,
'blog.phusion.nl': 0.912650575320633,
'thezvi.wordpress.com': 0.9126499253271542,
'nickjanetakis.com': 0.9126192502614521,
'stateofjs.com': 0.9126077344903017,
'hakim.se': 0.9126077344903017,
'blog.datomic.com': 0.9125894031840723,
'docs.racket-lang.org': 0.912488235007798,
'jamie-wong.com': 0.9124775693621691,
'mocko.org.uk': 0.9124699553919461,
'www.telegraph.co.uk': 0.9124491411165716,
'ag.ny.gov': 0.9124354337525871,
'billwadge.wordpress.com': 0.9124293382023613,
'www.matasano.com': 0.9124000351892007,
'okigiveup.net': 0.9123862469871238,
'logicmag.io': 0.9123859846898301,
'andrewchen.co': 0.9123448304091328,
'www.airpair.com': 0.9123437098236648,
'cowlark.com': 0.9123304746459454,
'www.pbs.org': 0.9123204159144807,
'semianalysis.com': 0.912313946546377,
'kk.org': 0.9123097221722443,
'log.nadim.cc': 0.9122965641952984,
'www.collaborativefund.com': 0.9122890867111917,
'www.ventusky.com': 0.912268380036006,
'www.nintendo.com': 0.9122375993317341,
'www.worldofindie.co.uk': 0.9122336940304385,
'jeroenjanssens.com': 0.9122336940304385,
'libreboot.org': 0.9122148556890368,
'blog.roboflow.com': 0.9121810251169891,
'www.msn.com': 0.9121613167344725,
'debugger.medium.com': 0.9121472531001502,
'www.pagetable.com': 0.9121460025296508,
'www.eejournal.com': 0.9121137219392826,
'blog.sanctum.geek.nz': 0.91211049118521,
'blog.elcomsoft.com': 0.9120714505588577,
'www.judegomila.com': 0.9120548967590474,
'cirw.in': 0.9119924579917503,
'corner.squareup.com': 0.9119829497403208,
'flightfox.com': 0.911934167943058,
'tip.golang.org': 0.9119192289295014,
'www.nzherald.co.nz': 0.9119181580628049,
'www.wezm.net': 0.911916042600115,
'crontab.guru': 0.9119103202923069,
'degoes.net': 0.9119022003821793,
'investors.modernatx.com': 0.9118967098234642,
'linustechtips.com': 0.9118886368935806,
'blog.tanyakhovanova.com': 0.911882662223509,
'amandapeyton.com': 0.9118468718524512,
'nypost.com': 0.9118217286764665,
'forumserver.twoplustwo.com': 0.9118191670930907,
'belkadan.com': 0.9117907816852344,
'nicolasgallagher.com': 0.9117507194110998,
'leowid.com': 0.9117097398796832,
'blog.izs.me': 0.9117080886266495,
'www.electrospaces.net': 0.9117024618532766,
'granta.com': 0.9116811248127611,
'blog.darksky.net': 0.911673049778484,
'lettersofnote.com': 0.9116636714221488,
'www.webkit.org': 0.9115845605402175,
'shipilev.net': 0.9115598314538631,
'www.nanex.net': 0.911539637720508,
'blog.whatsapp.com': 0.9115335907580295,
'leetcode.com': 0.9115051248737622,
'www.numfocus.org': 0.9115022081675324,
'onlinelibrary.wiley.com': 0.9114852476099333,
'www.tomscott.com': 0.9114829485455537,
'trmm.net': 0.9114358511601164,
'www.marketplace.org': 0.9113899909762714,
'www.codecademy.com': 0.911385022367728,
'news.vice.com': 0.9113079103997874,
'bit-player.org': 0.9112589244732427,
'postmake.io': 0.9112553086495423,
'journal.standardnotes.org': 0.9112172605590297,
'semantic-ui.com': 0.9111938765485481,
'blog.dshr.org': 0.9111782499363091,
'www.rogerebert.com': 0.9111691070786543,
'web.dev': 0.9111397506487405,
'bravenewgeek.com': 0.9111366238548788,
'benjiweber.co.uk': 0.9111133306489915,
'www.theinformation.com': 0.9111027630703883,
'jmmv.dev': 0.9110762745053466,
'www.nycresistor.com': 0.9110654315146017,
'bloomberg.github.io': 0.9110576284941153,
'glench.com': 0.9110421711549385,
'blog.kicksend.com': 0.9109887018713544,
'www.godaddy.com': 0.9109502513393147,
'neverworkintheory.org': 0.910947601730565,
'thebillfold.com': 0.9109433900278253,
'www.superhighway98.com': 0.9109274563820018,
'tylerneylon.com': 0.9109274563820018,
'gankro.github.io': 0.9109257416809919,
'thetruesize.com': 0.9108778939911146,
'zipcpu.com': 0.9108728406864637,
'www.cs.uni.edu': 0.910867930832846,
'www.whitehouse.gov': 0.9108673306488233,
'bradleytaunt.com': 0.9108617254932104,
'scripting.com': 0.9108577372151384,
'www.cybertec-postgresql.com': 0.9108466966632626,
'www.wcvb.com': 0.9107985466943099,
'www.dslreports.com': 0.9107964183891935,
'justi.cz': 0.9107667989469016,
'finance.yahoo.com': 0.9107655094145327,
'blog.data.gov.sg': 0.910764849456426,
'www.gutenberg.org': 0.9107167501810286,
'www.sciencedaily.com': 0.9106989174929755,
'blog.airbnb.com': 0.9106859675193685,
'www.sendwithus.com': 0.9106524760338102,
'chrismm.com': 0.9106400698754001,
'stevecorona.com': 0.9106182479757787,
'swanson.github.com': 0.910602900714108,
'eager.io': 0.9105915511196,
'www.insidesources.com': 0.9105882590524582,
'chrisseaton.com': 0.9105861690903986,
'cstheory.stackexchange.com': 0.9105756115852348,
'www.behind-the-enemy-lines.com': 0.9105542219435652,
'www.commondreams.org': 0.9105378564647107,
'tech.marksblogg.com': 0.91050910005871,
'blog.bitquabit.com': 0.9104538788044407,
'utkusen.com': 0.9104251525129977,
'lab.hakim.se': 0.9104228721342874,
'calacanis.com': 0.9104018184523797,
'blog.metaobject.com': 0.9103873178112856,
'blogs.discovermagazine.com': 0.9103541727829514,
'flutter.io': 0.9103411596660869,
'www.chriskrycho.com': 0.9103411596660869,
'www.washingtonmonthly.com': 0.9103284229318426,
'nora.codes': 0.9103242591900224,
'blog.professorbeekums.com': 0.9103190381466665,
'yro.slashdot.org': 0.9103086060368418,
'noncombatant.org': 0.9102799785330806,
'www.irrlicht3d.org': 0.9102781052287677,
'www.analog.com': 0.9102635806004643,
'bret.appspot.com': 0.9102570367702293,
'byuu.net': 0.9102490436100149,
'learntemail.sam.today': 0.9102460058415499,
'discuss.joelonsoftware.com': 0.9102271772174491,
'news.samsung.com': 0.9102218695714285,
'hamberg.no': 0.9101858671649673,
'culture.pl': 0.9101776992266171,
'godbolt.org': 0.9101759214909476,
'www.usgamer.net': 0.9101501554941474,
'openmymind.net': 0.9101212855439798,
'www.pgcon.org': 0.9101151487041538,
'blog.insightdatascience.com': 0.9101089594835049,
'blog.mapbox.com': 0.9100815042120244,
'heap.io': 0.9100756276392993,
'www.openttd.org': 0.910035809777446,
'www.gazehawk.com': 0.9100190649304827,
'blog.headius.com': 0.9100084850499096,
'jquerymobile.com': 0.9099865878820667,
'strlen.com': 0.9099808992547239,
'bellmar.medium.com': 0.9099735750567878,
'designfestival.com': 0.9099647154391645,
'blog.kowalczyk.info': 0.909946343995861,
'natewienert.com': 0.9099163091164866,
'blog.iron.io': 0.9099140278683325,
'www.optaplanner.org': 0.9098736627179512,
'blog.directededge.com': 0.9098673466271318,
'www.acm.org': 0.9098204575391435,
'postgres.heroku.com': 0.9098008938162723,
'netflix.github.io': 0.9097922949542527,
'www.upguard.com': 0.9097914092921688,
'dopeboy.github.io': 0.9097690974580366,
'www.panix.com': 0.909767441860465,
'www.arangodb.com': 0.9097488818296668,
'www.ctvnews.ca': 0.9097251357478893,
'superorganizers.substack.com': 0.9097041628130547,
'tips.ariyh.com': 0.9096731009765122,
'henrikwarne.com': 0.9096514841360387,
'web.media.mit.edu': 0.909634017448133,
'www.math.columbia.edu': 0.9096124004726523,
'sorbet.org': 0.9095992544268405,
'www.whattofix.com': 0.9095812789302404,
'lukeplant.me.uk': 0.90953747228724,
'blog.quora.com': 0.9095278802240517,
'areallybadidea.com': 0.909495363691211,
'www.gsb.stanford.edu': 0.9094535928364683,
'rednuht.org': 0.9094069027053294,
'cappuccino.org': 0.9093755871303836,
'blog.wallaroolabs.com': 0.9093612256561168,
'resobscura.blogspot.com': 0.9093403026878701,
'story.californiasunday.com': 0.9093105225503313,
'xato.net': 0.9092732945858745,
'thinkfaster.co': 0.9092409956171542,
'blog.algorithmia.com': 0.9091584066166702,
'n-o-d-e.net': 0.9091578728445029,
'www.talkingquickly.co.uk': 0.9091487127436086,
'blamcast.net': 0.9091487127436086,
'www.opendemocracy.net': 0.9091030079275115,
'www.pixelbeat.org': 0.9091017229185366,
'blog.2ndquadrant.com': 0.9090827165871289,
'klinger.io': 0.9090715471566316,
'invisibleup.com': 0.9090623220718969,
'www.mattgreer.org': 0.909028070014661,
'www.georgehart.com': 0.9090164458580612,
'gsuiteupdates.googleblog.com': 0.9090066640642879,
'programming.guide': 0.9090056285178236,
'www.revealnews.org': 0.9089822305000187,
'www.android.com': 0.9089431208516167,
'renderman.pixar.com': 0.9089238890253597,
'swtch.com': 0.9088645421539494,
'akkartik.name': 0.9088447989502915,
'ma.ttias.be': 0.9088072198608288,
'blog.freecodecamp.com': 0.9087962710873405,
'cognitiontoday.com': 0.9087409016865039,
'data.triplebyte.com': 0.9087226720990824,
'unix4lyfe.org': 0.9087151101306639,
'coconut-lang.org': 0.9087151101306639,
'discourse.nixos.org': 0.908662900188324,
'macwright.org': 0.908628352520185,
'www.varnish-cache.org': 0.9086150127257842,
'blog.lyft.com': 0.9085807144025437,
'martinrue.com': 0.9085670862102617,
'bittersoutherner.com': 0.9085029987755043,
'www.somebits.com': 0.9084860999243902,
'certsimple.com': 0.9084684115974558,
'www.calvertjournal.com': 0.9084635529608529,
'www.pentestpartners.com': 0.9084628705092087,
'wiki.gentoo.org': 0.9084338073440062,
'00f.net': 0.9083970094532444,
'www.weforum.org': 0.9083913811504614,
'evanmiller.org': 0.9083652366260793,
'www.geekwire.com': 0.9083231743190484,
'news.yahoo.com': 0.9083103718085543,
'books.google.com': 0.9082816718436784,
'georgestocker.com': 0.9082719861623312,
'www.yacoset.com': 0.9082542163067842,
'spencerfry.com': 0.9082262546419613,
'citusdata.com': 0.9082125231004685,
'www.thenewatlantis.com': 0.9081731363479266,
'j2kun.svbtle.com': 0.9081723260580993,
'www.bothsidesofthetable.com': 0.9081408393460154,
'www.thetimes.co.uk': 0.9081193478982206,
'corecursive.com': 0.90809594719532,
'www.wildml.com': 0.9080846001869393,
'smallcultfollowing.com': 0.9080827648147168,
'www.theartnewspaper.com': 0.9080723328672826,
'cdixon.posterous.com': 0.9080713598958169,
'www.readwriteweb.com': 0.9080580806277859,
'www.hsph.harvard.edu': 0.9080327405123425,
'www.state.gov': 0.908024664450505,
'planscope.io': 0.907995411751062,
'www.allthingsdistributed.com': 0.9079938413302127,
'www.boston.com': 0.9079342359331115,
'blog.kentcdodds.com': 0.9079295161962991,
'www.brianstorti.com': 0.9079295161962991,
'scala-lang.org': 0.9079245106804908,
'austinvernon.eth.link': 0.9078822412155745,
'pindancing.blogspot.com': 0.9078822412155745,
'www.residentadvisor.net': 0.9078629004665392,
'www.documentcloud.org': 0.9078579855885078,
'googledocs.blogspot.com': 0.9078489828529417,
'www.cdc.gov': 0.9078393365891005,
'www.howardism.org': 0.9078286053408602,
'fsharpforfunandprofit.com': 0.9078054396581823,
'semiaccurate.com': 0.907755821837259,
'www.hetzner.com': 0.9077473776003605,
'aiweirdness.com': 0.9077065010884221,
'www.bit-101.com': 0.9077018983499333,
'www.41latitude.com': 0.9077018983499333,
'www.amnh.org': 0.9076987359242361,
'infolab.stanford.edu': 0.9076647094055436,
'stories.californiasunday.com': 0.9076472906657365,
'pxlnv.com': 0.9076015494709667,
'batsov.com': 0.9075988217886798,
'www.gamedev.net': 0.9075987374715156,
'diff.wikimedia.org': 0.9075899793873731,
'rust-analyzer.github.io': 0.9075741551452703,
'www.mondaynote.com': 0.9075302424153898,
'www.gatesfoundation.org': 0.9075167456992206,
'meta.ath0.com': 0.9075048785901592,
'radicle.xyz': 0.9074969769110395,
'www.gigamonkeys.com': 0.9074847578759488,
'thetyee.ca': 0.9074067773987987,
'www.ucdavis.edu': 0.9073906291556996,
'annehelen.substack.com': 0.9073906291556995,
'www.undeadly.org': 0.9073802859341632,
'teenage.engineering': 0.9073802859341632,
'approximatelycorrect.com': 0.9073547174968317,
'console.dev': 0.9073543457497613,
'www.qemu.org': 0.9073039659002148,
'chadaustin.me': 0.907285573667598,
'javascript.info': 0.9072810118016286,
'crawshaw.io': 0.9072713221321198,
'anandtech.com': 0.9072582150847296,
'minireference.com': 0.9072555275680507,
'0x00sec.org': 0.9072265870419588,
'jtaby.com': 0.9072263761646095,
'hackernoon.com': 0.9072112942336243,
'aem1k.com': 0.9071770334928229,
'www.theatlanticwire.com': 0.9071768956517269,
'voice.mozilla.org': 0.9071764184712818,
'www.tennessean.com': 0.9071055766821053,
'graphicallinearalgebra.net': 0.9070506542876967,
'jpmens.net': 0.9070445763259152,
'www.eugenewei.com': 0.907025376171475,
'blog.packagecloud.io': 0.9070193601765509,
'zef.me': 0.9070189250200639,
'blog.makezine.com': 0.907016130343466,
'yannesposito.com': 0.9069508537217187,
'ubuntu.com': 0.9069426319310011,
'apolloinrealtime.org': 0.9069028892973577,
'leontrolski.github.io': 0.9069028892973576,
'divinations.substack.com': 0.9068786540871128,
'nicole.express': 0.9068478175005559,
'www.cnet.com': 0.9068461959910697,
'docs.browserless.io': 0.9068163026535472,
'joshondesign.com': 0.906770998121308,
'humanparts.medium.com': 0.906760680978119,
'aviationweek.com': 0.9067465504035637,
'philip.greenspun.com': 0.9067439389898158,
'huonw.github.io': 0.9066784039500629,
'ariadne.space': 0.90667782006653,
'blog.sia.tech': 0.9066295330975407,
'www.cato.org': 0.9065844637081601,
'blog.hipchat.com': 0.9065819832684133,
'ruben.verborgh.org': 0.9065777308846613,
'www.straightdope.com': 0.9064957391706991,
'www.joshbarczak.com': 0.9064865112228077,
'www.aarongreenspan.com': 0.9064448754833254,
'james-iry.blogspot.com': 0.9064382639201024,
'changelog.com': 0.9064361162234437,
'jdlm.info': 0.9064121095346035,
'www.csail.mit.edu': 0.9064063607235981,
'blog.cleancoder.com': 0.9063898821262703,
'kyrobeshay.com': 0.9063750617475119,
'js1k.com': 0.9063646805868381,
'socketsite.com': 0.9063204907590037,
'mobile.reuters.com': 0.9063074578535042,
'www.spinics.net': 0.9062822710883404,
'www.h-online.com': 0.9062819868634641,
'humanwhocodes.com': 0.9062381797427127,
'angular.io': 0.9062269960472951,
'www.digitalcameraworld.com': 0.906205656831627,
'www.amd.com': 0.9061918994722282,
'noahpinion.substack.com': 0.9061897254478186,
'scienceblogs.com': 0.9061724154956764,
'forge.medium.com': 0.9061379433537395,
'folklore.org': 0.9061180195674341,
'www.jonathan-cook.net': 0.906117767771791,
'randomwire.com': 0.9060987415295255,
'blog.crunchydata.com': 0.9060970567080769,
'www.timeshighereducation.co.uk': 0.9060939645760167,
'gds.blog.gov.uk': 0.9060670828649058,
'www.mihaileric.com': 0.9060240156895187,
'seatgeek.com': 0.9059982211064962,
'www.fanfiction.net': 0.9059866760601158,
'msrc-blog.microsoft.com': 0.9059631330906113,
'thecounter.org': 0.9059275361638426,
'www.haproxy.com': 0.905913947863991,
'movingfulcrum.com': 0.9058292135077338,
'onstartups.com': 0.9057830668516696,
'noyb.eu': 0.9057682181268583,
'millionshort.com': 0.9057645582241304,
'blog.sketchapp.com': 0.905760488137004,
'www.tillett.info': 0.9057526849990746,
'community.ui.com': 0.9057461537102033,
'adnauseam.io': 0.9057337220602527,
'intoli.com': 0.9056718941128913,
'www.jsoftware.com': 0.9056617340683936,
'www.oreilly.com': 0.9056506837308099,
'www.themacro.com': 0.905632500892492,
'www.chicagotribune.com': 0.9055936474876807,
'developer.squareup.com': 0.9055714474003402,
'pixelambacht.nl': 0.9055660472352755,
'lambdaschool.com': 0.9055457567735196,
'consumerist.com': 0.9055167142071894,
'lisp-univ-etc.blogspot.com': 0.905513555501623,
'www.reaktor.com': 0.9055111900863794,
'mars.jpl.nasa.gov': 0.9054761782517211,
'blog.rescuetime.com': 0.9054099923733353,
'dustycloud.org': 0.9053737646709764,
'the.exa.website': 0.9053731049780873,
'www.lareviewofbooks.org': 0.905357669166342,
'adventures.michaelfbryan.com': 0.9053432603320545,
'bigtechnology.substack.com': 0.9053275639886895,
'mashable.com': 0.9052624670823227,
'blog.xamarin.com': 0.9052537100443511,
'www.micahlerner.com': 0.905230894226323,
'software.intel.com': 0.9052122500296891,
'canonical.org': 0.9051821187623978,
'www.scmp.com': 0.9051295599614926,
'grantland.com': 0.9050923790278581,
'metarabbit.wordpress.com': 0.9050923774541669,
'education.github.com': 0.9050362272771562,
'www.thelancet.com': 0.905021932673904,
'www.nextplatform.com': 0.9050024178186049,
'home-assistant.io': 0.904995020548573,
'varnish-cache.org': 0.9049950205485728,
'blog.netlab.360.com': 0.9049426199082281,
'm.nautil.us': 0.9049309527012956,
'www.wikileaks.org': 0.9049234922760648,
'www.cell.com': 0.9049048617612142,
'cheapskatesguide.org': 0.9048941298503079,
'engineering.tumblr.com': 0.9048881466624363,
'w3techs.com': 0.9048639649614062,
'literaryreview.co.uk': 0.9048564834641546,
'explorabl.es': 0.9048086359175662,
'blog.felipe.rs': 0.9048079619547811,
'bitsup.blogspot.com': 0.9047993533025517,
'unherd.com': 0.9047865592498257,
'garry.posterous.com': 0.9047577664612345,
'worthdoingbadly.com': 0.9047131145041029,
'rob.conery.io': 0.9046723461377625,
'io9.com': 0.9046401640309223,
'nolanlawson.com': 0.9046163101493089,
'www.soundguys.com': 0.9046093349153457,
'www.starterstory.com': 0.9045594173223862,
'blog.jse.li': 0.9045467732931323,
'knowingless.com': 0.9045071158772837,
'decrypt.co': 0.9044787320123616,
'www.euronews.com': 0.9044514611283094,
'www.mattwarren.org': 0.9044367882909422,
'www.nirsoft.net': 0.9044100097952366,
'www.atlassian.com': 0.9043691389685379,
'exploringjs.com': 0.9043580712164743,
'www.recurse.com': 0.9043507328694445,
'www.legacy.com': 0.9043392504930966,
'docs.fcc.gov': 0.9043351544730283,
'channel9.msdn.com': 0.904284603417082,
'blog.ksplice.com': 0.9042701981747182,
'www.valvesoftware.com': 0.9042649125194085,
'blog.netbsd.org': 0.904253667752545,
'hyper.sh': 0.9042360279323705,
'blog.waleedkhan.name': 0.9042360279323705,
'computerhistory.org': 0.904233366454177,
'sugru.com': 0.9041590518924904,
'feefighters.com': 0.9041509269810387,
'dejanseo.com.au': 0.9040498577824194,
'www.ianww.com': 0.9040340548815767,
'blog.cr.yp.to': 0.9040072201081354,
'trac.nginx.org': 0.9039603960396039,
'kentnguyen.com': 0.903957880171432,
'jonathanturley.org': 0.9039575565206787,
'snyk.io': 0.9039222826692356,
'radar.oreilly.com': 0.9038574383300398,
'blog.sentry.io': 0.9037725052372675,
'pluto.jhuapl.edu': 0.9037348670913641,
'calendar.perfplanet.com': 0.903704307455864,
'graphics.stanford.edu': 0.903685214892998,
'singhkays.com': 0.9036579185992312,
'tinyapps.org': 0.9036401962155587,
'motherjones.com': 0.9036341732104931,
'blog.evjang.com': 0.903604799270119,
'www.lightbluetouchpaper.org': 0.9035449087212127,
'blog.arduino.cc': 0.9035388791817426,
'www.psychnewsdaily.com': 0.9035015745269135,
'gnu.org': 0.9035002162125149,
'pchiusano.github.io': 0.9034726569621144,
'hypercritical.co': 0.9034345666188365,
'0a.io': 0.9034135558240236,
'www.erowid.org': 0.9034051526175023,
'www.inklestudios.com': 0.9033800880640788,
'blog.soylent.com': 0.9033111216571915,
'www.fpcomplete.com': 0.9032736546035143,
'dmitryfrank.com': 0.9032706083767721,
'www.red-lang.org': 0.9032700232946759,
'www.parallella.org': 0.9032534786832978,
'habr.com': 0.903232673933909,
'www.musictech.net': 0.9031936127744511,
'josephwalla.com': 0.9031899305005244,
'hsivonen.fi': 0.9031735375682458,
'thoughtcatalog.com': 0.9031627630658119,
'www.scylladb.com': 0.9031607915889202,
'www.dreamhost.com': 0.9031378273716731,
'www.stochasticlifestyle.com': 0.9031310549200046,
'vaibhavsagar.com': 0.9031083985391525,
'searchengineland.com': 0.9030654738013855,
'www.nrk.no': 0.9030651285810807,
'blog.app.net': 0.903029680517345,
'mail.openjdk.java.net': 0.9030249968896539,
'sciencebulletin.org': 0.9029854747384259,
'www.paperplanes.de': 0.9029834819452879,
'research.facebook.com': 0.9029781526945183,
'jessenoller.com': 0.9029740692225192,
'blog.hirelite.com': 0.9029584423609692,
'www.bemmu.com': 0.9029304388261952,
'www.cam.ac.uk': 0.9029214208630774,
'adguard.com': 0.9028940843182978,
'www.protocol.com': 0.9028937484894913,
'redislabs.com': 0.9028875063757023,
'www.coding2learn.org': 0.9028312810037816,
'www.tonywright.com': 0.9027993696652322,
'writing.kemitchell.com': 0.902794255100101,
'support.microsoft.com': 0.9027883582257822,
'voxeu.org': 0.9027732002536809,
'math.ucr.edu': 0.9027587862028252,
'gekk.info': 0.9026852677733105,
'www.dwitter.net': 0.9026795513672158,
'nakedsecurity.sophos.com': 0.9026433950295091,
'www.tomdispatch.com': 0.9026213839350323,
'www.erichgrunewald.com': 0.9025809741606529,
'everydayastronaut.com': 0.9025525737662192,
'meta.stackexchange.com': 0.9025397509956542,
'dashbit.co': 0.9025100393249755,
'kroah.com': 0.9025051904560609,
'www.terrybisson.com': 0.902478233349385,
'www.whatsapp.com': 0.9024176929829565,
'ploum.net': 0.9024105508563706,
'explosion.ai': 0.9024029708934638,
'blog.repl.it': 0.9024021393458092,
'neurosciencenews.com': 0.9023952036309971,
'www.supermemo.com': 0.9023729956597971,
'www.diyphotography.net': 0.902367985362208,
'kcsrk.info': 0.902365879005221,
'security.stackexchange.com': 0.9023603426250738,
'developer.android.com': 0.9023376538238216,
'blog.rjmetrics.com': 0.9023127105964952,
'kakoune.org': 0.902272263107977,
'www.saurik.com': 0.9022715997521633,
'thetoolsweneed.com': 0.902259888208641,
'atthis.link': 0.9022431859413726,
'hg.python.org': 0.9021998686737284,
'www.softwarepreservation.org': 0.9021625352128104,
'www.kleinbottle.com': 0.9021190716448031,
'interrupt.memfault.com': 0.9020903788101069,
'dspace.mit.edu': 0.9020251184851814,
'philipwalton.com': 0.9020062837472365,
'www.foxbusiness.com': 0.9019313178215171,
'jrsinclair.com': 0.9019295383152887,
'www.space.com': 0.9019268835645871,
'sachin.posterous.com': 0.9018811076664313,
'khn.org': 0.901879302131479,
'www.pigdog.org': 0.901878363289037,
'zackkanter.com': 0.9018673516010077,
'www.informationdiet.com': 0.9018348908028704,
'giantrobotlasers.com': 0.9017894645998003,
'incolumitas.com': 0.9017770940926804,
'jeffkreeftmeijer.com': 0.9017611553399866,
'bostonreview.net': 0.9017262602947111,
'techreport.com': 0.901717492257879,
'www.baldurbjarnason.com': 0.9017151044394945,
'www.sicpers.info': 0.9017134966165734,
'blog.sfgate.com': 0.9016737116837992,
'write.as': 0.9016704517485878,
'www.aljazeera.com': 0.9016695698160121,
'fr.slideshare.net': 0.9016442121790157,
'everything2.com': 0.9016361459958601,
'typicalprogrammer.com': 0.9015788203543227,
'blog.scaleway.com': 0.9015776579003968,
'rethrick.com': 0.9015770143955718,
'inference-review.com': 0.9015413236700288,
'thedorkweb.substack.com': 0.9015280289106077,
'blog.23andme.com': 0.9015240824972036,
'www.mikealche.com': 0.9015168230317929,
'hashcat.net': 0.9015094649406228,
'madebyloren.com': 0.9014851889844163,
'www.blockchain.com': 0.901483519454099,
'www.computerworld.com': 0.9014820166011037,
'cen.acs.org': 0.901480955026061,
'lauren.vortex.com': 0.9014724896133568,
'oreilly.com': 0.901407801663114,
'blog.netflix.com': 0.9013957945143364,
'commoncrawl.org': 0.9013596330544786,
'www.svds.com': 0.9013257789593407,
'blog.goodstuff.im': 0.9013257789593407,
'www.lhup.edu': 0.9013191983110588,
'www.nist.gov': 0.9013044488448885,
'jasonevanish.com': 0.9012724811591895,
'nextcity.org': 0.9012682451009019,
'nuitka.net': 0.9012059293191188,
'dataswamp.org': 0.9011871538800621,
'www.reactos.org': 0.9011785225363149,
'limpet.net': 0.9011626533660011,
'nextjournal.com': 0.9011584604272797,
'sinews.siam.org': 0.9011379162640358,
'blog.zactownsend.com': 0.9011275964672605,
'm.guardiannews.com': 0.9011260854499006,
'mazzo.li': 0.9011141262538125,
'news.developer.nvidia.com': 0.9011001587785914,
'spideroak.com': 0.901093458914619,
'www.startribune.com': 0.9010814985872827,
'www.gitbook.com': 0.901071077131811,
'www.brusselstimes.com': 0.9010360879334309,
'blog.instagram.com': 0.9010324035185349,
'benedikt-bitterli.me': 0.9009740364816841,
'fedoramagazine.org': 0.9009664628347622,
'www.rte.ie': 0.9009526229516198,
'f-droid.org': 0.9008856641605046,
'blog.dwolla.com': 0.9008628892479406,
'news.trust.org': 0.9008305066882848,
'about.bnef.com': 0.9007424598766725,
'txti.es': 0.9007231608685488,
'samizdat.mines.edu': 0.9007085795324656,
'www.murilopereira.com': 0.9007078611650351,
'bitcoincharts.com': 0.9006914299992029,
'material.io': 0.9006442988271292,
'openradar.appspot.com': 0.9006237134629408,
'blog.rapportive.com': 0.9006217110635638,
'm-cacm.acm.org': 0.9005996462860704,
'blog.halide.cam': 0.9005839293957032,
'www.billthelizard.com': 0.9005839293957032,
'news.sciencemag.org': 0.9005769469610003,
'www.edmunds.com': 0.900563544204038,
'wiki.c2.com': 0.9005588568752737,
'thinkvitamin.com': 0.9005542231541395,
'ipython.org': 0.900531415561299,
'panelbear.com': 0.9005250466106081,
'stuartsierra.com': 0.9004498846216558,
'clojurescript.org': 0.900443619125688,
'maps.org': 0.9004208529632491,
'www.dextronet.com': 0.9004014616042372,
'blog.digitalocean.com': 0.9003805295727227,
'www.happyassassin.net': 0.9003455623339318,
'blockchain.info': 0.9003439746043133,
'realm.io': 0.9003295505994889,
'unixsheikh.com': 0.9003091480910362,
'zyan.scripts.mit.edu': 0.9002640065392239,
'til.simonwillison.net': 0.9002517190240483,
'fingerprintjs.com': 0.9002278669918841,
'cs-blog.khanacademy.org': 0.9001663145026005,
'eugene.kaspersky.com': 0.9000762462446584,
'www.bay12games.com': 0.9000684308983997,
'ginzametrics.com': 0.9000536631561354,
'blog.simpleanalytics.com': 0.9000166910943967,
'viewfromthewing.com': 0.9000166910943967,
'liam-on-linux.livejournal.com': 0.9000090039034558,
'softwaremaniacs.org': 0.9000031026474421,
'blog.mariadb.org': 0.8999936863202533,
'levels.io': 0.899970180177946,
'craigmod.com': 0.8999690838572385,
'news.spotify.com': 0.8999500411279416,
'forums.developer.apple.com': 0.8999431500814804,
'www.overcomingbias.com': 0.8999429014404717,
'users.rust-lang.org': 0.8999355740659959,
'scrimba.com': 0.8999254626887211,
'www.sbnation.com': 0.8998688532688766,
'benswann.com': 0.89986648239432,
'www.thelocal.dk': 0.8998638303612118,
'pitchfork.com': 0.8998590033239674,
'wekeroad.com': 0.899851220900418,
'animalnewyork.com': 0.8998375739356855,
'beets.io': 0.899813377379441,
'www.ibiblio.org': 0.8997974933028597,
'kernelnewbies.org': 0.899793879009676,
'www.fullstackreact.com': 0.8997933884297521,
'ricostacruz.com': 0.8997816625141879,
'www.digitalocean.com': 0.8997784555641042,
'www.maps.org': 0.8997736250289394,
'pointersgonewild.com': 0.89975961274635,
'512pixels.net': 0.8997410869841632,
'blog.wilshipley.com': 0.899720920102866,
'devlog.hexops.com': 0.8997114101311218,
'blogs.nvidia.com': 0.899693028608083,
'www.sacbee.com': 0.8996905748131163,
'www.publicbooks.org': 0.8996781736043378,
'hforsten.com': 0.8996749694919105,
'okmij.org': 0.899647639382427,
'www.dataquest.io': 0.8996134651823179,
'www.nocsdegree.com': 0.8995491494660874,
'pandas.pydata.org': 0.8995416268122183,
'doctorow.medium.com': 0.8995206430959863,
'blog.ezyang.com': 0.8995116025900058,
'cs.gmu.edu': 0.8995004521044683,
'weblog.raganwald.com': 0.8994722109240123,
'finance.fortune.cnn.com': 0.8994585795276846,
'pedestrianobservations.com': 0.8994403946669056,
'dangrover.com': 0.8994295897503085,
'nostarch.com': 0.8994103848538745,
'foreignpolicy.com': 0.8994068719520655,
'benchmarksgame-team.pages.debian.net': 0.8993756538078823,
'kvark.github.io': 0.8993354696336303,
'www.scummvm.org': 0.8993014570629587,
'www.jpl.nasa.gov': 0.8992820116565705,
'theweek.com': 0.8992433138556716,
'status.heroku.com': 0.8992414299043588,
'tech.grammarly.com': 0.8992224733195932,
'www.mensjournal.com': 0.8992186596239703,
'braythwayt.com': 0.8992151653414675,
'clayallsopp.com': 0.8991984891927328,
'blog.japaric.io': 0.8991683991683992,
'www.linuxuprising.com': 0.8991634552022689,
'www.cambridge.org': 0.8991459176334309,
'pando.com': 0.8991426686559758,
'nextcloud.com': 0.8991240467738062,
'blog.luap.info': 0.8991106158419904,
'goto.ucsd.edu': 0.8990952211205248,
'layervault.tumblr.com': 0.8990927364776206,
'www.preetamnath.com': 0.8990436000092332,
'www.loper-os.org': 0.8990428163259621,
'beta.latimes.com': 0.8990201091709796,
'status.linode.com': 0.8990191396542296,
'www.yahoo.com': 0.8990120070182801,
'www.mining.com': 0.8989843218989275,
'thestartuptoolkit.com': 0.8989617943814647,
'exple.tive.org': 0.8989583647726666,
'www.percona.com': 0.8989398873032174,
'tug.org': 0.8988675834004226,
'azure.microsoft.com': 0.8988119185957827,
'badassjs.com': 0.8987895474430552,
'florentcrivello.com': 0.8987782517021887,
'joelonsoftware.com': 0.8987694677031264,
'www.cntraveler.com': 0.8987494824459783,
'postmasculine.com': 0.8986927094085853,
'blade.nagaokaut.ac.jp': 0.8986856729161738,
'blog.demofox.org': 0.8986624269542334,
'www.zeit.de': 0.8986585350988878,
'louwrentius.com': 0.8986513425823265,
'www.jeffwofford.com': 0.8986246802875617,
'lists.canonical.org': 0.8985997713186075,
'blog.aboutamazon.com': 0.898569865878112,
'discourse.haskell.org': 0.8985468763312833,
'marginalrevolution.com': 0.8985222766062069,
'www.southampton.ac.uk': 0.8985031338241491,
'wccftech.com': 0.898464128485399,
'pages.nist.gov': 0.8984293193717278,
'blog.jcoglan.com': 0.8984277156180708,
'today.duke.edu': 0.898426650553988,
'blog.us.playstation.com': 0.8984266505539878,
'technet.microsoft.com': 0.8984178818074816,
'moultano.wordpress.com': 0.8983087054446331,
'forums.swift.org': 0.8983030929609286,
'pyfound.blogspot.com': 0.8982450409736414,
'sebastianraschka.com': 0.8982066640921482,
'www.michaelfogleman.com': 0.8982013504259682,
'www.confluent.io': 0.8981951121390044,
'www.maa.org': 0.8981932917194397,
'serverfault.com': 0.898185336292522,
'fitzgeraldnick.com': 0.8981340410428447,
'www.2600.com': 0.8981193270212353,
'rustyshelf.org': 0.898115004185606,
'svnweb.freebsd.org': 0.8981026654387363,
'www.argmin.net': 0.8980921557949308,
'www.theawl.com': 0.8980833367511953,
'keygen.sh': 0.8980696445122388,
'www.freightwaves.com': 0.8980574109093967,
'developers.facebook.com': 0.8980338313592694,
'www.randomhacks.net': 0.8980285406407745,
'susam.in': 0.8980164962202184,
'exercism.io': 0.8980136526006044,
'trekhleb.dev': 0.8980055706761685,
'securelist.com': 0.8979897593608089,
'air.mozilla.org': 0.897889243419127,
'yahooeng.tumblr.com': 0.8978817471988295,
'techrights.org': 0.8978658400805657,
'franciskim.co': 0.8978653593888846,
'ergoemacs.org': 0.8978533245623609,
'www.scs.stanford.edu': 0.8978490165530042,
'www.currentaffairs.org': 0.8978420868913102,
'arc.aiaa.org': 0.897839844709445,
'youtube.googleblog.com': 0.8978353788604845,
'www.pulitzer.org': 0.8977920724736956,
'computer.rip': 0.8977456785440727,
'blog.dgraph.io': 0.8977112410728876,
'blog.fastforwardlabs.com': 0.8977025051645369,
'www.bridgewater.com': 0.8976967291200032,
'mycroft.ai': 0.8976933371034872,
'open.salon.com': 0.8976917672314974,
'blog.longreads.com': 0.8976472678414861,
'lists.ubuntu.com': 0.897645348573024,
'nvie.com': 0.8976386294805203,
'adamdrake.com': 0.8976376774058287,
'www.cordcuttersnews.com': 0.8975937131370894,
'www.kayak.com': 0.8975688924284162,
'aiexperiments.withgoogle.com': 0.8975573917462177,
'gustedt.wordpress.com': 0.8975532085064689,
'www.adobe.com': 0.8975379400846643,
'miter.mit.edu': 0.8975316961912847,
'blog.statuspage.io': 0.8975269760160564,
'hyperpolyglot.org': 0.8975160197448814,
'www.scaleway.com': 0.8974907733168768,
'bennettfeely.com': 0.8974804776492202,
'joshwcomeau.com': 0.8974719675010521,
'www.ktvu.com': 0.8974664720007207,
'www.tomsguide.com': 0.897458826143719,
'www.jitbit.com': 0.8973926361586634,
'time.is': 0.8973510461083218,
'tinyclouds.org': 0.8973083698372893,
'gurjeet.singh.im': 0.8973075039780294,
'blog.mailgun.net': 0.8972959119527889,
'harry.me': 0.8972525836598453,
'archpaper.com': 0.8972054834238723,
'yoric.github.io': 0.897152918128832,
'community.atlassian.com': 0.8971438675267877,
'news.ycombinator.com': 0.8971152540222578,
'aviation.stackexchange.com': 0.8970966595562585,
'scholars-stage.blogspot.com': 0.8970933350643355,
'durmonski.com': 0.897088311771032,
'blog.zorinaq.com': 0.8970730785733094,
'blog.peepcode.com': 0.8970173769216392,
'kivy.org': 0.8969978306759151,
'www.crowdtilt.com': 0.8969827660481152,
'mathematica.stackexchange.com': 0.8969710211525342,
'www.wyden.senate.gov': 0.8969688474969271,
'watsi.org': 0.8969438303261918,
'igor-blue.github.io': 0.8968776201906612,
'medium.economist.com': 0.8968759451432184,
'www.wolframalpha.com': 0.8968667656586023,
'highlyscalable.wordpress.com': 0.8968525375322202,
'in.reuters.com': 0.8967974687633106,
'www.nethack.org': 0.8967951042087746,
'tidbits.com': 0.8967894768726519,
'theappendix.net': 0.8967463599409581,
'tech.fortune.cnn.com': 0.8967131583500147,
'www.fossjobs.net': 0.8966914843070399,
'blog.robinhood.com': 0.8966536394829941,
'www.realtimerendering.com': 0.8966481963801871,
'blog.theincredibleholk.org': 0.8966222436790824,
'www.ee.ryerson.ca': 0.8965601151332392,
'databricks.com': 0.8965596021498984,
'www.defensenews.com': 0.8965587750067616,
'rcoh.me': 0.896554024229452,
'mijailovic.net': 0.8965535170148166,
'gallium.inria.fr': 0.8965486747498262,
'blog.optimizely.com': 0.896542105323571,
'blog.replit.com': 0.8965338803932761,
'learn.adafruit.com': 0.8965131842089722,
'joelrunyon.com': 0.8964677519262857,
'blog.frizk.net': 0.8964121097275121,
'alarmingdevelopment.org': 0.8964094023716221,
'robots.thoughtbot.com': 0.8963990601096038,
'www.includeos.org': 0.8963879787342307,
'mrcoles.com': 0.8963879787342307,
'c2.com': 0.8963178119940578,
'www.whonix.org': 0.8963115990850473,
'fivebooks.com': 0.8963105340034757,
'reprog.wordpress.com': 0.8963054521368881,
'www.wilfred.me.uk': 0.8962945507182206,
'whydoeseverythingsuck.com': 0.8962216827001023,
'www.pcworld.com': 0.8962205591216735,
'www.bamsoftware.com': 0.8961964199305918,
'www.cs.unm.edu': 0.8961816656276443,
'www.idris-lang.org': 0.8961766972485017,
'www.raspberrypi.com': 0.8961620944863709,
'effectiviology.com': 0.8961424475553789,
'asciiflow.com': 0.8961224916388418,
'philosophically.com': 0.896116599326938,
'www.alexa.com': 0.8961037898681212,
'algorithmicbotany.org': 0.8960953021444669,
'yasoob.me': 0.8960526720715563,
'varun.ca': 0.896004489709373,
'medievalbooks.nl': 0.8959983974915444,
'www.kgw.com': 0.895989268551949,
'www.junglecoder.com': 0.8959808787419773,
'insideintercom.io': 0.8959693504237235,
'www.html5rocks.com': 0.8959295211749472,
'www.arxiv-vanity.com': 0.8959180319048996,
'micahflee.com': 0.8958958043171017,
'www.nccgroup.trust': 0.8958587838339611,
'c0de517e.blogspot.com': 0.8958503147286802,
'www.foreignpolicy.com': 0.8958091411993182,
'ashtonkemerling.com': 0.8957869981475534,
'www.amusingplanet.com': 0.8957821987841423,
'lea.verou.me': 0.895732027114529,
'nationalinterest.org': 0.8957204150819327,
'lists.webkit.org': 0.8956698653136842,
'opkode.com': 0.8956558833770757,
'www-formal.stanford.edu': 0.89562877673962,
'www.michaelgeist.ca': 0.8956008548612578,
'theory.stanford.edu': 0.8955882277276088,
'i.stanford.edu': 0.8955669355788611,
'exstreamist.com': 0.8955604388524536,
'citeseerx.ist.psu.edu': 0.8955488558889958,
'bugzilla.redhat.com': 0.8955216901346179,
'www.cmu.edu': 0.8954764227994559,
'0xax.gitbooks.io': 0.8954552987943115,
'jasoncrawford.org': 0.8954419392195877,
'www.lego.com': 0.8954257332135742,
'discourse.julialang.org': 0.8954216385277134,
'www.open-std.org': 0.8953539493751335,
'www.indiatoday.in': 0.8953429983961907,
'www.winestockwebdesign.com': 0.8953288508285574,
'mobiforge.com': 0.8953182171062833,
'www.regulations.gov': 0.895303818266352,
'www.salesforce.com': 0.8952981434270118,
'www.trackinghappiness.com': 0.8952717597076212,
'blackhole12.blogspot.com': 0.8952521384621974,
'blog.infinitenegativeutility.com': 0.8952329903252306,
'www.opera.com': 0.8952211159380423,
'blogs.balsamiq.com': 0.8952154811855707,
'treyhunner.com': 0.8952154811855707,
'constructionphysics.substack.com': 0.8951823466408914,
'beyondloom.com': 0.8951761937264585,
'www.lennyrachitsky.com': 0.8951749721627695,
'haskell.cs.yale.edu': 0.8951742276618677,
'raymii.org': 0.8951548310801417,
'llogiq.github.io': 0.8951452226380562,
'blog.amirkhella.com': 0.8950945164652512,
'www.mooreds.com': 0.8950941927209954,
'www.hudsandguis.com': 0.895021645021645,
'flask.pocoo.org': 0.89499976188154,
'www.dol.gov': 0.8949918028432319,
'wololo.net': 0.8949102546729428,
'dashes.com': 0.894908282336013,
'mortoray.com': 0.8948856187046399,
'www.mediawiki.org': 0.8948851399600354,
'developerblog.redhat.com': 0.8948826280677145,
'www.banterly.net': 0.8948384778162753,
'www.justin.tv': 0.8948269726237952,
'www.koreaherald.com': 0.8947866914467383,
'two-wrongs.com': 0.8947866914467383,
'www.3ders.org': 0.8947587499828596,
'millcomputing.com': 0.894753411664112,
'remysharp.com': 0.894728785876003,
'magenta.tensorflow.org': 0.8947055951377291,
'www.techradar.com': 0.8947023301779632,
'www.pushing-pixels.org': 0.8946997760143556,
'sagemath.blogspot.com': 0.8946383392529572,
'mymodernmet.com': 0.8946360917364401,
'jamesmcm.github.io': 0.8946346573429327,
'designyoutrust.com': 0.8945718808483918,
'boards.4chan.org': 0.8945658314381272,
'blog.framebase.io': 0.8945652173913043,
'openwrt.org': 0.8945648303938399,
'blog.codecombat.com': 0.8945386717554477,
'www.informit.com': 0.8945265591993962,
'www.evilsocket.net': 0.894519291668392,
'www.governing.com': 0.8945150467351556,
'www.masteringemacs.org': 0.894486414286821,
'tomassetti.me': 0.8944815712239113,
'gun.io': 0.8944737983455031,
'www.netlify.com': 0.8944433597452348,
'www.ntsb.gov': 0.8944261516293643,
'www.cs.kent.ac.uk': 0.8943581679658725,
'www.kernel.org': 0.8943363945858583,
'www.flyertalk.com': 0.8943293407889205,
'www.unlimitednovelty.com': 0.8943013001071519,
'math.andrej.com': 0.8942887015711656,
'healeycodes.com': 0.8942310355364101,
'blog.jmdawson.co.uk': 0.894210852992969,
'metaoptimize.com': 0.8942001103153372,
'mynorthwest.com': 0.894171800552717,
'asciinema.org': 0.8941695108843633,
'pydanny.com': 0.8941547728795499,
'warisboring.com': 0.8940977556210705,
'blog.serverfault.com': 0.8940908548836543,
'narratively.com': 0.8940876544990076,
'freakonomics.com': 0.8940592156715558,
'www.joindiaspora.com': 0.8940510018896056,
'www.caixinglobal.com': 0.8940493754864369,
'guccifer2.wordpress.com': 0.8940429649134846,
'www.freelists.org': 0.894038057617328,
'devblog.bu.mp': 0.8940332414895527,
'mislav.uniqpath.com': 0.894024162032998,
'www.heydonworks.com': 0.8939995146383073,
'glyph.twistedmatrix.com': 0.8939986046314382,
'paulirish.com': 0.8939908517980819,
'a1k0n.net': 0.8939890710382514,
'www.vldb.org': 0.893980099250682,
'www.squeakland.org': 0.8939334940011854,
'america.aljazeera.com': 0.8939301873013243,
'www.epi.org': 0.8939255228575806,
'tryolabs.com': 0.8938904483824683,
'willcrichton.net': 0.8938879993790215,
'news.lenovo.com': 0.8938868214729069,
'paulbernal.wordpress.com': 0.8938730853391685,
'www.psmag.com': 0.8938507630159833,
'meiert.com': 0.8938177244811635,
'www.ghacks.net': 0.8938115780468081,
'www.shamusyoung.com': 0.8938032619934929,
'zeptobars.com': 0.893764616861537,
'0x0f0f0f.github.io': 0.893764616861537,
'cm.bell-labs.com': 0.8937058283807041,
'www.fastcolabs.com': 0.8936928790642708,
'aging.nautil.us': 0.8936740461671302,
'hackerone.com': 0.8936672524995389,
'storage.courtlistener.com': 0.893640350877193,
'danlebrero.com': 0.893623066742338,
'mapzen.com': 0.8936063685534883,
'bernsteinbear.com': 0.8935751450877738,
'austinmann.com': 0.8935501465273382,
'www.vinaysahni.com': 0.8935481547848754,
'burakkanber.com': 0.8935378844831082,
'mollyrocket.com': 0.8935354120568186,
'feeding.cloud.geek.nz': 0.8935244667500523,
'cliffle.com': 0.8934905766791413,
'www.mikeperham.com': 0.8934692786771169,
'www.sequoiacap.com': 0.8934381407790684,
'www.thestranger.com': 0.8934172181855746,
'python.org': 0.8933810789765779,
'www.datacenterdynamics.com': 0.893374339023219,
'www.nongnu.org': 0.8933522846439509,
'blog.pkh.me': 0.8933285357815479,
'greenteapress.com': 0.8933064903929183,
'debarghyadas.com': 0.8933044930066333,
'jashkenas.github.com': 0.8933044930066333,
'jborichevskiy.com': 0.8932713559026754,
'www.saturdayeveningpost.com': 0.8932203537854467,
'www.engineyard.com': 0.8931369083416167,
'jeremymikkola.com': 0.8931346519747027,
'chriseidhof.tumblr.com': 0.893107113349865,
'www.the-scientist.com': 0.8931050183937893,
'heterodoxacademy.org': 0.8930676522754398,
'daveceddia.com': 0.8930562709346127,
'eighty-twenty.org': 0.8929575630252592,
'itvision.altervista.org': 0.8928942922265454,
'pogue.blogs.nytimes.com': 0.892885991203199,
'science.sciencemag.org': 0.8928844657775074,
'www.citypages.com': 0.8928735399686778,
'www.houstonchronicle.com': 0.8928722277760239,
'www.perl.com': 0.8928505576437601,
'www.mcclatchydc.com': 0.892841447269041,
'www.over-yonder.net': 0.8928301959201767,
'donatstudios.com': 0.8928301959201767,
'ucsdnews.ucsd.edu': 0.8928192288277865,
'www.johnmyleswhite.com': 0.8928134392551893,
'vorner.github.io': 0.8928096417930931,
'nikic.github.com': 0.8928090193925269,
'samsclass.info': 0.8927237027135478,
'markmaunder.com': 0.8926951731429821,
'tauday.com': 0.8926800567850355,
'riscv.org': 0.8926238746067623,
'blog.blazingdb.com': 0.8926082454543174,
'garrett.damore.org': 0.892590826266076,
'www.ilikebigbits.com': 0.8925607594811928,
'mediagoblin.org': 0.8925318168757916,
'keithclark.co.uk': 0.8925217211946057,
'reviews.llvm.org': 0.8925217211946057,
'www.census.gov': 0.8925111327414155,
'www.carbonbrief.org': 0.8925108441158517,
'1000memories.com': 0.8924866571941482,
'tommorris.org': 0.8924689354696166,
'csswizardry.com': 0.8924685362435004,
'simonmar.github.io': 0.8924611973392461,
't3x.org': 0.8924420140543968,
'alpinelinux.org': 0.8924276788770722,
'news.blizzard.com': 0.8924063008658305,
'lowendmac.com': 0.892380727879953,
'jakubmarian.com': 0.8923725482586792,
'smackdown.blogsblogsblogs.com': 0.8923365014823826,
'www.gamefromscratch.com': 0.8922886748833608,
'www.history.com': 0.892284563335802,
'www.aps.org': 0.8922636239768308,
'bitbashing.io': 0.8922618659350428,
'thecreativeindependent.com': 0.8922525349154249,
'abcnews.go.com': 0.8922472028702633,
'lilianweng.github.io': 0.8922138047648334,
'filecoin.io': 0.8921882836554222,
'kristerw.blogspot.com': 0.8921827368433277,
'www.archdaily.com': 0.8921786726941157,
'critter.blog': 0.8921605201397375,
'blog.uber.com': 0.8921471734712938,
'blog.darklang.com': 0.8921367030876631,
'youteam.io': 0.8921269513814725,
'developer.oculus.com': 0.8921230866971346,
'www.airbnb.com': 0.8921210442927137,
'www.caitlingreen.org': 0.8921201754031437,
'stuffin.space': 0.8921201754031437,
'www.logicmatters.net': 0.8921008427968538,
'blogs.atlassian.com': 0.8920598317674283,
'www.coindesk.com': 0.8920474723589183,
'blog.checklyhq.com': 0.8920323362989564,
'www.reifyworks.com': 0.8920283981813807,
'www.tensorflow.org': 0.8919880973555573,
'medicalxpress.com': 0.8919405231029118,
'lists.wikimedia.org': 0.8919231093801725,
'www.cabinetmagazine.org': 0.8919061013347211,
'vez.mrsk.me': 0.8918998921036764,
'tenderlovemaking.com': 0.8918462339278702,
'timdettmers.com': 0.8918366190595837,
'jpkoning.blogspot.com': 0.8918251594506603,
'www.timesonline.co.uk': 0.8917973535504273,
'pdfs.semanticscholar.org': 0.8917971680021309,
'www.trustwave.com': 0.8917680987246398,
'ambrevar.xyz': 0.8917657796004286,
'www.theroot.com': 0.8917542491581687,
'tylertringas.com': 0.8917090979583869,
'www.ncsc.gov.uk': 0.8916672988164659,
'globenewswire.com': 0.8916669637212149,
'www.applevis.com': 0.8916621885801436,
'www.metafilter.com': 0.8916619189169765,
'erickhun.com': 0.8916502534520422,
'blog.wearewizards.io': 0.8916476923173318,
'cpj.org': 0.8916369716407726,
'boyter.org': 0.891577544859255,
'm.facebook.com': 0.8915642265642092,
'www.calnewport.com': 0.8915632480334919,
'www.profgalloway.com': 0.8915466752832707,
'mentallandscape.com': 0.8915429443685636,
'm.theatlantic.com': 0.891523653721213,
'hackertarget.com': 0.8915152085122275,
'www.pyret.org': 0.8914988814317674,
'www.obdev.at': 0.8914980426595512,
'dash.harvard.edu': 0.8914784869901488,
'engineeringblog.yelp.com': 0.8914654287085163,
'www.mercurial-scm.org': 0.8914323327846669,
'datanitro.com': 0.8914234373706179,
'www.huffingtonpost.in': 0.8914121923586303,
'fangpenlin.com': 0.891404181051816,
'discovermagazine.com': 0.8914023623580617,
'streetlifesolutions.blogspot.com': 0.8914002514918445,
'blog.documentfoundation.org': 0.8913787017013197,
'dotsies.org': 0.8913773796192609,
'tails.boum.org': 0.8913724994898894,
'randfishkin.com': 0.8912627828106103,
'tomasp.net': 0.8911820823224039,
'www.taiwannews.com.tw': 0.8911580981808408,
'modern-sql.com': 0.891151176073313,
'carlos.bueno.org': 0.8911385230462937,
'www.kansascity.com': 0.8910854933759852,
'english.stackexchange.com': 0.8910824248780256,
'cdm.link': 0.8910804711300865,
'nighthacks.com': 0.8910750341672654,
'blog.vrypan.net': 0.8910540018083172,
'nyxt.atlas.engineer': 0.8910326186239637,
'blog.evernote.com': 0.8909942969284109,
'news.bbc.co.uk': 0.89099404579875,
'securitylab.github.com': 0.8909843257796506,
'pjreddie.com': 0.8909657104816301,
'fury.com': 0.8909653233291642,
'blog.skylight.io': 0.8909427640646718,
'kickstarter.engineering': 0.8909427640646718,
'www.darpa.mil': 0.8909203771658175,
'psyarxiv.com': 0.8909128838846586,
'spottedsun.com': 0.89089016985033,
'matt.aimonetti.net': 0.8907810868277886,
'news.dartlang.org': 0.8907684232451024,
'musescore.org': 0.8907008278994514,
'www.apa.org': 0.8906903176590568,
'www.flightglobal.com': 0.8906785542989906,
'www.theatlanticcities.com': 0.8906598207125368,
'search.cpan.org': 0.8906114753639363,
'www.p01.org': 0.8906084333939481,
'blog.dantup.com': 0.8906084333939481,
'thedailywtf.com': 0.8905424089852495,
'e.foundation': 0.8905340551839913,
'cookieplmonster.github.io': 0.8905252719934627,
'news.php.net': 0.8905175241100567,
'toroid.org': 0.8905145635041263,
'www.kk.org': 0.8904878920768752,
'forum.sublimetext.com': 0.890452944414247,
'praxis.fortelabs.co': 0.890452944414247,
'leostartsup.com': 0.8904132589825239,
'web.cecs.pdx.edu': 0.890412029578113,
'www.ofbrooklyn.com': 0.8903954802259887,
'www.revk.uk': 0.8903240899854906,
'rawkes.com': 0.8903199760734138,
'furbo.org': 0.8903142798666888,
'nickoneill.com': 0.8902859706962722,
'compilers.iecc.com': 0.8902788378277958,
'www.climatechangenews.com': 0.8902788378277958,
'avichal.com': 0.8902714932126697,
'www.healthcaredive.com': 0.8902714932126697,
'lucasfcosta.com': 0.8902709782517574,
'stevecoast.com': 0.8902478204961621,
'dariusforoux.com': 0.8902302723631834,
'msdn.microsoft.com': 0.8901788121850751,
'www.nzz.ch': 0.8901568608802078,
'trac.webkit.org': 0.8901554708541648,
'www.waze.com': 0.8901472253680635,
'www.liip.ch': 0.8901472253680635,
'ericlippert.com': 0.8901127718368207,
'matpalm.com': 0.8901030282480019,
'roshfu.com': 0.8900305511261404,
'adeshpande3.github.io': 0.8900188928475846,
'www.factmag.com': 0.8900054823385649,
'grafana.org': 0.8899942321734828,
'www.divio.com': 0.889923339928156,
'www.law.cornell.edu': 0.8899173371280885,
'www.cs.huji.ac.il': 0.8899161000679049,
'gothamist.com': 0.8899111843846533,
'www.ajc.com': 0.8899040092884533,
'mathpix.com': 0.8899039505478837,
'www.vulture.com': 0.8899034479557729,
'bair.berkeley.edu': 0.8898667877223271,
'www.goldsborough.me': 0.8897894354492999,
'arkadiyt.com': 0.8897865060368211,
'flowingdata.com': 0.8897859193800971,
'grist.org': 0.8897796023630788,
'clusterhq.com': 0.889770984328511,
'postmarketos.org': 0.8897648371785116,
'monkey.org': 0.8897626687452419,
'www.coderholic.com': 0.8897409044066217,
'blog.silktide.com': 0.8897409044066216,
'docs.house.gov': 0.88972591966882,
'colab.research.google.com': 0.8897224911816017,
'www.nhs.uk': 0.8896805121516772,
'insideclimatenews.org': 0.8896793812620282,
'torvalds-family.blogspot.com': 0.8896665476254435,
'www.holloway.com': 0.8896662494413392,
'www.airspacemag.com': 0.8896507942259744,
'grsecurity.net': 0.889646047305093,
'www.kevinrooke.com': 0.8896372799856362,
'engineering.gusto.com': 0.8896147691691174,
'blog.christianperone.com': 0.8896015309843164,
'www.badscience.net': 0.889589692438837,
'www.jerf.org': 0.889547641907146,
'vfoley.xyz': 0.8895216400911161,
'www.dangrossman.info': 0.8895193470455725,
'blog.timac.org': 0.8895027671338473,
'www.nme.com': 0.8894941034343402,
'withblue.ink': 0.8894380029474798,
'www.modernperlbooks.com': 0.8894320825157749,
'guix.gnu.org': 0.8894056608506781,
'vasilishynkarenka.com': 0.8894015273075921,
'chrome.google.com': 0.8893920540006104,
'startupboy.com': 0.8893837967279286,
'thesoundingline.com': 0.8893544871283143,
'blog.kotowicz.net': 0.8893507542723823,
'www.ubuntuvibes.com': 0.8893113229443974,
'www.tindie.com': 0.8892920494549288,
'www.falstad.com': 0.8892263646054737,
'alisdair.mcdiarmid.org': 0.8892079775948455,
'www.brennancenter.org': 0.8892079775948455,
'theanarchistlibrary.org': 0.8891911407644053,
'www.cftc.gov': 0.8891651323421873,
'lisp-journey.gitlab.io': 0.8891428571428571,
'www.namepros.com': 0.8891410371836007,
'eu.usatoday.com': 0.8890682694542983,
'blog.canonical.com': 0.8890615085402753,
'sahandsaba.com': 0.88904655549824,
'blog.rstudio.org': 0.8890367675160065,
'neilkakkar.com': 0.8890307804355818,
'news.nationalpost.com': 0.8890115573593319,
'www.snopes.com': 0.8889924719688463,
'www.brainpickings.org': 0.8889686116306911,
'mitmproxy.org': 0.8889639851594255,
'blog.balancedpayments.com': 0.888958456392269,
'breakingdefense.com': 0.8889416410999423,
'markets.businessinsider.com': 0.8889221685107183,
'jerrygamblin.com': 0.8889200836058827,
'angularjs.blogspot.com': 0.8888770420833161,
'www.shopify.com': 0.8888551514369234,
'www.helloerik.com': 0.8888428887518348,
'languagelog.ldc.upenn.edu': 0.8888040353901259,
'diogomonica.com': 0.888789480686917,
'oversight.house.gov': 0.8887677241378522,
'blog.yossarian.net': 0.8887466744525089,
'tim.blog': 0.8887179571174896,
'blog.marcocantu.com': 0.8887010787021495,
'www.stilldrinking.org': 0.8886969078588848,
'plentyoffish.wordpress.com': 0.8886858465608289,
'troubles.md': 0.8886733614347104,
'8bitworkshop.com': 0.8886716751051769,
'plato.stanford.edu': 0.8886226501737253,
'blog.watsi.org': 0.8886099090504368,
'www.vatican.va': 0.8886001974662521,
'www.gridsagegames.com': 0.8885999017945436,
'blog.wakemate.com': 0.8885718128809637,
'www.guardian.co.uk': 0.8885578778095446,
'www.eastbaytimes.com': 0.8885533867034545,
'insights.ubuntu.com': 0.8885369737806736,
'www.uber.com': 0.888502360897741,
'hharnisc.github.io': 0.8884780095069668,
'blog.eyas.sh': 0.8884641430053347,
'adlrocha.substack.com': 0.8884402828343574,
'www.agwa.name': 0.8883681305045957,
'lexfridman.com': 0.8883669315681367,
'www.hp.com': 0.8883541180846748,
'chargen.matasano.com': 0.888350093124338,
'docs.github.com': 0.888346612251366,
'blogs.esa.int': 0.8883268060163584,
'hackage.haskell.org': 0.8882716929118732,
'www.askthepilot.com': 0.8882206989905654,
'investor.fb.com': 0.8881776846453903,
'hackertourism.com': 0.8881603547360667,
'varianceexplained.org': 0.8881303906523573,
'www.stanford.edu': 0.8881280238738776,
'googleappengine.blogspot.com': 0.8881149640019546,
'blog.spark.io': 0.8880887280047254,
'danielmiessler.com': 0.8880772991848912,
'blog.cloudant.com': 0.8880720742931111,
'help.netflix.com': 0.888032506658475,
'www.consumer.ftc.gov': 0.8880113570637971,
'thechangelog.com': 0.8880056295154011,
'mrdoob.com': 0.8879829721117504,
'www.saveur.com': 0.8879678355351808,
'www.moderntreasury.com': 0.8879300370907698,
'blogs.suntimes.com': 0.8879083159007929,
'staff.tumblr.com': 0.8878771427989084,
'archive.wired.com': 0.8878447305921355,
'www.ironicsans.com': 0.8878289779246122,
'blog.capitalandgrowth.org': 0.8878130504973006,
'www.thisiscolossal.com': 0.8878105455839786,
'benchling.engineering': 0.8878049780860752,
'www.symmetrymagazine.org': 0.8877876402671063,
'www.feld.com': 0.8877667891871323,
'everynoise.com': 0.8877314814814815,
'input.fontbureau.com': 0.8877209761748434,
'qiao.github.io': 0.8877030948951334,
'factordaily.com': 0.887680355039973,
'ecc-comp.blogspot.com': 0.887651452722378,
'addyosmani.com': 0.8876225519241614,
'reallifemag.com': 0.8876055526528979,
'niedzejkob.p4.team': 0.8875742399340828,
'www.cs.purdue.edu': 0.887560497701347,
'vercel.com': 0.8875553045329022,
'blog.sesse.net': 0.887545005792023,
'mynoise.net': 0.8875240702669522,
'nevalalee.wordpress.com': 0.8874709976798144,
'www.justingarrison.com': 0.8874550740684668,
'www.cse.buffalo.edu': 0.887431083352139,
'jamesgolick.com': 0.8874059947114532,
'gnunet.org': 0.887393102678099,
'www.tampabay.com': 0.8873751025878571,
'www.jide.com': 0.887368855995013,
'producthabits.com': 0.8873611172314394,
'www.fourhourworkweek.com': 0.8873408966020546,
'refer.ly': 0.8873384440828381,
'www.aei.org': 0.8873351640879104,
'www.mentalfloss.com': 0.8873305275256577,
'izbicki.me': 0.8873178158351525,
'mathiasbynens.be': 0.8872781506210743,
'www.pcjs.org': 0.8872494045912365,
'arcan-fe.com': 0.887245072438425,
'www.americamagazine.org': 0.8872386352495559,
'mailman.nanog.org': 0.8872223978170161,
'bluejekyll.github.io': 0.8872093023255814,
'www.mysliderule.com': 0.8872093023255814,
'blogs.oracle.com': 0.8871952202335835,
'leonardofed.io': 0.8871763351184384,
'www.sourcetrail.com': 0.8871714606492734,
'www.linuxfromscratch.org': 0.8871714606492734,
'bariweiss.substack.com': 0.8871587544964651,
'moot.it': 0.8871570315823085,
'popula.com': 0.887128977917166,
'press.web.cern.ch': 0.887082617470916,
'ownyourbits.com': 0.8870779976717112,
'getstream.io': 0.8870262897116983,
'scpd.stanford.edu': 0.8870121302239351,
'dcgross.com': 0.8868939399089152,
'unsupervisedmethods.com': 0.8868826572298285,
'www.willmcgugan.com': 0.8868766525394289,
'wiki.tcl-lang.org': 0.8868748835508599,
'blog.disqus.com': 0.8868414318888773,
'blogs.technet.microsoft.com': 0.8868323608446661,
'tromp.github.io': 0.8867756037157464,
'blog.ircmaxell.com': 0.8867580271160181,
'mux.com': 0.8867579106919085,
'bugs.mysql.com': 0.8867572809726417,
'www.irishtimes.com': 0.8867543138710346,
'dpc.pw': 0.8867103394982209,
'dusted.codes': 0.8866966751368195,
'quanticdev.com': 0.8866903311266173,
'www.neowin.net': 0.8866503890472428,
'chrislord.net': 0.886641393239182,
'blog.otoro.net': 0.8866334244061989,
'www.thenational.scot': 0.8866271172090046,
'www.inquirer.com': 0.8865967718863557,
'modelviewculture.com': 0.8865845558030554,
'info.iet.unipi.it': 0.8865497076023392,
'www.storehouse.co': 0.886548015153011,
'www.anfractuosity.com': 0.8865113264892915,
'danwang.co': 0.8864808931374831,
'www.researchgate.net': 0.8864516461055575,
'www.doxsey.net': 0.8864376154183435,
'bugs.ruby-lang.org': 0.8864368986388406,
'whyy.org': 0.8864168618266979,
'itch.io': 0.8864120842802148,
'amigalove.com': 0.8863992170741539,
'people.mozilla.com': 0.8863952990075262,
'medium.freecodecamp.com': 0.8863811065750393,
'blog.doist.com': 0.8863606429850415,
'www.swyx.io': 0.8863599205692547,
'www.eidel.io': 0.886301279533414,
'senrigan.io': 0.8862368459608536,
'www.pvk.ca': 0.8861969908332605,
'www.ccs.neu.edu': 0.8861380071332806,
'cturt.github.io': 0.8861102984083253,
'www.lysator.liu.se': 0.8860844730714155,
'philcalcado.com': 0.8859388762978412,
'lists.dragonflybsd.org': 0.8859388762978412,
'microkerneldude.wordpress.com': 0.8859288068767922,
'www.theringer.com': 0.8858962060628648,
'bzg.fr': 0.8858719540548398,
'www.crypto.com': 0.8858494594971499,
'fibery.io': 0.8858217047605946,
'kimonote.com': 0.885805964130514,
'news.softpedia.com': 0.8857837864929501,
'successfulsoftware.net': 0.8857784923477241,
'blog.intercom.com': 0.8857763495575585,
'www.embeddedrelated.com': 0.8857676106438698,
'www.psychologytoday.com': 0.8857585049003665,
'haveibeenpwned.com': 0.8856992747056773,
'blog.ryotak.me': 0.8856929365305167,
'blog.piston.rs': 0.8856929365305167,
'joshmitteldorf.scienceblog.com': 0.8855962693776582,
'www.reviews.com': 0.8855218005991683,
'svpow.com': 0.8855122462369361,
'blog.nahurst.com': 0.8855122462369361,
'engineering.kablamo.com.au': 0.8854939252840581,
'blog.adafruit.com': 0.8854467055810114,
'earth.nullschool.net': 0.8854306983539961,
'inventwithpython.com': 0.885393853053717,
'chalkdustmagazine.com': 0.8853197575419666,
'blog.objectmentor.com': 0.8852948669873847,
'www.aholme.co.uk': 0.8852815790604109,
'www.dkriesel.com': 0.8852756720656075,
'vanemden.wordpress.com': 0.8852604927216077,
'investor.google.com': 0.8852562818553033,
'pipeline.corante.com': 0.8852387066889292,
'action.eff.org': 0.8852266245403357,
'blog.twitch.tv': 0.8852196608142338,
'www.nand2tetris.org': 0.8852071005917159,
'gpfault.net': 0.8851652261576463,
'meshedinsights.com': 0.8851628232067584,
'probcomp.csail.mit.edu': 0.8851438011782188,
'missiveapp.com': 0.8851280394688669,
'blog.sendgrid.com': 0.8851277655579799,
'mitadmissions.org': 0.8851045058363526,
'info.meteor.com': 0.8850985466953342,
'thewirecutter.com': 0.8850944093158075,
'www.aboveavalon.com': 0.8850896030144491,
'haxe.org': 0.8850332564592883,
'www.snarky.ca': 0.8850298847276258,
'blog.mapillary.com': 0.8849756787053948,
'www.alternet.org': 0.8849651154014133,
'martinheinz.dev': 0.884952570972269,
'www.stevebenjamins.com': 0.8849347568208777,
'derwiki.tumblr.com': 0.8849077586898806,
'www.kuro5hin.org': 0.8848948828327496,
'www.archlinux.org': 0.8848794541675421,
'www.bennjordan.com': 0.8848672501882364,
'tromey.com': 0.8848672501882364,
'medlifestyle.news': 0.8848240975158286,
'hackpad.com': 0.8848240112405613,
'zinascii.com': 0.8848232963497987,
'heredragonsabound.blogspot.com': 0.8848232963497987,
'pycon.blogspot.com': 0.8847972407140463,
'thegenesisblock.com': 0.8847466884910953,
'jayriverlong.github.io': 0.8846854627897014,
'rkoutnik.com': 0.8846701041470997,
'moalquraishi.wordpress.com': 0.8846503798740485,
'blog.angularjs.org': 0.8846476203665491,
'learnyousomeerlang.com': 0.8846416196747935,
'blog.rubymotion.com': 0.8846052536362908,
'www.berkshirehathaway.com': 0.8845734385329344,
'bigstory.ap.org': 0.8845644349212194,
'www.dji.com': 0.8845531462931538,
'homes.cs.washington.edu': 0.884536530239123,
'momjian.us': 0.8845249482528319,
'anderspitman.net': 0.8845238095238095,
'ubiquity.acm.org': 0.8845123969111347,
'archive.vector.org.uk': 0.8844858003347821,
'bitcoinfoundation.org': 0.8844841517497576,
'esr.ibiblio.org': 0.884465525984111,
'www.caranddriver.com': 0.8844455469996317,
'www.bristol.ac.uk': 0.8844312836410105,
'www.keacher.com': 0.8844188041447092,
'www.planet.com': 0.8843849199350312,
'www.braintreepayments.com': 0.8843684120647333,
'www.topic.com': 0.8843258553540223,
'www.lyx.org': 0.8843101335952557,
'hal.inria.fr': 0.884278551246099,
'blog.powerdns.com': 0.8842750467710181,
'www.discovermeteor.com': 0.8842406817313907,
'www.statista.com': 0.884226458408693,
'diveintohtml5.org': 0.884224840703763,
'shkrobius.livejournal.com': 0.8842248407037628,
'danwin.com': 0.8842240470222248,
'thingist.com': 0.8841700109782021,
'sf.curbed.com': 0.8841684697837997,
'www.artima.com': 0.8840774930170289,
'www.gregegan.net': 0.8840639202320028,
'veed.io': 0.8840295485188058,
'www.kguttag.com': 0.8840223725080164,
'www.passwordstore.org': 0.8839871084196906,
'www.time.com': 0.8839776340050455,
'patft.uspto.gov': 0.8839759157295332,
'brikis98.blogspot.com': 0.8839185792660761,
'luckypeach.com': 0.8839078718615478,
'www.256kilobytes.com': 0.8839065999741562,
'kamalmarhubi.com': 0.8838909108920787,
'www.levels.fyi': 0.8838529264451486,
'www.grubstreet.com': 0.8838363484488945,
'kubernetes.io': 0.8838168079942494,
'privpapers.ssrn.com': 0.8837957120914789,
'finledger.com': 0.8837957120914789,
'ircv3.net': 0.8837369391764082,
'webtorrent.io': 0.8836930455635492,
'animagraffs.com': 0.8836872834930155,
'www.coinbase.com': 0.8836702182178287,
'jobs.apple.com': 0.8836631610037271,
'lifehacker.com': 0.88361719985225,
'blog.mongodb.org': 0.883606109678017,
'lobe.ai': 0.8835655842394832,
'blogs.ams.org': 0.8835451368221627,
'www.psychologicalscience.org': 0.8834931862797779,
'www.autonews.com': 0.8834926699108666,
'google-styleguide.googlecode.com': 0.8834824574967941,
'venomousporridge.com': 0.8834619351905811,
'www.bennorthrop.com': 0.8834521090229532,
'blog.lusis.org': 0.8834395848799916,
'www.defcon.org': 0.8834198543134106,
'railstips.org': 0.8834172428920269,
'sitn.hms.harvard.edu': 0.8834134615384616,
'www.johnwdefeo.com': 0.8834118239436308,
'www.metabase.com': 0.8833568908825655,
'jordansmith.io': 0.8833386064130131,
'blog.brokep.com': 0.8833292707180577,
'davnicwil.com': 0.883290787723497,
'daverupert.com': 0.8832817007531839,
'www.internalpointers.com': 0.8832731648616126,
'blog.scottnonnenberg.com': 0.8832680173548925,
'webmasters.googleblog.com': 0.8832159517897961,
'ivanceras.github.io': 0.8831795673307369,
'www.inference.vc': 0.8831370240507179,
'white.ucc.asn.au': 0.883132530120482,
'dis.4chan.org': 0.8831325301204819,
'mta.openssl.org': 0.8831205226464401,
'blog.sourcerer.io': 0.8831047861347778,
'callumprentice.github.io': 0.8830898580872119,
'www.techtransparencyproject.org': 0.8830469812134585,
'blog.adobe.com': 0.883040341397881,
'agfundernews.com': 0.8830325567122732,
'www.centauri-dreams.org': 0.8829955449207836,
'documentcloud.github.com': 0.8829915560916767,
'rust-lang.github.io': 0.8829915560916767,
'portswigger.net': 0.8829877514947376,
'tagide.com': 0.8829745187840874,
'creativecommons.org': 0.8829668579998763,
'flak.tedunangst.com': 0.8829644456502317,
'ferrous-systems.com': 0.8829609951535142,
'shapecatcher.com': 0.8829477368055727,
'solovyov.net': 0.8829399933646876,
'blog.pmarca.com': 0.8829123044832685,
'linusakesson.net': 0.8829080057671795,
'www.benfrederickson.com': 0.8828991573081899,
'maps.stamen.com': 0.8828934647050843,
'techcommunity.microsoft.com': 0.8828546011216626,
'blog.pyston.org': 0.8828536336243845,
'blog.freshdesk.com': 0.8828536336243845,
'www.radionz.co.nz': 0.8828494689784901,
'translate.googleusercontent.com': 0.882847058520687,
'stephenmann.io': 0.8828104499622798,
'www.berkeleyside.com': 0.8828101152681945,
'engineering.zenpayroll.com': 0.8827759389076459,
'googlefiberblog.blogspot.com': 0.8827390835242562,
'nadiaeghbal.com': 0.8827255975365081,
'chartio.com': 0.882704073135325,
'slidetocode.com': 0.8826822967578939,
'mic.com': 0.8826760832749148,
'ayende.com': 0.8826726588726063,
'sixrevisions.com': 0.8826715498136648,
'www.ligo.caltech.edu': 0.8826619298426813,
'www.insider.com': 0.8826223554956923,
'grahamc.com': 0.8825699143811585,
'www.sbcl.org': 0.882563699339324,
'www-m.cnn.com': 0.8825634181290516,
'advances.sciencemag.org': 0.8825504380314206,
'codingvc.com': 0.8825483021452769,
'blog.buoyant.io': 0.8825343736402197,
'tools.google.com': 0.8825286182855726,
'www.thisamericanlife.org': 0.8825206320161334,
'www.livescience.com': 0.8825187990685426,
'blog.udacity.com': 0.88248770629827,
'www.mint.com': 0.8824829228307717,
'www.quantstart.com': 0.8824744656067158,
'i3wm.org': 0.882461673497088,
'lugassy.net': 0.882461673497088,
'renato.athaydes.com': 0.8824351262415586,
'john.ankarstrom.se': 0.8824318706073884,
'deis.com': 0.882398256845465,
'blog.travis-ci.com': 0.8823494388558837,
'akshat.posterous.com': 0.882348785313154,
'blog.danieljanus.pl': 0.882305963719186,
'www.lambdassociates.org': 0.882291549111701,
'www.paypal-engineering.com': 0.8822900726859743,
'macdaddy.io': 0.8822875933955345,
'homepages.inf.ed.ac.uk': 0.882263750625994,
'events.google.com': 0.8822491129403885,
'isocpp.org': 0.882201264152672,
'www.secjuice.com': 0.8821793252738344,
'www.martinvigo.com': 0.8821771335964247,
'osblog.stephenmarz.com': 0.8821656713562592,
'business.financialpost.com': 0.8821599714093322,
'thetech.com': 0.8821385176184691,
'berkshirehathaway.com': 0.882138517618469,
'harvardlawreview.org': 0.8821355848021567,
'leafo.net': 0.8821225145186178,
'curia.europa.eu': 0.882094739471072,
'ckeditor.com': 0.8820783917249119,
'www.topgear.com': 0.8820554259493689,
'thebaffler.com': 0.8820445386842816,
'wilwheaton.net': 0.8820435463334623,
'andyadams.org': 0.8820328528134829,
'applidium.com': 0.8820326538487594,
'techspecs.blog': 0.8819755098672539,
'cafbit.com': 0.8819755098672539,
'community.spotify.com': 0.8819640976966642,
'jacobinmag.com': 0.8819625349712696,
'www.indiegogo.com': 0.881946367637731,
'hueniverse.com': 0.8819393022613744,
'tent.io': 0.8819180061465067,
'blog.caranddriver.com': 0.8818992762989476,
'www.avweb.com': 0.8818262516609122,
'kieranhealy.org': 0.8817976429098323,
'webrtchacks.com': 0.8817875203232487,
'www.penzba.co.uk': 0.8817713693955629,
'www.onbeing.org': 0.8817382469287067,
'www.makeartwithpython.com': 0.8817227794803292,
'storagemojo.com': 0.8816983007282381,
'rentzsch.tumblr.com': 0.8816906014355783,
'www.chrisnorstrom.com': 0.881678989119816,
'mir.aculo.us': 0.8816552237217568,
'automatetheboringstuff.com': 0.8815884675997907,
'doc.cat-v.org': 0.8815699328665041,
'skilldrick.github.io': 0.8815628815628815,
'www.azulsystems.com': 0.8815545999084712,
'googledevelopers.blogspot.com': 0.8815293986804532,
'shopify.engineering': 0.8815078480198789,
'www.windy.com': 0.8814945016309822,
'gen.medium.com': 0.8814868144618807,
'mygengo.com': 0.8814826810568291,
'microship.com': 0.8814701304564254,
'jroweboy.github.io': 0.881441670329882,
'www.isro.gov.in': 0.8814323797659261,
'amitness.com': 0.8814233401080115,
'journal.neilgaiman.com': 0.8813000878314081,
'blog.longnow.org': 0.8812928039894358,
'developers.lyst.com': 0.881286326309607,
'threads2.scripting.com': 0.8812146696862522,
'www.intego.com': 0.8812141651552359,
'abc7news.com': 0.8812015459094202,
'ac.els-cdn.com': 0.8811985270068678,
'zwischenzugs.wordpress.com': 0.8811838816065725,
'ooni.torproject.org': 0.8811640746997138,
'rodneybrooks.com': 0.8811464380886621,
'mentalfloss.com': 0.8811323569439224,
'www.elie.net': 0.8810667382996152,
'www.eecs.berkeley.edu': 0.881035190029909,
'surjan.substack.com': 0.8810020145474288,
'metrics.torproject.org': 0.880999086515737,
'www.scss.tcd.ie': 0.8809783370810794,
'www.zdziarski.com': 0.8809560496711293,
'blog.lizzie.io': 0.8809270824147091,
'www.nntp.perl.org': 0.8809140700339988,
'www.skorks.com': 0.8809070694481402,
'www.lastweekinaws.com': 0.880898941183668,
'projectgus.com': 0.8808911111787601,
'togelius.blogspot.com': 0.8808652592717593,
'www.thedailymuse.com': 0.880861164752333,
'blog.mailchimp.com': 0.8808342908333803,
'www.seas.harvard.edu': 0.8808287698175389,
'kernelmag.dailydot.com': 0.8807834100859487,
'www.audubon.org': 0.8807603148427023,
'juretriglav.si': 0.8807541586341653,
'www.ias.edu': 0.8806635755411035,
'www.startupsopensourced.com': 0.8806386086637549,
'scott.a16z.com': 0.880594330778045,
'www.webdistortion.com': 0.8805671581809865,
'coinbase.com': 0.8805536570003393,
'my.matterport.com': 0.8805531859476551,
'world.hey.com': 0.8805499381704851,
'freenetproject.org': 0.8805418719211823,
'ez.substack.com': 0.8805347456971939,
'www.cryptologie.net': 0.880520955636949,
'www.ewherry.com': 0.8804202710724481,
'www.aaronklein.com': 0.8804202710724481,
'www.metachris.com': 0.8803487337679655,
'www.info.ucl.ac.be': 0.8803332113342701,
'www.maritime-executive.com': 0.8802634318319039,
'nationalpost.com': 0.8802495762864062,
'stu.mp': 0.8802212225414684,
'pcwalton.github.io': 0.8801825251719878,
'www.lensrentals.com': 0.8801530918564299,
'programmingzen.com': 0.8801076222388025,
'applied-data.science': 0.8800802934751344,
'www.2uo.de': 0.8800661016526783,
'www.americanscientist.org': 0.8800620422490765,
'vault.fbi.gov': 0.8800567908408685,
'www.mymodernmet.com': 0.880048242675238,
'www.stripes.com': 0.8799620726032865,
'aralbalkan.com': 0.8799619967876905,
'www.washingtonian.com': 0.8799468909598039,
'jamey.thesharps.us': 0.8799345914752841,
'visualwebsiteoptimizer.com': 0.8799281242616495,
'blogs.uw.edu': 0.879925783709813,
'blog.sqlizer.io': 0.8799186267288787,
'workshop.chromeexperiments.com': 0.8799186267288787,
'www.jamieweb.net': 0.8799112718655394,
'royal.pingdom.com': 0.8798976823547263,
'oag.ca.gov': 0.8798952586106802,
'www.askamanager.org': 0.8798868933154113,
'briarproject.org': 0.8798570681672088,
'everythingsysadmin.com': 0.8798435781150284,
'serokell.io': 0.8798171504945399,
'mmcgrana.github.com': 0.8797647395096098,
'ruslanspivak.com': 0.8797628893146511,
'wesmckinney.com': 0.8797444121875875,
'blog.mattbierner.com': 0.8797205044550149,
'mit.edu': 0.8796628612944964,
'www.asimovinstitute.org': 0.8796512842876961,
'press.stripe.com': 0.8796512842876961,
'technicshistory.wordpress.com': 0.8796210680092151,
'common-lisp.net': 0.8796152653799312,
'www.u-tokyo.ac.jp': 0.8796017299329183,
'www.dailykos.com': 0.8795999531727511,
'www.pagerduty.com': 0.8795889070772409,
'austinkleon.com': 0.8795887664062606,
'blog.last.fm': 0.8795848845621085,
'blog.clearbit.com': 0.8795162875960222,
'blog.thinkst.com': 0.8794805645736704,
'hivelogic.com': 0.87947663783247,
'hechingerreport.org': 0.8794286991256227,
'www.fortressofdoors.com': 0.8794194991885873,
'docs.docker.com': 0.8794114856568306,
'kushagragour.in': 0.879405287790302,
'ultimatehackingkeyboard.com': 0.8793982349555475,
'developer.ibm.com': 0.8793941898423207,
'rushter.com': 0.879388928994475,
'www.tarsnap.com': 0.8793327066997211,
'www.wtsp.com': 0.8793263869529092,
'brohrer.github.io': 0.8793263869529092,
'elementary.io': 0.8793165731880934,
'destevez.net': 0.8793048124903404,
'www.gentoo.org': 0.879269905778943,
'nbviewer.jupyter.org': 0.879269721586867,
'blog.angular.io': 0.879268923349318,
'www.techrepublic.com': 0.8792309439284567,
'www.virgin.com': 0.8792219903286642,
'blog.getspool.com': 0.8792029887920298,
'www.lucidchart.com': 0.8791838979662947,
'blog.readability.com': 0.8791595821198683,
'narrative.ly': 0.8791473124131408,
'davidgerard.co.uk': 0.879133623229212,
'www.tldp.org': 0.8791060662671288,
'www.chillingeffects.org': 0.8790911678663772,
'mattbruenig.com': 0.8790858191942302,
'blog.coursera.org': 0.8790756346109743,
'engineering.khanacademy.org': 0.8790360280222413,
'aurellem.org': 0.8790251342720007,
'www8.hp.com': 0.8789932584616722,
'www.mrmoneymustache.com': 0.8789801725348548,
'themetricsystem.rjmetrics.com': 0.8789799435157434,
'popey.com': 0.8789713929868309,
'xlinux.nist.gov': 0.8789408088929164,
'ben-evans.com': 0.8789322960390088,
'www.datapointed.net': 0.8789114224720275,
'www.net.in.tum.de': 0.8788892296044377,
'stratos.seas.harvard.edu': 0.8788892296044377,
'www.smbc-comics.com': 0.878826000548314,
'blog.gittip.com': 0.8788213217536158,
'www.cloudflare.com': 0.8788059374287456,
'rmurphey.com': 0.8787968488987687,
'www.tuhs.org': 0.8787968488987687,
'blog.openstreetmap.org': 0.8787897231673689,
'www.istartedsomething.com': 0.8787889297497098,
'ilyabirman.net': 0.8787732444342291,
'www.github.com': 0.8787717233938399,
'amix.dk': 0.8787529849051614,
'www.insidehighered.com': 0.8787455322695712,
'actu.epfl.ch': 0.8787380577275826,
'www.buyforlife.com': 0.8787356591099948,
'staysaasy.com': 0.8786604306758078,
'www.okayhq.com': 0.8786525461110833,
'engineering.mixpanel.com': 0.878632488020514,
'www.netresec.com': 0.8786288106631085,
'jott.live': 0.8786251117226983,
'www.health.harvard.edu': 0.8786179702092969,
'plot.ly': 0.8785945787217043,
'www.edwardtufte.com': 0.8785389757836423,
'pzemtsov.github.io': 0.8785192835169581,
'grack.com': 0.878506371619462,
'www.aeonmagazine.com': 0.878499144479296,
'lowlevelbits.org': 0.8784722367844,
'blog.dotcloud.com': 0.8784660513882162,
'robwalling.com': 0.8784461152882206,
'luajit.org': 0.8784380131632007,
'mccormick.cx': 0.8784273467707013,
'www.usehaystack.io': 0.8784273467707013,
'stedolan.github.io': 0.8784164208081071,
'supermemo.guru': 0.8784156694332751,
'jackschaedler.github.io': 0.8783768191039716,
'mgba.io': 0.87835268885913,
'forum.nim-lang.org': 0.8783083943148878,
'sympa.inria.fr': 0.8783048144562394,
'lists.nongnu.org': 0.8783041478638947,
'lispcookbook.github.io': 0.8782936010037641,
'ds9a.nl': 0.8782772345442594,
'thinkpol.ca': 0.8782726019439232,
'itunes.apple.com': 0.8782672365580176,
'www.coursera.org': 0.8782670165903833,
'www.nrc.nl': 0.8782340201533003,
'www.faa.gov': 0.8781736885382665,
'artsandculture.google.com': 0.8781425787274354,
'smerity.com': 0.8781378214625035,
'www.omaha.com': 0.8781181772597978,
'www.wweek.com': 0.8781063159893508,
'www.stroustrup.com': 0.8780938693821273,
'sacks.substack.com': 0.8780908854852072,
'www.heraldsun.com.au': 0.8780821695385574,
'wiki.alopex.li': 0.8780784699174555,
'www.dartlang.org': 0.8780551661146612,
'bugs.php.net': 0.878032279338754,
'www.cyclingnews.com': 0.8779874213836478,
'www.minix3.org': 0.8779874213836478,
'www.cryptomuseum.com': 0.8779834887038765,
'blog.sendhub.com': 0.8779734555774865,
'web.williams.edu': 0.8779619274642066,
'sourceware.org': 0.8779595104655152,
'blog.revolutionanalytics.com': 0.8779228861519864,
'blog.feabhas.com': 0.877905135054773,
'rubenerd.com': 0.8778819789027571,
'longnow.org': 0.8778804834133449,
'www.kevinslin.com': 0.8778337531486146,
'testing.googleblog.com': 0.8778159226448956,
'merbist.com': 0.8778006932561861,
'academic.oup.com': 0.8777955202058847,
'www.romancortes.com': 0.8777775637794132,
'ottverse.com': 0.8777664608276228,
'panopticlick.eff.org': 0.8777635360894769,
'www.presslabs.com': 0.8777093280795232,
'www.archaeology.org': 0.8777019277787409,
'www.qt.io': 0.8776867090078414,
'www.rust-lang.org': 0.8776640621382866,
'komonews.com': 0.8776519468194085,
'petewarden.typepad.com': 0.877650269187265,
'redecentralize.org': 0.8776475140888633,
'spencerdailey.com': 0.8776383740955995,
'ryanbigg.com': 0.8776375837266788,
'codeascraft.com': 0.8776350192149127,
'davidsimon.com': 0.8776245936362499,
'nodered.org': 0.8775609145071156,
'releases.llvm.org': 0.8775465677232488,
'blog.npmjs.org': 0.8775321171093747,
'nothings.org': 0.8775219068062873,
'www.minecraft.net': 0.8775149658488136,
'www.nber.org': 0.8774912434535365,
'docs.keydb.dev': 0.8774559444971969,
's3.amazonaws.com': 0.8774127148800118,
'www.startuplessonslearned.com': 0.8773991243759531,
'lostinjit.blogspot.com': 0.8773967208056322,
'unity3d.com': 0.8773730560374686,
'rtpg.co': 0.8773704171934261,
'meteuphoric.wordpress.com': 0.8773701347020316,
'c9.io': 0.8772877729098074,
'davidad.github.io': 0.8772551751572703,
'www.cs.tufts.edu': 0.8772142384084949,
'patents.google.com': 0.8771976309077076,
'carlschwan.eu': 0.8771903919813752,
'www.disneyresearch.com': 0.8771847497268329,
'kilianvalkhof.com': 0.8771650377417883,
'easylang.online': 0.8771572953832314,
'simonschreibt.de': 0.87714784842864,
'answers.onstartups.com': 0.8771043002987845,
'casetext.com': 0.8770972548772688,
'thekidshouldseethis.com': 0.8770970221777706,
'blog.bittorrent.com': 0.8770863340546824,
'www.phrack.org': 0.8770711101073438,
'blog.deepgram.com': 0.8770646346636887,
'grooveshark.com': 0.8770646346636887,
'www.stat.cmu.edu': 0.8770412624073355,
'matthewrayfield.com': 0.8770299470017662,
'blog.merovius.de': 0.8769629784425906,
'jordan-wright.github.io': 0.8769261031957445,
'www.michael-noll.com': 0.876870919176529,
'www.detroitnews.com': 0.876865566511843,
'www.chromestatus.com': 0.8768237137831741,
'spacy.io': 0.8767936477263165,
'blog.theoldreader.com': 0.8767904369417293,
'larrysummers.com': 0.8767904369417293,
'devo.ps': 0.8767685151051207,
'jackg.org': 0.8767448498103232,
'www.leancrew.com': 0.8767166755854882,
'blog.smartdec.net': 0.876696975867677,
'aaronparecki.com': 0.8766851750370125,
'eandt.theiet.org': 0.8766793394897466,
'www.lennysnewsletter.com': 0.8766464219782513,
'blog.factual.com': 0.8766406692255209,
'archlinux.me': 0.8765956164893757,
'cglab.ca': 0.8765746912638428,
'meteor.com': 0.8765739666365363,
'blog.jpalardy.com': 0.8765005864209046,
'amitkapila16.blogspot.com': 0.8764939037562264,
'en.yna.co.kr': 0.8764898808598593,
'linux.slashdot.org': 0.8764680273690748,
'www.trevormckendrick.com': 0.8764571075800416,
'cmcenroe.me': 0.8764460213566705,
'robertovitillo.com': 0.8764453376914981,
'metacademy.org': 0.8764331210191083,
'lethain.com': 0.8764062511281379,
'www.news.com.au': 0.876395323072595,
'www.severe-weather.eu': 0.8763844603500262,
'www.federalreserve.gov': 0.8763690497257808,
'passcode.csmonitor.com': 0.8763573796830383,
'chrismorgan.info': 0.8763466464048606,
'www.kurzweilai.net': 0.8763277380739884,
'www.federalregister.gov': 0.8763029852387336,
'carolchen.me': 0.8762960630948078,
'scholar.princeton.edu': 0.8762929484619637,
'limi.net': 0.8762759528715298,
'sijmen.ruwhof.net': 0.8761846124088934,
'winaero.com': 0.8761846124088934,
'www.techquila.co.in': 0.8761680078483304,
'www.theamericanconservative.com': 0.876167214287758,
'faculty.washington.edu': 0.8761620699999401,
'marijnhaverbeke.nl': 0.876139575410831,
'swizec.com': 0.8761258858184888,
'www.thebestpageintheuniverse.net': 0.876117496807152,
'www.epsilontheory.com': 0.8761124613754698,
'www.exurbe.com': 0.8761030084533606,
'www.motortrend.com': 0.8760487843348314,
'blog.seattlepi.com': 0.8759960447121663,
'www.enlightenment.org': 0.8759950518819543,
'www.seriouseats.com': 0.8759880790728855,
'appleinsider.com': 0.8759788729692606,
'stilldrinking.org': 0.8759642287018524,
'www.holistics.io': 0.8759580367004831,
'www.dannyguo.com': 0.8759212674654512,
'letsworkshop.com': 0.8759027598733787,
'www.androidauthority.com': 0.8758978540777702,
'visalist.io': 0.8758718179278925,
'www.makelinux.net': 0.8758439962639901,
'avikdas.com': 0.8758367116558062,
'www.linkedin.com': 0.8757903442097913,
'wiki.freebsd.org': 0.8757706173362011,
'www.tandfonline.com': 0.8757698347226249,
'www.pouet.net': 0.8757326866058377,
'blocksandfiles.com': 0.8756986319076105,
'www.docracy.com': 0.8756913215448502,
'www.useit.com': 0.8756472373396756,
'www.sapiens.org': 0.8756019742944366,
'www.ccn.com': 0.8755951918062195,
'www.mattblodgett.com': 0.8755948781981208,
'www.france24.com': 0.8755846335815239,
'www.zacharyburt.com': 0.8755387620665042,
'www.congress.gov': 0.8755256407121746,
'fab.cba.mit.edu': 0.8755115130516663,
'newsroom.mastercard.com': 0.8754825096495608,
'youexec.com': 0.8754809119780987,
'pragmaticpineapple.com': 0.8754604704436907,
'ruder.io': 0.8754604704436906,
'www.cpomagazine.com': 0.8754507502798615,
'fontsinuse.com': 0.8754438392621019,
'www.uscis.gov': 0.8754248448695789,
'youtube.com': 0.8754200542585205,
'movio.co': 0.8754182523171723,
'www.justwatch.com': 0.8754182523171723,
'www.seeedstudio.com': 0.8754129248508049,
'electrospaces.blogspot.com': 0.8754013788662113,
'journal.dedasys.com': 0.8753840466834827,
'css-tricks.com': 0.8753766177219949,
'googleenterprise.blogspot.com': 0.8753419411917504,
'radian.org': 0.8753349762136738,
'futurism.com': 0.8752759686541421,
'brandon.invergo.net': 0.875259685377418,
'blogs.bmj.com': 0.8752587898620211,
'cs.stanford.edu': 0.8752572315765397,
'www.law.nyu.edu': 0.8752557472640224,
'madebymike.com.au': 0.8752557472640223,
'www.seomoz.org': 0.8752420123288777,
'blog.invisiblethings.org': 0.8752416201881567,
'www.greghendershott.com': 0.8752189373052796,
'frankchimero.com': 0.8752104961969256,
'www.dodgycoder.net': 0.8752104961969256,
'txt.fyi': 0.8751961676198574,
'www.cisa.gov': 0.8751827659203362,
'www.failory.com': 0.8751736016299931,
'jamanetwork.com': 0.8751682514227755,
'www.objc.io': 0.8751399221981017,
'duo.com': 0.875136824237271,
'gs.statcounter.com': 0.8751360927111078,
'futurice.com': 0.8751180437027989,
'sifted.eu': 0.8751010675262672,
'laravel-news.com': 0.8750758486776293,
'archagon.net': 0.8750478337445297,
'tinyletter.com': 0.8750075762218684,
'blog.scoutapp.com': 0.875005539344818,
'tomorrowcorporation.com': 0.8750053569816645,
'blogs.opera.com': 0.8749949083342682,
'm.sfgate.com': 0.8749578015755867,
'blog.scottlogic.com': 0.8749532936904543,
'www.vincit.fi': 0.8749482978568669,
'www.hiringlab.org': 0.8749298773541315,
'adioso.com': 0.8749298773541315,
'technology.blog.gov.uk': 0.8748916088022032,
'intelligence.org': 0.874874784745018,
'musicmachinery.com': 0.874874767784566,
'www.spaceref.com': 0.8748480825746138,
'raisingfutureadults.blogspot.com': 0.8748200301964755,
'fantasticanachronism.com': 0.8747762139154207,
'calculist.org': 0.8747762139154207,
'lizthedeveloper.com': 0.8747656784427211,
'elliotjaystocks.com': 0.8747563598722315,
'www.julian.com': 0.8747501335716515,
'garbagecollected.org': 0.8747266158827341,
'timdaub.github.io': 0.8746432384113364,
'www.good.is': 0.8745679480603733,
'www.elidedbranches.com': 0.8745591524175645,
'people.gnome.org': 0.8745420458214501,
'goteleport.com': 0.8745075530736341,
'rome.ro': 0.8744782906989623,
'molily.de': 0.8744677514854211,
'www.douglas-self.com': 0.8744450764890996,
'www.rubymotion.com': 0.874400769900326,
'www.shubhro.com': 0.8743836350202087,
'blog.jim-nielsen.com': 0.8743167857862288,
'www.deseret.com': 0.8743016225209655,
'www.dallasnews.com': 0.8742922264773864,
'purecss.io': 0.874271743877152,
'lowercasecapital.com': 0.874271743877152,
'blog.getpaint.net': 0.874271743877152,
'google.com': 0.8742545289746798,
'betabeat.com': 0.874238729033988,
'dawn.cs.stanford.edu': 0.8742346653711291,
'futureofcoding.org': 0.8741693203974006,
'www.ocregister.com': 0.87415339402842,
'materialize.com': 0.8741062320515165,
'teddziuba.com': 0.8740975697255843,
'www.thelocal.de': 0.8740894106328552,
'webaim.org': 0.8740351003371842,
'www.firstthings.com': 0.8739883726292156,
'www.yodaiken.com': 0.8739612387765462,
'www.goodcheapandfast.com': 0.87394184329453,
'fastcompression.blogspot.com': 0.8739308258375799,
'perl6advent.wordpress.com': 0.8739142963181532,
'www.shirky.com': 0.873897843474887,
'www.baltimoresun.com': 0.873865725506546,
'www.greaterwrong.com': 0.8738154479524215,
'bastian.rieck.me': 0.8738148485847019,
'toastytech.com': 0.8738123717532574,
'raw.githubusercontent.com': 0.8738090863060529,
'www.wealthsimple.com': 0.8737884336019249,
'sanfrancisco.cbslocal.com': 0.8737831715062644,
'blog.palantir.com': 0.8736979166666666,
'learngitbranching.js.org': 0.8736979166666666,
'arr.am': 0.8736979166666666,
'i.blackhat.com': 0.8736899056814162,
'lolware.net': 0.873689463390705,
'sherylcanter.com': 0.8736870447303569,
'andrewgelman.com': 0.8736823027610198,
'www.stat.columbia.edu': 0.8736693395358077,
'pouchdb.com': 0.8736594278511989,
'www.compose.io': 0.8736483544069877,
'mleverything.substack.com': 0.8736159033763847,
'work.qz.com': 0.8736112922383825,
'lucasmreis.github.io': 0.8736055963750089,
'news.itsfoss.com': 0.8736019609008807,
'wiki.nikitavoloboev.xyz': 0.8735941094898686,
'developer.nvidia.com': 0.873569853920636,
'antigonejournal.com': 0.8734640566717642,
'spectator.us': 0.8734631581578451,
'www.businesswire.com': 0.8734611836531004,
'stressgrid.com': 0.87343280677026,
'pgbovine.net': 0.8734068690708945,
'forums.tigsource.com': 0.8733979453959029,
'thenewstack.io': 0.8733971641405377,
'blog.adamsmith.cc': 0.8733885158313304,
'www.gallup.com': 0.8733177322868378,
'www.institutionalinvestor.com': 0.873174323391906,
'allendowney.blogspot.com': 0.8731681454070807,
'www.lowrisc.org': 0.873141850865328,
'jshakespeare.com': 0.8731286306597288,
'www.nasaspaceflight.com': 0.8731126994667253,
'divan.github.io': 0.873106828361247,
'emacsconf.org': 0.873106828361247,
'www.scottporad.com': 0.8731061096250692,
'cam.ly': 0.8731039371172824,
'en.wikibooks.org': 0.8730999086484691,
'peak5390.wordpress.com': 0.8730838690655434,
'www.boredpanda.com': 0.8730796077702506,
'adam.heroku.com': 0.8730645256354151,
'transportevolved.com': 0.8730606539065758,
'jamesfriend.com.au': 0.8730468292001098,
'sethrobertson.github.io': 0.8730063430759272,
'kennethreitz.com': 0.872966789255468,
'nigeltao.github.io': 0.8728857599922942,
'www.damtp.cam.ac.uk': 0.8728857599922942,
'blog.chewxy.com': 0.872876574828518,
'savannah.gnu.org': 0.8728754513879475,
'www.sfexaminer.com': 0.872849372895737,
'engineering.pinterest.com': 0.8727923546520051,
'hitchdev.com': 0.8727791664778902,
'www.airlineratings.com': 0.8727662828334155,
'dt.deviantart.com': 0.8727436710427438,
'blog.dustincurtis.com': 0.8727333641616176,
'andymatuschak.org': 0.8727040209571875,
'ccc.de': 0.8727034120734908,
'www.cs.bham.ac.uk': 0.8726465691109567,
'www.robertputt.co.uk': 0.8726417417469949,
'www.miamiherald.com': 0.8725798697059735,
'blog.fitbit.com': 0.8725623923381794,
'adrianchadd.blogspot.com': 0.8725623923381794,
'developer.amazon.com': 0.8725538657953651,
'blog.codeship.com': 0.8725534393008655,
'www.attn.com': 0.8725475818041069,
'paleotronic.com': 0.8725451217616845,
'www.cdixon.org': 0.8725371731911513,
'cuddly-octo-palm-tree.com': 0.8725361366622865,
'www.flownet.com': 0.8725345822755496,
'www.sacrideo.us': 0.8725261893403989,
'newyorker.com': 0.8725219622684063,
'marmelab.com': 0.8725088072483861,
'help.twitter.com': 0.8724643326336655,
'browser.geekbench.com': 0.8724643326336655,
'blairreeves.me': 0.8724601687001106,
'thesquareplanet.com': 0.872441172969379,
'thesephist.com': 0.8724167847627841,
'www.freenas.org': 0.8723746839925903,
'kwokchain.com': 0.872332880121703,
'bostinnovation.com': 0.8723328354214289,
'love2d.org': 0.8723183994033862,
'en.people.cn': 0.872305563134482,
'blog.sucuri.net': 0.8722868151723697,
'tink.uk': 0.87225641869954,
'developer.palm.com': 0.8722310220579464,
'inessential.com': 0.8722306989053726,
'latimesblogs.latimes.com': 0.8722297510115247,
'startuplessonslearned.blogspot.com': 0.8721812850079301,
'blog.qt.digia.com': 0.8721548264341702,
'cssdeck.com': 0.8721274930836064,
'medium.freecodecamp.org': 0.8720631693223321,
'www.philly.com': 0.8720354776247603,
'nat.org': 0.8719937064288694,
'www.architecturaldigest.com': 0.871986927054336,
'adactio.com': 0.8719856028794241,
'anvil.works': 0.8719506746013774,
'accidentallyquadratic.tumblr.com': 0.8719251801589324,
'arduino.cc': 0.8718637797679766,
'blog.embed.ly': 0.8718082655702977,
'anyvite.com': 0.8718025672263069,
'www.eater.com': 0.871748228284988,
'moxie.org': 0.8717373817367493,
'www.jacquescorbytuech.com': 0.8717101335367482,
'n8n.io': 0.8716898518049913,
'commonsware.com': 0.871674162722329,
'www.snoyman.com': 0.8716637912464283,
'blog.patricktriest.com': 0.87161890252714,
'leaf.dragonflybsd.org': 0.8716189025271399,
'www.gamedeveloper.com': 0.8716189025271399,
'www.decisionsciencenews.com': 0.8716189025271399,
'primitivetechnology.wordpress.com': 0.8716187818992139,
'www.marinetraffic.com': 0.8716117229392578,
'www.dallasobserver.com': 0.8716085858849908,
'www.consumeraffairs.com': 0.871556493468071,
'english.aljazeera.net': 0.8715487596633263,
'eng.paxos.com': 0.871523178807947,
'scobleizer.com': 0.8715026146236272,
'endler.dev': 0.871459748976304,
'www.assemblyai.com': 0.871459748976304,
'spaceweatherarchive.com': 0.8714580347584702,
'www.todayifoundout.com': 0.871448243747315,
'www.itprotoday.com': 0.8714363693988826,
'physics.stackexchange.com': 0.8714320610312986,
'www.lifespan.io': 0.8714320342847328,
'pspdfkit.com': 0.8714289642359972,
'www.slackware.com': 0.8714078524585236,
'economix.blogs.nytimes.com': 0.8713770585361653,
'imba.io': 0.8713614361461072,
'blog.wizsec.jp': 0.8713527851458885,
'neal.fun': 0.8713508609752882,
'shift.newco.co': 0.8713309282655746,
'm50d.github.io': 0.8712974347864724,
'c.learncodethehardway.org': 0.8712925320326951,
'hosted.ap.org': 0.8712897256940902,
'www.onebigfluke.com': 0.8712867098269154,
'getmagicnow.com': 0.8712820888223272,
'projects.thestar.com': 0.8712523001988615,
'www.ralfj.de': 0.8712321679229643,
'www.vantage.sh': 0.871207076528529,
'www.linuxfoundation.org': 0.8711369679250435,
'openclassroom.stanford.edu': 0.8711347101538987,
'usefyi.com': 0.8711273706889674,
'www.bingocardcreator.com': 0.8711151425360408,
'blog.robenkleene.com': 0.8711151425360408,
'mako.cc': 0.8711144740115356,
'blog.frontapp.com': 0.8710729457593172,
'virtuallyfun.com': 0.8710620996646494,
'www.csmonitor.com': 0.8710243224833312,
'www.nintendo.co.uk': 0.8710106382978724,
'codon.com': 0.8709949725685586,
'www.sixthtone.com': 0.870972915708058,
'statweb.stanford.edu': 0.8709617791625288,
'www.selectedintelligence.com': 0.8709463428439262,
'www.elasticsearch.org': 0.8709352679197123,
'math.mit.edu': 0.870930298916698,
'dwheeler.com': 0.870930298916698,
'kim.com': 0.8709205575405985,
'www.nbcsandiego.com': 0.8708959915522153,
'paragonie.com': 0.8708785801404334,
'bostonherald.com': 0.8708776881584194,
'mathwithbaddrawings.com': 0.8708565076256994,
'www.openshot.org': 0.8708388814913449,
'bugs.python.org': 0.8708284581145338,
'blog.aerofs.com': 0.8708080233174353,
'www.chicagomag.com': 0.8707562016036294,
'www.250bpm.com': 0.8707223328636163,
'datagenetics.com': 0.8706854622162113,
'blogs.embarcadero.com': 0.870674156132316,
'www.ispreview.co.uk': 0.8706343121704596,
'www.twistlock.com': 0.8706274964354553,
'www.lukew.com': 0.8705263186498488,
'news.efinancialcareers.com': 0.8705226226571788,
'www.travelandleisure.com': 0.8705014801192988,
'blog.okturtles.com': 0.8704939919893191,
'ilsr.org': 0.8704939919893191,
'www.usb.org': 0.8704796759014012,
'dish.andrewsullivan.com': 0.8704773071162746,
'likewise.am': 0.8704734916123844,
'blog.idonethis.com': 0.8704583890982127,
'blog.talosintelligence.com': 0.8704526519003233,
'googlesystem.blogspot.com': 0.8704373858850157,
'garrysub.posterous.com': 0.8704372803921456,
'insinuator.net': 0.8704372803921456,
'hypothes.is': 0.8704167324964841,
'cdt.org': 0.8704025091193442,
'apod.nasa.gov': 0.8703900488888665,
'herman.bearblog.dev': 0.8703870042603329,
'archiveteam.org': 0.8703718279113771,
'seths.blog': 0.8703521175616137,
'blog.internetcases.com': 0.8703208556149733,
'olifro.st': 0.870316353450948,
'www.commerce.gov': 0.8703148755143479,
'cosmosmagazine.com': 0.8703108288261813,
'onemileatatime.com': 0.8703062566192509,
'blog.cardiogr.am': 0.8702780879767176,
'www.nsf.gov': 0.8702217832404014,
'michaelfeathers.silvrback.com': 0.8702081612825068,
'cse.buffalo.edu': 0.8701754328744842,
'www.cardiff.ac.uk': 0.8701754328744842,
'stuartkhall.com': 0.8701745691078596,
'www.ablogtowatch.com': 0.8701496552122142,
'kripken.github.io': 0.8701188385936537,
'www.submarinecablemap.com': 0.8700961746199228,
'blog.pandora.com': 0.8700581031783425,
'undark.org': 0.8700047089370274,
'indiestatik.com': 0.8699989528159702,
'www.ldeming.com': 0.8699989528159702,
'www.shadertoy.com': 0.8699979474186827,
'blog.forecast.io': 0.8699840133880231,
'www.thumbtack.com': 0.8699483761483322,
'neopythonic.blogspot.com': 0.8699126451206446,
'blog.yoshuawuyts.com': 0.8699103901809758,
'meaningness.com': 0.8698796561423923,
'www.climatecentral.org': 0.8698732077608144,
'phasenoise.livejournal.com': 0.8698692075920079,
'solarianprogrammer.com': 0.8698554554648813,
'www.allenpike.com': 0.869850220276853,
'prometheus.io': 0.8698088579434151,
'www.zocalopublicsquare.org': 0.8698034656609108,
'jsonresume.org': 0.8697986577181208,
'ts.data61.csiro.au': 0.8697986577181208,
'downloadsquad.switched.com': 0.8697589511534172,
'chris-said.io': 0.8697498026681764,
'www.eurekalert.org': 0.8697479182812706,
'rise.cs.berkeley.edu': 0.8697428562146504,
'nav.al': 0.8696282036759441,
'www.contextis.com': 0.8696220176972772,
'alpha.app.net': 0.8695846584298926,
'simon.shimmerproject.org': 0.8695798593001582,
'blog.andyet.com': 0.8695385830681973,
'harvardmagazine.com': 0.8695167508879433,
'montrealgazette.com': 0.8695147149611512,
'www.ossblog.org': 0.8694501584181511,
'www.snazz.xyz': 0.8694481830417228,
'securethoughts.com': 0.8694070114575367,
'www.dcrainmaker.com': 0.8694035765566716,
'static.googleusercontent.com': 0.869401063536612,
'www.cyberscoop.com': 0.8693847613582892,
'engineering.imvu.com': 0.8693718520967112,
'cs.nyu.edu': 0.869371790902293,
'github.hubspot.com': 0.8693590367674685,
'zenodo.org': 0.8693126875353983,
'www.nczonline.net': 0.8693117689767528,
'awealthofcommonsense.com': 0.8692734628001552,
'jcp.org': 0.8692722371967655,
'www.tabletmag.com': 0.869255146058989,
'www.spritzinc.com': 0.8692337048510355,
'www.pl-enthusiast.net': 0.8692303984396749,
'blog.rtwilson.com': 0.8692255308065057,
'parislemon.com': 0.8692052486630297,
'www.perimeterx.com': 0.869171185060124,
'blog.ndepend.com': 0.8691625738297151,
'www.gregreda.com': 0.8691455800391642,
'www.cooper.com': 0.8691426419439985,
'newsroom.ibm.com': 0.869132356539317,
'www.core77.com': 0.8691279702829294,
'paraschopra.com': 0.869122539211249,
'openspace.sfmoma.org': 0.8691093062023977,
'blog.stephenmarz.com': 0.8691082953129042,
'blog.workshape.io': 0.8690653365846199,
'ngrok.com': 0.8690653365846199,
'code.flickr.net': 0.8690543850923815,
'uk.pcmag.com': 0.8690031200561459,
'superpowered.com': 0.868989779225976,
'hunterwalk.medium.com': 0.8689812194478416,
'www.cs.tau.ac.il': 0.8689490020361462,
'littlemaninmyhead.wordpress.com': 0.8689216938948838,
'www.excamera.com': 0.8689189189189189,
'bjorn.tipling.com': 0.8689189189189189,
'people.idsia.ch': 0.8688594113666208,
'www.olark.com': 0.8688594113666208,
'blog.sigplan.org': 0.8688223104414097,
'www.volokh.com': 0.8687744918859344,
'betterprogramming.pub': 0.8687724684006338,
'www.consumerreports.org': 0.8687417291110767,
'spin.atomicobject.com': 0.8687241428724921,
'opencv.org': 0.8687019086527302,
'evincarofautumn.blogspot.com': 0.8687019086527302,
'blog.imgtec.com': 0.868681889755656,
'pauladamsmith.com': 0.86866369377526,
'neosmart.net': 0.8686249309907671,
'reprints.longform.org': 0.8686085856898127,
'www.xda-developers.com': 0.8685957498546732,
'reefpoints.dockyard.com': 0.8685213588812543,
'benguild.com': 0.8684664223004914,
'www.naughtycomputer.uk': 0.8684379703659447,
'www.kb.cert.org': 0.8684268393607657,
'www.insidescience.org': 0.868400521356853,
'motorola-blog.blogspot.com': 0.8683887736567877,
'op-co.de': 0.8683853459972863,
'pvk.ca': 0.8683840412350319,
'roadsandkingdoms.com': 0.8683501222568618,
'www.netbsd.org': 0.8682992634857859,
'lite.cnn.io': 0.8682859224905237,
'redeye.firstround.com': 0.8682663428825431,
'forum.xda-developers.com': 0.8682587280913919,
'posthog.com': 0.8682495044484383,
'ftp.iza.org': 0.868245075106375,
'craphound.com': 0.8682258194103606,
'leancrew.com': 0.8682221213291426,
'www.rijksmuseum.nl': 0.8682073993067194,
'blog.newsweek.com': 0.8681920308533846,
'www.hipmunk.com': 0.8681741501104386,
'fuchsia.dev': 0.8681336649460718,
'infrequently.org': 0.8680443914407565,
'www.nationalreview.com': 0.8680234764373578,
'www.futilitycloset.com': 0.8680090490995264,
'blogs.apache.org': 0.8679900476857304,
'www.hayabusa2.jaxa.jp': 0.8679812862496842,
'www.lenovo.com': 0.867894246896281,
'malisper.me': 0.8678702820090507,
'people.debian.org': 0.8678402113450426,
'blog.jwhitham.org': 0.8678305149745539,
'hai.stanford.edu': 0.8677844697267021,
'hubicka.blogspot.com': 0.8677805560584517,
'bemmu.posterous.com': 0.8677440846959065,
'www.gizmodo.com.au': 0.8677203766504066,
'meyerweb.com': 0.8676812093352608,
'www.novaspivack.com': 0.8676756246000905,
'www.theamericanscholar.org': 0.8676692275096932,
'www.pjrc.com': 0.8676602643652892,
'googleforwork.blogspot.com': 0.8676558691706335,
'get.slack.help': 0.8676558691706335,
'mwi.usma.edu': 0.8676236271344514,
'blogs.zoho.com': 0.8676226844533295,
'www.aaronstannard.com': 0.8676196165257838,
'www.messynessychic.com': 0.867572253713323,
'uk.businessinsider.com': 0.8675716785772413,
'www.defectivebydesign.org': 0.867569136323548,
'bash.org': 0.8675094991983033,
'venturehacks.com': 0.8675094413288004,
'www.thepublicdiscourse.com': 0.8674973517990539,
'browser.engineering': 0.8674925582906508,
'www.peachesnstink.com': 0.8674597307626667,
'www.ics.uci.edu': 0.8674482943120847,
'www.cs.utah.edu': 0.8674397196409499,
'www.inkstonenews.com': 0.8674036054993348,
'www.privacytools.io': 0.8673748578488399,
'brianknapp.me': 0.8673589537571962,
'www.mediaite.com': 0.867339960916398,
'www.theinquirer.net': 0.8673345006090067,
'www.dhruvonmath.com': 0.867331172642827,
'lucb1e.com': 0.8672987335530669,
'www.hetzner.de': 0.8672930834878059,
'blogs.nature.com': 0.8672878554675312,
'newsfeed.time.com': 0.8672878554675312,
'davemart.in': 0.8672750734639297,
'cryptopals.com': 0.8672750734639297,
'paperswithcode.com': 0.8672082397249838,
'blog.hipmunk.com': 0.8671864413349992,
'citizen428.net': 0.8671759938812859,
'shitpost.plover.com': 0.8671759938812859,
'developer.github.com': 0.8671757822917692,
'www.lispcast.com': 0.86717529476292,
'www.webofstories.com': 0.8671647243985804,
'blog.asana.com': 0.8671537327631247,
'vvvvalvalval.github.io': 0.8671420654436023,
'news.northeastern.edu': 0.8671175664071602,
'www.digitalmars.com': 0.8671085676412111,
'www.mysk.blog': 0.8671085676412111,
'www.janestreet.com': 0.8671085676412111,
'francescodilorenzo.com': 0.8671085676412111,
'www.binghamton.edu': 0.8671020117333935,
'measuringmeasures.blogspot.com': 0.8671015509192971,
'blog.businessofsoftware.org': 0.8670987836087559,
'www.dougengelbart.org': 0.8670867654048934,
'tkainrad.dev': 0.8670687156162842,
'www.machinelearningplus.com': 0.8670617301641067,
'200ok.ch': 0.8670503631928145,
'quizlet.com': 0.8669907032962705,
'www.singlelunch.com': 0.8669898915969994,
'chairnerd.seatgeek.com': 0.8669768798830123,
'www.govtech.com': 0.8669681804482694,
'news.mongabay.com': 0.8669281682982936,
'erikmcclure.com': 0.8669236451292578,
'aelkus.github.io': 0.8669193008503213,
'www.pyimagesearch.com': 0.8669169539583714,
'www.thejournal.ie': 0.8669063286580795,
'reaktor.com': 0.8669062156202597,
'christophermeiklejohn.com': 0.8668972800697652,
'globaleconomicanalysis.blogspot.com': 0.8668972800697652,
'reasonml.github.io': 0.8668408053269024,
'www.csiro.au': 0.8668352643092083,
'www.maxmasnick.com': 0.8668020425330096,
'www.bitmatica.com': 0.8667582417582418,
'bbs.archlinux.org': 0.8667582417582418,
'bookofhook.blogspot.com': 0.8667582417582418,
'rappers.mdaniels.com.s3-website-us-east-1.amazonaws.com': 0.8667582417582418,
'milapneupane.com.np': 0.8667582417582418,
'lists.randombit.net': 0.8667581999829783,
'tyrrrz.me': 0.8667581999829783,
'research.checkpoint.com': 0.8667536287753966,
'avdi.org': 0.8667515410500497,
'tuxphones.com': 0.8667177933546236,
'wolfstreet.com': 0.8667059036925511,
'www.danielmiessler.com': 0.8666932411921048,
'klarasystems.com': 0.8666904333167735,
'blog.leahhanson.us': 0.8666699206546021,
'lua-users.org': 0.8666542828163675,
'blog.tjll.net': 0.866647835447029,
'massivelyop.com': 0.8665522518442965,
'kuow.org': 0.8665345404342943,
'www.chinafile.com': 0.8665345404342943,
'www.allaboutcircuits.com': 0.8665332702488167,
'www.physorg.com': 0.8665324818274975,
'www.altdevblogaday.com': 0.8665315510980502,
'codeahoy.com': 0.8664780210212021,
'perl.plover.com': 0.8664701135603498,
'www.wpri.com': 0.8664576956212064,
'www.brianchristner.io': 0.8664469380333186,
'srconstantin.wordpress.com': 0.8664373069106099,
'yeoman.io': 0.8663968845512826,
'pixelastic.github.io': 0.8663968845512826,
'www.structuredprocrastination.com': 0.8663817900462257,
'www.galactanet.com': 0.8663440641046102,
'alexmuir.com': 0.8663020356625262,
'www.cs.washington.edu': 0.866214599696902,
'gapingvoid.com': 0.866214599696902,
'doisinkidney.com': 0.8662068965517241,
'jeffknupp.com': 0.866196580751142,
'weblogs.asp.net': 0.8661868876649387,
'www.eclipse.org': 0.8661815854172948,
'www.twitter.com': 0.8661688444094342,
'www.gmo.com': 0.8660899350618185,
'h14s.p5r.org': 0.8660899350618185,
'www.privatdozent.co': 0.8660524023972933,
'conorpp.com': 0.8660524023972933,
'dev-blog.apollodata.com': 0.8660261935749697,
'www.bof.nl': 0.8660220994475138,
'www.potaroo.net': 0.8660076718528887,
'www.disneyanimation.com': 0.8659912587210011,
'perspectives.mvdirona.com': 0.8659285828108513,
'digiconomist.net': 0.8659107168851126,
'futhark-lang.org': 0.8658910633267001,
'blog.stephanbehnke.com': 0.8658778264013361,
'www.themillions.com': 0.8658146270171843,
'www.charlespetzold.com': 0.8657610099215972,
'dev.opera.com': 0.8657226869029186,
'www.contrast.ie': 0.8657095900925946,
'threatpost.com': 0.8656950945183971,
'diginomica.com': 0.8656889877908728,
'www.iro.umontreal.ca': 0.865673655242536,
'goodui.org': 0.8656691714697559,
'kevinmahoney.co.uk': 0.8656053786488649,
'den.dev': 0.8655508375322314,
'blog.danlew.net': 0.8655009175546347,
'insidermag.net': 0.8654928424374734,
'eviltrout.com': 0.8654668848568644,
'bartwronski.com': 0.8653908212138248,
'asserttrue.blogspot.com': 0.8653811565244239,
'laughtonelectronics.com': 0.8653462441760643,
'www.toomas.net': 0.8653462441760643,
'www.peoplespolicyproject.org': 0.8653036598492724,
'www.kpcb.com': 0.865280579032659,
'engineering.stanford.edu': 0.8652651824846332,
'www.mcsweeneys.net': 0.8652414926783502,
'www.weeklystandard.com': 0.8652325002095317,
'scratch.mit.edu': 0.8652264286658259,
'therideshareguy.com': 0.8651696061186824,
'hyperboleandahalf.blogspot.com': 0.8651696061186824,
'evhead.com': 0.8651553668857692,
'praeclarum.org': 0.8651529057353353,
'riowang.blogspot.com': 0.8651391622821514,
'www.rs-online.com': 0.86513559976471,
'blog.linuxmint.com': 0.8651351630976869,
'blog.soylent.me': 0.8651258007804876,
'cabel.me': 0.8650904033379694,
'wildbit.com': 0.8650667180545558,
'www.huffingtonpost.com': 0.86501806564583,
'code.nasa.gov': 0.8649895088434025,
'zenhabits.net': 0.8649439383347666,
'dl.dropbox.com': 0.8649350287137462,
'devblog.avdi.org': 0.8648862222006666,
'blog.phpfog.com': 0.8648344406451537,
'www.thetransportpolitic.com': 0.8648252523162882,
'forestry.io': 0.8648063640773068,
'www.dabeaz.com': 0.8647748270793637,
'www.brightball.com': 0.8647346279381389,
'style.org': 0.8647346279381389,
'www.elezea.com': 0.8647230821512976,
'www.aarp.org': 0.864718332495695,
'themillions.com': 0.8646781650668347,
'saastr.quora.com': 0.8646637505280962,
'ottawacitizen.com': 0.8646581352452648,
'mkremins.github.io': 0.8646201560571354,
'edge.org': 0.8646127068735515,
'micropython.org': 0.8645605097088179,
'www.metamorphblog.com': 0.8645587566462556,
'blog.jez.io': 0.8645309820862669,
'aatishb.com': 0.8644462651203315,
'www.celinehh.com': 0.8644462651203315,
'davidyat.es': 0.8644462651203315,
'code.joejag.com': 0.8644356837498463,
'www.washingtonexaminer.com': 0.864373005099176,
'www.resetera.com': 0.8643497804356822,
'quantum.country': 0.8643112341108541,
'savingjournalism.substack.com': 0.8642759030508148,
'ahungry.com': 0.8642759030508148,
'gamozolabs.github.io': 0.8642719268946735,
'jamespanderson.tumblr.com': 0.8642680643637904,
'elementaryos.org': 0.8642208492393608,
'www.debuggex.com': 0.8642032789469397,
'www.planetizen.com': 0.8641967719680171,
'www.eweek.com': 0.864172798382722,
'tatiyants.com': 0.8641038586159879,
'discover.dc.nihr.ac.uk': 0.8640971396521386,
'web.cs.ucdavis.edu': 0.8640703175243611,
'xmpp.org': 0.8640502790628004,
'chriskohlhepp.wordpress.com': 0.8640478440979216,
'mystartuphas30daystolive.tumblr.com': 0.8640252337365474,
'rhelblog.redhat.com': 0.8640126486430193,
'www.electricitymap.org': 0.8639997125170845,
'americasmarkets.usatoday.com': 0.8639946073685961,
'projects.propublica.org': 0.8639224455926199,
'e-estonia.com': 0.8639219016557921,
'blog.basho.com': 0.8639181305846296,
'ungleich.ch': 0.863900787970001,
'squid314.livejournal.com': 0.8638782213420932,
'www.cassandracapital.net': 0.8638782213420932,
'aiyprojects.withgoogle.com': 0.8638467210374888,
'bsdimp.blogspot.com': 0.8638147888998915,
'jwt.io': 0.8638117466836764,
'www.esri.com': 0.8638089133245348,
'www.gaborcselle.com': 0.8637930884909581,
'venam.nixers.net': 0.8637807788335469,
'www.7pace.com': 0.8637364419671971,
'www.joshmcguigan.com': 0.8637220704617038,
'promarket.org': 0.8636701430867854,
'tav.espians.com': 0.8636677390061345,
'www.datastax.com': 0.8636570251868989,
'www.cmyr.net': 0.8636232605564672,
'mainisusuallyafunction.blogspot.com': 0.8636109472670038,
'www.hisutton.com': 0.8635963039783531,
'lsub.org': 0.8635700664090139,
'alpblog.heroku.com': 0.8635306327438487,
'gpuopen.com': 0.8635262100242561,
'nvd.nist.gov': 0.8634973542011698,
'english.elpais.com': 0.8634616688985162,
'www.craigslist.org': 0.863445932794428,
'nostalgebraist.tumblr.com': 0.8634392689015953,
'www.hakaimagazine.com': 0.8634375812585431,
'marcan.st': 0.8634342519721849,
'wiki.debian.org': 0.8633915097045765,
'php.net': 0.8633877833122457,
'benjaminreinhardt.com': 0.8633802816901408,
'blog.pnas.org': 0.86336052660942,
'www.adityaravishankar.com': 0.86336052660942,
'mobile.twitter.com': 0.8633312362141495,
'www.alistapart.com': 0.863319649821934,
'msgpack.org': 0.8633096782467932,
'keionline.org': 0.8633096782467932,
'www.logicalclocks.com': 0.8632631941253387,
'svedic.org': 0.8632309919400734,
'newsroom.unsw.edu.au': 0.8632254714408825,
'maxwellito.github.io': 0.8632001968821827,
'www.commandlinefu.com': 0.8631417710298782,
'www.clockworkpi.com': 0.8631140242937411,
'www.typography.com': 0.8630969520774667,
'alumni.stanford.edu': 0.8630912751881218,
'www.sigarch.org': 0.8630849989436014,
'www.countbayesie.com': 0.8630846691596683,
'blog.echen.me': 0.8630803489219858,
'groovehq.com': 0.863062038855598,
'calteches.library.caltech.edu': 0.863062038855598,
'www.rochester.edu': 0.8630082813677696,
'blog.bjrn.se': 0.8629931779073361,
'lawandcrime.com': 0.8629659867085913,
'betashop.com': 0.862958801056521,
'www.ericsink.com': 0.8629424736221599,
'engineering.linkedin.com': 0.8629291758709224,
'cleantechnica.com': 0.8629063952275409,
'meltingasphalt.com': 0.8628160615079228,
'amiunique.org': 0.8628087618596978,
'bvckup2.com': 0.8628005657708628,
'ieftimov.com': 0.8627657236548011,
'ewanvalentine.io': 0.8627020659905763,
'secalerts.co': 0.8627018556487716,
'www.patrick-wied.at': 0.8626866200547734,
'www.pitt.edu': 0.8626432815245133,
'hothardware.com': 0.8626338662999138,
'www.appleinsider.com': 0.8626260620914765,
'hillelwayne.com': 0.8626238486818072,
'gridstylesheets.org': 0.8626062322946175,
'sciencex.com': 0.8626007339716192,
'developers.slashdot.org': 0.86255044240195,
'blog.confluent.io': 0.8624968367523541,
'glandium.org': 0.8624955638682134,
'arandomurl.com': 0.8624894125730017,
'blog.nugget.one': 0.8624812956348196,
'www.rd.com': 0.8624301242273775,
'slifty.com': 0.8624113475177305,
'www.lighttable.com': 0.8624113475177305,
'payments.amazon.com': 0.8624113475177305,
'www.lurklurk.org': 0.862401566562782,
'derstandard.at': 0.862401566562782,
'workstuff.tumblr.com': 0.8623257751778021,
'www.lvh.io': 0.8623241846862385,
'rs.io': 0.8623129582427793,
'eslint.org': 0.862298463525373,
'www.cucumbertown.com': 0.8622891062009898,
'jquery.com': 0.8622837910960226,
'z.cash': 0.8622770331746279,
'www.ottawacitizen.com': 0.8622229843756593,
'nanonets.com': 0.8621765227456938,
'www.crunchgear.com': 0.8621559229318876,
'mattruby.substack.com': 0.8621556881643411,
'mbuffett.com': 0.8621556881643411,
'cybergibbons.com': 0.8621443416178289,
'www.lua.org': 0.8621439982277107,
'monodraw.helftone.com': 0.8621272923907152,
'vicki.substack.com': 0.8621255550009437,
'www.davidgcohen.com': 0.8621062295375266,
'jfrog.com': 0.8620661061833018,
'joostdevblog.blogspot.com': 0.8620661061833018,
'johnkary.net': 0.8620331171141896,
'jakobgreenfeld.com': 0.8620331171141896,
'bosker.wordpress.com': 0.8619844782836177,
'encosia.com': 0.8619710029888306,
'martinkool.com': 0.8619319552747774,
'use-the-index-luke.com': 0.8619185059392996,
'siteanalytics.compete.com': 0.8619154305011097,
'madewithlove.be': 0.8619031112487354,
'blog.digg.com': 0.8619019650013046,
'wildlyinaccurate.com': 0.8618675915764481,
'news.agu.org': 0.8618622951257247,
'www.hnsearch.com': 0.8618535032580247,
'bbc.com': 0.8618329035365963,
'thevaluable.dev': 0.8618329035365963,
'cve.mitre.org': 0.8618136566919955,
'www.r2d3.us': 0.8618128425706462,
'www.wsbtv.com': 0.8617435427240897,
'wiki.php.net': 0.8617039994341269,
'regex101.com': 0.861640779434757,
'www.fastcodesign.com': 0.8616335728873852,
'waitbutwhy.com': 0.8616048195074264,
'wrapbootstrap.com': 0.8615996055432654,
'www.oracle.com': 0.8615524291467332,
'angrystaffofficer.com': 0.8615283953190601,
'pythonprogramming.net': 0.8615283953190601,
'codemirror.net': 0.8615237207033272,
'erthalion.info': 0.8615038163918213,
'news.wsu.edu': 0.8615031765236854,
'jmtd.net': 0.8614607909822787,
'www.businessinsider.com.au': 0.8614418925622166,
'www.keithschwarz.com': 0.8614202461594263,
'blog.testdouble.com': 0.8614030589984097,
'rationalconspiracy.com': 0.8613615814645382,
'michaelochurch.wordpress.com': 0.8613601391266915,
'blogs.berkeley.edu': 0.8613538103460412,
'blog.superfeedr.com': 0.8613308060874829,
'hn-books.com': 0.8613200444920518,
'techsolidarity.org': 0.8613153663268618,
'annualletter.gatesfoundation.org': 0.8613153663268618,
'www.ardanlabs.com': 0.8612708268948052,
'www.mapbox.com': 0.8612700375538782,
'protonvpn.com': 0.8612360355266332,
'www.winrumors.com': 0.8611989655340138,
'www.arcticstartup.com': 0.8611695091473522,
'blogs.akamai.com': 0.8611395195993665,
'penntoday.upenn.edu': 0.8611371780676135,
'www.howitactuallyworks.com': 0.8611264027197298,
'www.theage.com.au': 0.8611116664910229,
'blogs.computerworld.com': 0.8610945749876592,
'news.artnet.com': 0.8610390836029311,
'yongfook.com': 0.8610327646216055,
'ask.metafilter.com': 0.8610302755520367,
'blog.zencoder.com': 0.8609595040886218,
'labs.spotify.com': 0.8609558167629312,
'ashfurrow.com': 0.8609452045938362,
'rachelandrew.co.uk': 0.8609370906301821,
'xem.github.io': 0.860936923468393,
'natureofcode.com': 0.8609280600555332,
'www.daedtech.com': 0.8609275762383382,
'jnd.org': 0.8609065111890492,
'cryptoslate.com': 0.8608950862608346,
'googlepolicyeurope.blogspot.com': 0.8608494490780891,
'electricliterature.com': 0.8608434109180494,
'www.cs.toronto.edu': 0.8608065788416125,
'mtgox.com': 0.8608060590690018,
'laughingsquid.com': 0.8607982963555786,
'nickdesaulniers.github.io': 0.8607761141579771,
'news.netcraft.com': 0.8607587653123538,
'www.mattfriz.com': 0.8607469264593318,
'blogs.computerworlduk.com': 0.8606535841782695,
'weblog.jamisbuck.org': 0.8606463446127512,
'particletree.com': 0.860619018859172,
'www.loopinsight.com': 0.8605648031182908,
'blog.qualys.com': 0.86055640956746,
'moz.com': 0.860495291636908,
'vimeo.com': 0.8604892664547054,
'www.chalmers.se': 0.860431654676259,
'homepage.ntlworld.com': 0.8604242343472561,
'www.jpost.com': 0.8604154523480785,
'www.governor.ny.gov': 0.8603434360710707,
'www.teenvogue.com': 0.8603394119266393,
'blog.bufferapp.com': 0.8603306431458172,
'pharr.org': 0.8603247905476864,
'www.cforcoding.com': 0.8603197271193488,
'mixpanel.com': 0.8603064014189074,
'bits-please.blogspot.com': 0.860301774508791,
'waypoint.vice.com': 0.8602882797773976,
'elpais.com': 0.8602682993832683,
'www.thevintagenews.com': 0.8602498276023521,
'bdickason.com': 0.8602305475504323,
'immunant.com': 0.8601526618229086,
'madhadron.com': 0.8601439808675682,
'www.dwheeler.com': 0.8600331290988332,
'www.vim.org': 0.8600002647579501,
'blog.prototypr.io': 0.859998158168068,
'webpolicy.org': 0.8599719205993361,
'coding.smashingmagazine.com': 0.8599291407790329,
'wizardzines.com': 0.8598885971042833,
'www.itnews.com.au': 0.8598724037037537,
'www.codethinked.com': 0.8598208830600297,
'www.technologynetworks.com': 0.8598085045605989,
'abclocal.go.com': 0.8598024771455696,
'blog.wolframalpha.com': 0.8597915197356469,
'blog.juxt.pro': 0.8597891002608672,
'newsroom.tiktok.com': 0.8597628867114886,
'defector.com': 0.8597558066643536,
'apps.apple.com': 0.8597457842210768,
'www.jamesyu.org': 0.8597029099342595,
'www.dailytech.com': 0.8596655487230563,
'elie.net': 0.8596315170135848,
'linear.app': 0.8596237337192475,
'clerk.house.gov': 0.8596237337192475,
'keepachangelog.com': 0.8596075354545292,
'featherweightmusings.blogspot.com': 0.8596075354545292,
'eng.localytics.com': 0.8595938002959398,
'emacsrocks.com': 0.8595762853230908,
'dassur.ma': 0.8595420928346226,
'www.sightline.org': 0.859531902937568,
'www.gamesradar.com': 0.8595242144121686,
'sysdig.com': 0.8595202139290916,
'www.cloudkick.com': 0.8595183443817482,
'devopsu.com': 0.859493942235387,
'believermag.com': 0.8594892467285535,
'unhandledexpression.com': 0.8594741803219609,
'efficiencyiseverything.com': 0.8594569005504344,
'www.therobotreport.com': 0.859433089374823,
'www.howtographql.com': 0.8594202898550725,
'archive.today': 0.8593643247719568,
'www.jacobinmag.com': 0.8593589358679861,
'progrium.com': 0.8593468157466189,
'redfin.engineering': 0.8593466877552149,
'www.zillow.com': 0.8593304013971459,
'www.nexedi.com': 0.8593210096380903,
'madaidans-insecurities.github.io': 0.8593210096380903,
'home.treasury.gov': 0.8592910289545252,
'scholar.harvard.edu': 0.8592733512209211,
'opensourceconnections.com': 0.8592682967350698,
'bsdly.blogspot.com': 0.8592666292050983,
'aras-p.info': 0.8592595047254012,
'battlepenguin.com': 0.8592397001892514,
'solokeys.com': 0.8592264014510466,
'www.alexkyte.me': 0.8591861061610623,
'www.digitalmusicnews.com': 0.8591609711742595,
'kenneth.io': 0.8591232018072773,
'mywiki.wooledge.org': 0.8590993783573315,
'epic.org': 0.8590931061006938,
'behavioralscientist.org': 0.8590931061006938,
'jack.minardi.org': 0.8590798567329014,
'zeptobars.ru': 0.8590350573746124,
'bramcohen.com': 0.8590132997607233,
'frankforce.com': 0.8590074352869126,
'erniemiller.org': 0.8590074352869125,
'robservatory.com': 0.8589982689488246,
'engineering.squarespace.com': 0.8589982689488246,
'www.artofmanliness.com': 0.8589698957080373,
'blogs.bing.com': 0.8589523074975127,
'www.world-nuclear-news.org': 0.8589473343420543,
'openmirage.org': 0.8589305193408414,
'prism-break.org': 0.8589248369205337,
'slab.com': 0.8589172698916326,
'monkeylearn.com': 0.8589030969473112,
'journals.sagepub.com': 0.8588825279469426,
'metaredux.com': 0.8588307379036221,
'www.eurogamer.net': 0.8588239617965691,
'www.wisdomination.com': 0.8587259126214145,
'src.chromium.org': 0.8586820383432011,
'www.tudelft.nl': 0.8586674667651873,
'blog.project-insanity.org': 0.8586633233128,
'hadihariri.com': 0.8586515656051527,
'blog.xot.nl': 0.8586508044813143,
'www.nypost.com': 0.8586366736605365,
'andrewchenblog.com': 0.8586246895566572,
'jordan-wright.com': 0.8586221711446911,
'journal.burningman.org': 0.858622171144691,
'blog.tensorflow.org': 0.8586139292072289,
'googlewebmastercentral.blogspot.com': 0.8586030875489932,
'standblog.org': 0.858600583090379,
'ant.design': 0.858600583090379,
'overapi.com': 0.8585933703160669,
'www.suntimes.com': 0.8585797957168473,
'thecable.foreignpolicy.com': 0.8585540700891416,
'earther.gizmodo.com': 0.8584865305709988,
'www.bestinclass.dk': 0.8584597924729722,
'hn.algolia.com': 0.8584545053239319,
'gwillem.gitlab.io': 0.8584541318402437,
'elbenshira.com': 0.8584541318402437,
'bankunderground.co.uk': 0.8583675490979057,
'www.crowdsupply.com': 0.8583166788787248,
'lists.linuxfoundation.org': 0.8583007929896874,
'mikehadlow.blogspot.com': 0.8583007929896874,
'prideout.net': 0.8582618132398327,
'joinup.ec.europa.eu': 0.8582375872301639,
'coderoom.wordpress.com': 0.8582225134679472,
'andrewoneverything.com': 0.8582199611092117,
'peterlyons.com': 0.8581871345029239,
'facebook.com': 0.8581759828028657,
'www.vogue.com': 0.8581630406503709,
'future.a16z.com': 0.8581407364644609,
'www.thedotpost.com': 0.8581294866962557,
'davidwalsh.name': 0.85807314827283,
'www.heise.de': 0.8580470280497292,
'regex.info': 0.8580444001760147,
'blog.jayway.com': 0.8580313921739998,
'www.discovermagazine.com': 0.8580086852563162,
'www.gamespot.com': 0.858006803378577,
'www.mattmazur.com': 0.8579544667792158,
'0x80.pl': 0.8579258153670124,
'pine64.com': 0.8579245737044574,
'www.gog.com': 0.8579147324679443,
'jdh.hamkins.org': 0.857912419834021,
'openjdk.java.net': 0.8578856841040577,
'code.djangoproject.com': 0.8578761757124823,
'airbnb.io': 0.8578759796916386,
'www.neatorama.com': 0.8578511503394977,
'intridea.com': 0.8578447662759853,
'techchannel.att.com': 0.8578445745748121,
'mule.substack.com': 0.8578445745748121,
'www.fastmailstatus.com': 0.857833367118916,
'nas.sr': 0.8578228253764041,
'cloudblogs.microsoft.com': 0.8578026799771654,
'danielbmarkham.com': 0.8577690138809076,
'blogs.kde.org': 0.8577593606755378,
'graphics.latimes.com': 0.8577592314786884,
'www.wordfence.com': 0.8577572254294423,
'quariety.com': 0.8577565674618444,
'www.ritholtz.com': 0.8577423661740396,
'jasonadriaan.com': 0.8576809443893051,
'joelcalifa.com': 0.8576809443893051,
'www.cs.umd.edu': 0.857661758811538,
'zacharyvoase.com': 0.857656134895176,
'www.decisionproblem.com': 0.8576347888613035,
'steamdb.info': 0.8576311826444905,
'blog.chef.io': 0.8576130873561624,
'techraptor.net': 0.8575933683258181,
'www.nanog.org': 0.857579845060974,
'www.cs.yale.edu': 0.8575769410337604,
'blogs.skype.com': 0.8575392141865088,
'igoro.com': 0.8575204710620608,
'www.math.toronto.edu': 0.8574671949652505,
'tether.to': 0.8574671949652504,
'adrianmejia.com': 0.8573817628591522,
'adblockplus.org': 0.8573641935296868,
'techxplore.com': 0.8573012751875033,
'professional-troublemaker.com': 0.8572977062348858,
'idealab.talkingpointsmemo.com': 0.8572223861972004,
'nesslabs.com': 0.8572196165500797,
'www.nationalgeographic.co.uk': 0.8571954684569548,
'iridakos.com': 0.8571705482234938,
'theagilepanda.com': 0.8571428571428571,
'togototo.wordpress.com': 0.8571428571428571,
'blog.skyliner.io': 0.8571354628699216,
'eu.techcrunch.com': 0.8571248597503665,
'datastudio.google.com': 0.8571140921734792,
'www.django-cms.org': 0.8570965178487575,
'blog.path.com': 0.8570914140075269,
'www.villagevoice.com': 0.857081275011955,
'somehowmanage.com': 0.857055205499945,
'www.mnn.com': 0.8570470364584489,
'abovethelaw.com': 0.8570463120033517,
'blog.logrocket.com': 0.8570182021331599,
'www.apress.com': 0.8569721488001009,
'www.news.cornell.edu': 0.8569706892788174,
'www.curbed.com': 0.8569597747993827,
'digitizor.com': 0.856955455369854,
'blog.lareviewofbooks.org': 0.8569521195010459,
'www.cambus.net': 0.8569380680580816,
'blog.rize.io': 0.8568954207368967,
'2ality.com': 0.8568509445047684,
'ux.stackexchange.com': 0.856847117644151,
'www.thedomains.com': 0.8568397055760405,
'timesofindia.indiatimes.com': 0.856835710146828,
'examine.com': 0.8568332988347195,
'speirs.org': 0.8568142257362032,
'learnbyexample.github.io': 0.8568010189328112,
'cs.brown.edu': 0.8567908824604603,
'www.dnalounge.com': 0.8567908041212858,
'www.nbcbayarea.com': 0.8567028634763063,
'macperformanceguide.com': 0.8566332090796122,
'social.techcrunch.com': 0.856633209079612,
'gocardless.com': 0.8566325894779986,
'www.driverless.id': 0.8565431663151448,
'simonsarris.substack.com': 0.856507155170115,
'blog.l0cal.com': 0.856499398164963,
'poy.time.com': 0.8564993981649629,
'labs.ripe.net': 0.8564206791243404,
'britishlibrary.typepad.co.uk': 0.8564116348023619,
'directorsblog.nih.gov': 0.8563781052354208,
'www.trustedsec.com': 0.8563276469242246,
'www.nicholaswindsorhoward.com': 0.8563276469242246,
'wefunder.com': 0.8563033480182963,
'grpc.io': 0.8563005637859097,
'www.innovationaus.com': 0.8562719072138797,
'algorithmwatch.org': 0.856263282232775,
'kylewritescode.com': 0.8562626989641361,
'www.iquilezles.org': 0.8562567878094767,
'www.doc.ic.ac.uk': 0.8561753420541418,
'www.r-bloggers.com': 0.8561384095933265,
'www.kxan.com': 0.8561304599253935,
'blog.shodan.io': 0.856101606479241,
'blog.couchdb.org': 0.8560857595989197,
'randsinrepose.com': 0.8560683987937139,
'blogs.igalia.com': 0.8560658162982266,
'erikhoel.substack.com': 0.8560552035555758,
'www.columbia.edu': 0.8560502120089583,
'adambard.com': 0.8560134134615376,
'xania.org': 0.8559909491087049,
'pinboard.in': 0.8559460897276601,
'lobste.rs': 0.8559156074776079,
'interestingengineering.com': 0.8558912746106632,
'www.dilbert.com': 0.855888464977386,
'jcdav.is': 0.8558808822556393,
'zwbetz.com': 0.8558715995865966,
'npr.org': 0.8558692421991085,
'compellingsciencefiction.com': 0.8558593149932564,
'headmelted.com': 0.8558229432148337,
'blog.itlater.com': 0.8558223030664946,
'www.joystiq.com': 0.8557770050631482,
'blog.loadzero.com': 0.8557371228445998,
'www.exploringbinary.com': 0.8557365123930964,
'www.reason.com': 0.8557121404409874,
'towardsdatascience.com': 0.8556587797122474,
'ostif.org': 0.855654761904762,
'articles.latimes.com': 0.855649649030829,
'www.cs.ucr.edu': 0.8556340456009993,
'www.transportenvironment.org': 0.8556309211595626,
'www.sciencenews.org': 0.8556293530573256,
'blog.portswigger.net': 0.855625404687521,
'dshipper.posterous.com': 0.8556180321566556,
'abopen.com': 0.8555419933626652,
'www.twitch.tv': 0.8555327933571086,
'www.realclearscience.com': 0.8555038318613801,
'www.llnl.gov': 0.8554731239960308,
'brandmark.io': 0.8554590545897166,
'www.eflorenzano.com': 0.8554573688896427,
'play.google.com': 0.8554165451041797,
'www.ethanzuckerman.com': 0.8553991433840825,
'lunduke.substack.com': 0.8553770673931268,
'www.montulli.org': 0.8553552610577579,
'www.dvorak.org': 0.8553507205142731,
'pythoninsider.blogspot.com': 0.855339221122335,
'octopart.com': 0.8553291642797942,
'ithare.com': 0.8553097001258215,
'isc.sans.edu': 0.8552998663643834,
'doi.org': 0.8552929122812928,
'www.starling-software.com': 0.855240263021274,
'sedimental.org': 0.8552238805970149,
'www.nitrokey.com': 0.8552078417856235,
'www.fool.com': 0.8551227533080648,
'ntietz.com': 0.8551173351343933,
'thepointmag.com': 0.8551125688254727,
'www.symantec.com': 0.8551104394486714,
'highline.huffingtonpost.com': 0.8550936296505218,
'blog.adacore.com': 0.8550882114429865,
'techland.time.com': 0.8550381807790381,
'www-03.ibm.com': 0.855023796089594,
'blog.davidtate.org': 0.8550204472368161,
'blog.gitlab.org': 0.8549666512282963,
'blog.dijit.sh': 0.8549645308164788,
'boredzo.org': 0.8549332224976328,
'www.bishopfox.com': 0.8549242562625317,
'blog.sproutcore.com': 0.8548755303332917,
'www.nakedcapitalism.com': 0.8548518987841709,
'hackaday.io': 0.8548507362141421,
'www.nccgroup.com': 0.8548006867421194,
'docs.microsoft.com': 0.8547764550021021,
'www.kansas.com': 0.8547447271228046,
'www.myfonts.com': 0.8547443058333942,
'www.databreaches.net': 0.8547388607511365,
'design.google': 0.8546474649453593,
'www.bvp.com': 0.8546145212136451,
'eriwen.com': 0.854565525324503,
'flightaware.com': 0.854535067640072,
'us2.campaign-archive1.com': 0.854531163801279,
'engineering.skroutz.gr': 0.8544744689256348,
'arcentry.com': 0.8544398366164802,
'redmonk.com': 0.8544165230751263,
'jesseduffield.com': 0.8543827046117866,
'www.foodandwine.com': 0.8543802950724103,
'www.jacobsschool.ucsd.edu': 0.8543584327039565,
'www.vandenoever.info': 0.8543584327039565,
'tubetime.us': 0.8543543543543544,
'xenbits.xen.org': 0.85435049306025,
'blogs.sun.com': 0.8543496525648886,
'nofilmschool.com': 0.8543378482464506,
'joi.ito.com': 0.8543316742759597,
'blog.garlicsim.org': 0.8543018539678975,
'akr.am': 0.854281488649609,
'www.nypl.org': 0.8542056110271361,
'blog.opendns.com': 0.8542056110271361,
'shmuplations.com': 0.8542031037522354,
'blog.statwing.com': 0.8542005291800069,
'm.bbc.com': 0.8541979742875379,
'boston.conman.org': 0.8541978420073731,
'lilium.com': 0.8541923613616496,
'phk.freebsd.dk': 0.8541720175820052,
'underthehood.meltwater.com': 0.8541376157274252,
'www.airliners.net': 0.8541362938673078,
'www.gobankingrates.com': 0.8541353383458646,
'maxplanck.nautil.us': 0.8540985777081339,
'leandromoreira.com.br': 0.8540707480810653,
'24ways.org': 0.8540629407199574,
'www.federaltimes.com': 0.8540125885303304,
'blog.trello.com': 0.8540087030052186,
'www.slash7.com': 0.8539974953863674,
'www.projectrho.com': 0.8539630717568143,
'yahoopolicy.tumblr.com': 0.8539187064781893,
'pencil.evolus.vn': 0.8539187064781893,
'getkerf.wordpress.com': 0.8539187064781893,
'newfoodeconomy.org': 0.853885872320658,
'www.geekosystem.com': 0.8538783892446405,
'bengtan.com': 0.8538510433274764,
'dustin.github.com': 0.8538486125176874,
'jasonlefkowitz.net': 0.8538441285736149,
'teamtreehouse.com': 0.8538268240476388,
'blog.quarkslab.com': 0.853807684540482,
'www.artnome.com': 0.8537866111759274,
'blog.chron.com': 0.8537460072706317,
'blog.close.io': 0.8537188598181367,
'lonesysadmin.net': 0.8536953242835595,
'www.lemonde.fr': 0.8536772759118291,
'www.guernicamag.com': 0.8536142442449905,
'codeascraft.etsy.com': 0.8536109251434499,
'blog.oleganza.com': 0.8535993869171182,
'www.poetryfoundation.org': 0.8535856365759876,
'archive.computerhistory.org': 0.8535651705660577,
'www.philzimmermann.com': 0.8535638214444478,
'www.sqlalchemy.org': 0.8535347372406801,
'www.rubyinside.com': 0.8535314579904915,
'rjbs.manxome.org': 0.8534743202416919,
'www.itl.nist.gov': 0.8534657928784027,
'lab.simurai.com': 0.8534578819472406,
'www.shortcutfoo.com': 0.8534538731861366,
'www.asymco.com': 0.8534535683227724,
'johnmathews.eu': 0.8534414067561554,
'www.msnbc.msn.com': 0.8534205452840495,
'www.cs.brown.edu': 0.8534113013764459,
'blog.vivekhaldar.com': 0.8534010422454464,
'www.t3.org': 0.8533802695967294,
'coloradosun.com': 0.8533683929679808,
'www.gamesindustry.biz': 0.8533666288339562,
'www.benjoffe.com': 0.8533494797004003,
'windows.microsoft.com': 0.8533490913172264,
'codeblog.jonskeet.uk': 0.8533155100042978,
'mattdesl.svbtle.com': 0.8532886204160572,
'wilshipley.com': 0.8532666831748213,
'lunduke.com': 0.8532546707850448,
'blog.documentcloud.org': 0.8532526475037822,
'pratt.duke.edu': 0.8532505235078635,
'forum.palemoon.org': 0.8532499119645807,
'ring.cx': 0.8532409206090049,
'www.healthimaging.com': 0.8532409206090049,
'blog.lawnstarter.com': 0.853234538226177,
'www.drorpoleg.com': 0.853234538226177,
'www.abajournal.com': 0.8532334026694913,
'volokh.com': 0.8532246743179585,
'www.akitasoftware.com': 0.8532127948822967,
'blog.algolia.com': 0.8531662937825892,
'cgit.freedesktop.org': 0.8531345095634363,
'creativegood.com': 0.8531247497260012,
'www.bizjournals.com': 0.8530952665861291,
'krypt.co': 0.85304727815771,
'david-peter.de': 0.8530425774087601,
'www.bitcoin.org': 0.853030303030303,
'www.markdotto.com': 0.853030303030303,
'donsbot.wordpress.com': 0.8530217272624623,
'joyent.com': 0.8530153570981687,
'googleplusplatform.blogspot.com': 0.8530153000371625,
'modern-products.tumblr.com': 0.8530153000371625,
'hashicorp.com': 0.8529426622739795,
'www.tuaw.com': 0.8529238643536575,
'jessepollak.me': 0.8528766579532876,
'www.spectrum.ieee.org': 0.8528722385076482,
'evasi0n.com': 0.8527890991537469,
'continuum.io': 0.8527523717085685,
'inconshreveable.com': 0.8527291982231584,
'washingtonmonthly.com': 0.8527169016435672,
'www.go350.com': 0.8526967137013025,
'newhumanist.org.uk': 0.8526876181950579,
'www.thelocal.se': 0.8526535411240088,
'arielelkin.github.io': 0.8526249122103083,
'ldpreload.com': 0.8526076535424412,
'blog.valerieaurora.org': 0.8526007849075418,
'tympanus.net': 0.8526005626429696,
'www.pewresearch.org': 0.8525869052062965,
'gcaptain.com': 0.8525025061695275,
'www.fireeye.com': 0.8524966902771423,
'mkdev.me': 0.8524961199751095,
'thoughtmaybe.com': 0.8524858010387742,
'blog.lukaszolejnik.com': 0.8524707074061916,
'getmondo.co.uk': 0.852437604863552,
'open.bufferapp.com': 0.852434953761713,
'alumni.berkeley.edu': 0.8524296840496752,
'forums.macrumors.com': 0.8523683544548385,
'alastairreid.github.io': 0.8523644066969532,
'www.scottberkun.com': 0.852359208523592,
'ontwik.com': 0.8523482955385304,
'git-scm.com': 0.8523082962459579,
'news.un.org': 0.8522696996619386,
'www.adafruit.com': 0.8522616020934412,
'www.mathjax.org': 0.8522588583128298,
'www.chinalawblog.com': 0.8522208346251817,
'yahoolabs.tumblr.com': 0.8522048661669432,
'www.dw.de': 0.8521590754317577,
'www.udacity.com': 0.8521539143105872,
'dianeravitch.net': 0.8521341463414633,
'www.narrator.ai': 0.852097863618245,
'broadbandnow.com': 0.852097863618245,
'genius.com': 0.8520940394549934,
'theaircurrent.com': 0.8519939439576096,
'www.bytecellar.com': 0.8519428423381916,
'www.rifters.com': 0.8519032600490041,
'www.math.upenn.edu': 0.8518649443865483,
'algorithmsoup.wordpress.com': 0.8518190799236097,
'ellegriffin.substack.com': 0.8518190799236097,
'mikewest.org': 0.8517824188629416,
'loyaltylobby.com': 0.8516821355411338,
'blog.hut8labs.com': 0.8516819571865444,
'www.metmuseum.org': 0.8516671986689323,
'www-users.cs.umn.edu': 0.8516590199664924,
'undo.io': 0.8516245163512932,
'news.discovery.com': 0.8515880430654605,
'hulry.com': 0.8515702882943715,
'www.greenheartgames.com': 0.8515702882943715,
'ie.microsoft.com': 0.8515602032387396,
'www.thehindu.com': 0.8515477928389076,
'www.ably.io': 0.8515404721157463,
'chrispenner.ca': 0.8514969974322367,
'www.logodesignlove.com': 0.8514733938627516,
'people.ischool.berkeley.edu': 0.8514695464544928,
'timothycrosley.github.io': 0.8514548238897397,
'www.neocomputer.org': 0.8514548238897397,
'm.motherjones.com': 0.8514463981383273,
'ro-che.info': 0.8514172067988788,
'www.ofnumbers.com': 0.8514160013087799,
'rdist.root.org': 0.8514077817668315,
'blog.alexellis.io': 0.8514045841413571,
'www.notebookcheck.net': 0.8513578715363501,
'new.livestream.com': 0.8513411202985242,
'www.sonyaellenmann.com': 0.8513367975602456,
'builtin.com': 0.8513357366412594,
'thecritic.co.uk': 0.8513136478418425,
'bitworking.org': 0.851299858555184,
'iwantmyname.com': 0.8512894868376749,
'www.circleid.com': 0.851254578211912,
'woboq.com': 0.8512444088430177,
'www.instructables.com': 0.8512139095333348,
'markmail.org': 0.8512136385469307,
'da-data.blogspot.com': 0.8511940319698232,
'www.inverse.com': 0.8511748736762975,
'robinhood.com': 0.8511710385891778,
'blog.tojicode.com': 0.8511685296886505,
'usa.streetsblog.org': 0.8511588204743716,
'sanderknape.com': 0.851058381116051,
'technical.ly': 0.8510465053531708,
'www.trek10.com': 0.8510328925328215,
'theprint.in': 0.8510207129082807,
'000fff.org': 0.8509998804696824,
'journal.frontiersin.org': 0.8509377801906453,
'on.wsj.com': 0.850920176035645,
'sircmpwn.github.io': 0.8508709080357458,
'www.networkworld.com': 0.8508447666287313,
'apps.bostonglobe.com': 0.8508308486054927,
'pfrazee.hashbase.io': 0.8508225328373329,
'www.commentarymagazine.com': 0.8508100679431329,
'petewarden.com': 0.8507826894953255,
'www.exeter.ac.uk': 0.8507809015713883,
'www.statwing.com': 0.8507692307692307,
'www.unenvironment.org': 0.8507692307692307,
'www.paintcodeapp.com': 0.8507513412377187,
'andreschweighofer.com': 0.8507309300531638,
'forums.unrealengine.com': 0.8507306931858649,
'www.wikidata.org': 0.8507275608332324,
'draculatheme.com': 0.8507209915723823,
'hongkongfp.com': 0.8507154665320323,
'psmag.com': 0.8507120935262249,
'blog.flightcaster.com': 0.850683949182937,
'vancouversun.com': 0.8506812814664019,
'www.cc.gatech.edu': 0.8506656522986229,
'news.wisc.edu': 0.8506650572190997,
'madebymany.com': 0.850661432792337,
'www.afp.com': 0.8506381100610537,
'cudl.lib.cam.ac.uk': 0.8505963477950123,
'www.huyng.com': 0.8505963477950121,
'www.jeremymorgan.com': 0.8505908966433915,
'thereformedbroker.com': 0.8505772906521217,
'bigocheatsheet.com': 0.8505365958446607,
'matduggan.com': 0.850500461698309,
'deis.io': 0.8504704364122011,
'poolp.org': 0.8504003265009309,
'alteredqualia.com': 0.8503991615824809,
'product.hubspot.com': 0.8503749082035804,
'www.erlang-factory.com': 0.8503149886076794,
'paranoia.dubfire.net': 0.850273893946855,
'concurrencyfreaks.blogspot.com': 0.850256626665339,
'www.cato-at-liberty.org': 0.8502310137099974,
'databasearchitects.blogspot.com': 0.8502084998761508,
'www.netflixprize.com': 0.8501934998584227,
'www.utilitydive.com': 0.8501445783462607,
'www.olioapps.com': 0.850137018218358,
'rcrowley.org': 0.850105596274167,
'kentonv.github.io': 0.8500969474584005,
'hastebin.com': 0.8500573664501953,
'www.dotnetfoundation.org': 0.850018648725371,
'mattmazur.com': 0.8499955866420006,
'breakoutstartups.substack.com': 0.8499877911408882,
'www.platformonomics.com': 0.8499799037669202,
'www.groovehq.com': 0.8499412891580047,
'alf.nu': 0.8499136585227424,
'antoinevastel.com': 0.8498761519134193,
'liveweave.com': 0.8498511311102708,
'causal.app': 0.8498452012383901,
'liw.fi': 0.8498452012383901,
'www.paypal.com': 0.8497695268466062,
'redis.io': 0.8497424748729502,
'collectiveidea.com': 0.8496851714802387,
'felleisen.org': 0.8496847068727177,
'blog.nobugware.com': 0.8496603118181073,
'www.windowscentral.com': 0.8496154708571939,
'www.xobni.com': 0.8495968094932279,
'outage.report': 0.8495880396284066,
'www.justsecurity.org': 0.849566454817199,
'blog.rfox.eu': 0.8495467369935737,
'way-cooler.org': 0.8495467369935737,
'adamierymenko.com': 0.849538361390885,
'mesosphere.com': 0.8495051866724436,
'munchies.vice.com': 0.8495023041501157,
'meatfighter.com': 0.8495018599221447,
'dsc.discovery.com': 0.8494918847267826,
'sciencenordic.com': 0.8494773842761618,
'uxmovement.com': 0.8494740695810459,
'death.andgravity.com': 0.8494550556394679,
'www.idlewords.com': 0.8493746001892146,
'staffeng.com': 0.8493531997757849,
'godottutorials.pro': 0.8493096657310281,
'earthobservatory.nasa.gov': 0.8493083274285178,
'www.gladwell.com': 0.8492393112893071,
'www.cgpgrey.com': 0.8492221809584313,
'www.dailymail.co.uk': 0.8492114987186158,
'source.opennews.org': 0.8492058945166755,
'mikecanex.wordpress.com': 0.849180109488917,
'unsharpen.com': 0.8491592756377232,
'haseebq.com': 0.8491526060447331,
'blog.diasporafoundation.org': 0.849136592576036,
'www.pugetsystems.com': 0.8491094043930865,
'roadmap.sh': 0.8491002596731841,
'www.rawstory.com': 0.8490683893365794,
'www.nylas.com': 0.8490240168878735,
'www.tunetheweb.com': 0.8489936415581292,
'www.subtraction.com': 0.8489862226571359,
'adam.blog.heroku.com': 0.8489616756127251,
'www.gdcvault.com': 0.8489549021982677,
'n26.com': 0.8489269016604418,
'www.passportindex.org': 0.8489269016604418,
'www.sfweekly.com': 0.8489155778996055,
'wiki.csswg.org': 0.8489096573208723,
'remoteok.io': 0.8488395634361555,
'www.guru3d.com': 0.8488079037622022,
'www.wisdom.weizmann.ac.il': 0.8487791265218381,
'fred.stlouisfed.org': 0.8487687545118902,
'www.quantcast.com': 0.8487674782234258,
'www.dspguide.com': 0.8487415764723351,
'monitor.firefox.com': 0.8487326723774654,
'www.manchester.ac.uk': 0.8487326070292353,
'www.ecowatch.com': 0.8487187193792909,
'exponents.co': 0.8487052872629832,
'ubere.ng': 0.8486844576029813,
'lists.cs.uiuc.edu': 0.8486469141688284,
'www.erasmatazz.com': 0.8485843687777063,
'prestonbyrne.com': 0.8485821394042593,
'fakesteve.blogspot.com': 0.8485803980992321,
'andyet.net': 0.8485375978388094,
'blog.getfirebug.com': 0.8485293903506373,
'joshtronic.com': 0.8485255486310321,
'ngm.nationalgeographic.com': 0.8484980049334815,
'www.cat-bus.com': 0.8484880761635857,
'oos.moxiecode.com': 0.8484880761635857,
'dlo.me': 0.8484699284663549,
'www.openbazaar.org': 0.8484375,
'tumbledry.org': 0.8484375,
'boris-marinov.github.io': 0.8484375,
'www.pcmag.com': 0.8484334813622002,
'wsj.com': 0.8484234837206328,
'ssd.eff.org': 0.8484099350915519,
'www.worldometers.info': 0.8483883403559179,
'www.denverpost.com': 0.8483741516385197,
'blog.teambox.com': 0.8482961926347361,
'www.staff.science.uu.nl': 0.8482766434239408,
'chromium-review.googlesource.com': 0.8482702404753355,
'www.antoarts.com': 0.8482404035930796,
'www.redcode.nl': 0.8482404035930796,
'www.morningstar.com': 0.8482320069770474,
'www.nomachetejuggling.com': 0.8481585934322239,
'influxdata.com': 0.8481227202820945,
'ericholscher.com': 0.8481140078411193,
'www.nfx.com': 0.8481012921108911,
'wellpreparedmind.wordpress.com': 0.8480622757690688,
'kara.allthingsd.com': 0.8480371905314535,
'taptaptap.com': 0.8480285039672184,
'rhizome.org': 0.8479990358005195,
'c-faq.com': 0.8479835930179948,
'lighttable.com': 0.8478713990382907,
'www.ocweekly.com': 0.8478713990382906,
'www.yesmagazine.org': 0.8478423076321161,
'boston.cbslocal.com': 0.8477838372324837,
'sethvargo.com': 0.847722900935325,
'www.teamliquid.net': 0.8477100468877423,
'kn100.me': 0.8476871371630261,
'carsonified.com': 0.8476327362582359,
'makecode.com': 0.8476290144215641,
'www.leejamesrobinson.com': 0.8476290144215641,
'journals.lww.com': 0.8476245506194432,
'philfreo.com': 0.8476129588207604,
'insights.som.yale.edu': 0.8475823333113045,
'fredwu.me': 0.8475685309018646,
'www.desertsun.com': 0.8475476379514038,
'blog.noctua-software.com': 0.8475078292773957,
'engineering.twitter.com': 0.8474442201925468,
'elazzabi.com': 0.847403765595545,
'camendesign.com': 0.8473892733773356,
'spaceflightnow.com': 0.8473844409998718,
'www.cs.uic.edu': 0.8473371233913075,
'www.lpalmieri.com': 0.8473049751829188,
'www.call-with-current-continuation.org': 0.8472612335852765,
'pub.gajendra.net': 0.8472440944881889,
'write.privacytools.io': 0.8471964059191153,
'blog.gitprime.com': 0.8471964059191153,
'deepstreamhub.com': 0.8471964059191153,
'fosspost.org': 0.8471934496110164,
'www.staradvertiser.com': 0.8471904258144719,
'daggle.com': 0.8471392026866065,
'text.sourcegraph.com': 0.8471325439607902,
'graphics.reuters.com': 0.8470805937710428,
'www.pulumi.com': 0.8470801636194034,
'deepstream.io': 0.8470226785546888,
'shop.oreilly.com': 0.8469734259689878,
'www.rebol.com': 0.8469659113684828,
'nylas.com': 0.8469368267588653,
'sethgodin.typepad.com': 0.8469363395112108,
'oilprice.com': 0.8469313973547731,
'www.itsnicethat.com': 0.8469219290040122,
'www.theblackvault.com': 0.8468972040529525,
'jboriss.wordpress.com': 0.8468972040529525,
'jeena.net': 0.8468879391324495,
'www.mixnode.com': 0.8468817868572549,
'danco.substack.com': 0.8468470187947723,
'dyno-might.github.io': 0.8468364833167144,
'extremelearning.com.au': 0.8467447980712459,
'news.crunchbase.com': 0.8467020525586185,
'the-japan-news.com': 0.8466850596011346,
'www.tor.com': 0.8466742350710085,
'graphics.pixar.com': 0.8466553705876948,
'jackmott.github.io': 0.846643484398615,
'www.deekit.com': 0.8466140516084654,
'gkbrk.com': 0.8466121179458332,
'washingtonexaminer.com': 0.846542334844308,
'www.jayhuang.org': 0.8465047427289335,
'lunchbag.ca': 0.8464844037983796,
'www.vanschneider.com': 0.8464718715474767,
'www.seobook.com': 0.8464370807656654,
'lfe.io': 0.8464044390292613,
'www.unitedstatescourts.org': 0.8463713866332889,
'www.listennotes.com': 0.8463577921119287,
'techblog.livingsocial.com': 0.8463371185563517,
'blogoscoped.com': 0.8463347353586499,
'www.density.io': 0.8463125089091263,
'dbasch.posterous.com': 0.8462528298095997,
'martiancraft.com': 0.8462422366153325,
'blog.meldium.com': 0.8461981486766226,
'blog.codeclimate.com': 0.8461758267783162,
'www.abtinforouzandeh.com': 0.8461646472697207,
'rachelnabors.com': 0.8461583302534714,
'zencoder.com': 0.8461094533062384,
'www.itworld.com': 0.8460477806046857,
'blog.ipinfo.io': 0.846031746031746,
'www.popsci.com': 0.8460295003872216,
'www.autoweek.com': 0.8460205561195527,
'bbot.org': 0.8459877925136515,
'crookedtimber.org': 0.8459783522151496,
'www.cultofmac.com': 0.8459180081962092,
'en.wikisource.org': 0.8458715274833618,
'www.kaggle.com': 0.8458458101824604,
'www.europarl.europa.eu': 0.8458356717488885,
'benchmarksgame.alioth.debian.org': 0.8458109252610799,
'librepcb.org': 0.8457869634340223,
'www.nationalobserver.com': 0.8457869634340223,
'skitch.com': 0.8457433438672782,
'www.bfilipek.com': 0.8457240963150365,
'scottaaronson.com': 0.8456937780186609,
'fee.org': 0.8456866411435615,
'www.doomworld.com': 0.8456828105532045,
'librivox.org': 0.845661772928326,
'www.crispymtn.com': 0.845661772928326,
'www.nydailynews.com': 0.8455856575577351,
'brilliantmaps.com': 0.8455577844591506,
'tiamat.tsotech.com': 0.8455576461751069,
'npf.io': 0.8455538969123215,
'jonathan.bergknoff.com': 0.8455538969123215,
'belitsoft.com': 0.8455477879801876,
'blog.thecybershadow.net': 0.8455414012738853,
'blog.privacytools.io': 0.8455414012738853,
'hardmath123.github.io': 0.8455383067302876,
'elifesciences.org': 0.8455333018767387,
'www.nemil.com': 0.845471203066534,
'web.eecs.umich.edu': 0.8454557317112926,
'500hats.typepad.com': 0.8454234389310107,
'jsonfeed.org': 0.8454122902628886,
'opendatastructures.org': 0.8454122902628886,
'tex.stackexchange.com': 0.8454020068893373,
'yifan.lu': 0.8453802898253059,
'www.sierraclub.org': 0.8453802898253059,
'axisofeval.blogspot.com': 0.845371248915586,
'www.lbatalha.com': 0.8453195049851959,
'theferrett.livejournal.com': 0.8452950558213717,
'libmill.org': 0.8452950558213717,
'www.epa.gov': 0.8452050900524932,
'tricycle.org': 0.8451979459749552,
'gking.harvard.edu': 0.8451920342583992,
'seattletimes.nwsource.com': 0.8451827104383299,
'killscreen.com': 0.8451534193795461,
'www.pixeltrue.com': 0.8451429290751707,
'www.randalolson.com': 0.8451252770638041,
'en.blog.wordpress.com': 0.8450869071274824,
'blog.thijsalkema.de': 0.845084400536627,
'www.zephoria.org': 0.8450267106265884,
'paulromer.net': 0.8450058741046084,
'blog.gtk.org': 0.8449976566750211,
'blog.flattr.net': 0.8449113529737324,
'seedmagazine.com': 0.8448457357890011,
'www.rt.com': 0.8448330221761664,
'technosophos.com': 0.8448317661673463,
'monzo.com': 0.844824966969565,
'www.imperial.ac.uk': 0.8448169716366182,
'andreyf.tumblr.com': 0.8448123793732512,
'renesd.blogspot.com': 0.8448109932557698,
'www.secretgeek.net': 0.8447678779550777,
'alistapart.com': 0.8447602813793536,
'facebook.design': 0.8446854923268982,
'sachachua.com': 0.8446808335848872,
'www.codesections.com': 0.8446790832287159,
'davidwong.fr': 0.844658979317956,
'blogs.office.com': 0.8446378605952511,
'root.cern.ch': 0.8446152758619639,
'indieweb.org': 0.8446147837430709,
'blog.reemer.com': 0.8446120410396897,
'www.change.org': 0.844610318859491,
'www.mrspeaker.net': 0.8446002192794787,
'www.sun-sentinel.com': 0.8445943201382862,
'gorban.org': 0.844551282051282,
'www.bgr.com': 0.8445425484905292,
'coinmarketcap.com': 0.8445032134621009,
'0day.work': 0.8444799253517218,
'blog.realkinetic.com': 0.8444710920604793,
'm.bbc.co.uk': 0.844465510423491,
'earther.com': 0.8444605232524242,
'link.springer.com': 0.8444514328769025,
'positech.co.uk': 0.8444262068957591,
'www.onthemedia.org': 0.8444066696037038,
'www.cleveralgorithms.com': 0.8444062415001815,
'mailchi.mp': 0.8443098020754962,
'www.pixelapse.com': 0.8443017656500802,
'berniesanders.com': 0.8443017656500802,
'a.wholelottanothing.org': 0.8442952471954357,
'events.ccc.de': 0.8442791650208269,
'www.chicagobusiness.com': 0.8442628765380391,
'nsl.com': 0.844239240640813,
'srcco.de': 0.844239240640813,
'testpilot.firefox.com': 0.8442195629693428,
'noahveltman.com': 0.8441367921512271,
'jamiequint.com': 0.8441367921512271,
'reverb.com': 0.8441367921512271,
'www.pri.org': 0.8441332017534635,
'www.trustnodes.com': 0.8441221359793255,
'blogs.mathworks.com': 0.8440748728050913,
'www.cncf.io': 0.8440600291289487,
'opensource.com': 0.844058190643116,
'about.digg.com': 0.8440360228264422,
'news.brown.edu': 0.8440209497481723,
'www.forth.com': 0.8440090068688062,
'blockstream.com': 0.843993466124853,
'getaether.net': 0.8439867218274119,
'www.hooktheory.com': 0.8439867218274119,
'irssi.org': 0.8439867218274119,
'jemma.dev': 0.8439780808263159,
'gittup.org': 0.8439780808263159,
'www.dropboxforum.com': 0.8439738032049753,
'blog.neocities.org': 0.8439723488535399,
'www.aopa.org': 0.8439413757138091,
'sappingattention.blogspot.com': 0.8439256178572813,
'math-blog.com': 0.843908007189824,
'meetup.hackernewsdc.org': 0.8438982906033671,
'www.ibtimes.sg': 0.8438980758876966,
'curiousrat.com': 0.8438980758876966,
'www.norvig.com': 0.8438704067825865,
'michaelnielsen.org': 0.8438489596573671,
'www.ponylang.io': 0.843843768902312,
'www.noisebridge.net': 0.843843768902312,
'apps.fcc.gov': 0.8438354432862912,
'www.eenews.net': 0.8437777995615074,
'www.dhs.gov': 0.8437568931274587,
'www.uu.se': 0.8437434035754188,
'blog.streak.com': 0.8437430051067691,
'bluishcoder.co.nz': 0.8436990841434638,
'trello.com': 0.8436685239481871,
'prolost.com': 0.8436572621196788,
'corgibytes.com': 0.843643069430385,
'cbloomrants.blogspot.com': 0.8436415581159902,
'grouptalent.com': 0.8436259181114076,
'news.sky.com': 0.8435750425055707,
'dankaminsky.com': 0.8435552716799266,
'blog.kaspersky.com': 0.8435542294083143,
'blog.rubbingalcoholic.com': 0.8435483870967742,
'paul.copplest.one': 0.8435464913857005,
'andrewhy.de': 0.8435425442649045,
'www.techfounder.net': 0.8435095761996304,
'spinroot.com': 0.8435095761996304,
'garethrees.org': 0.8435072199487013,
'blog.gedrap.me': 0.8434993251797909,
'oceans.nautil.us': 0.8434897723987279,
'www.autonlab.org': 0.8434766589029354,
'kallithea-scm.org': 0.8433870123342087,
'www.getrevue.co': 0.8433604592288586,
'www.whatsonweibo.com': 0.8433258307054031,
'tass.com': 0.843318150706682,
'paste.lisp.org': 0.8433142035934179,
'www.rollcall.com': 0.8433069334851104,
'reference.wolfram.com': 0.8432544830728409,
'opendotdotdot.blogspot.com': 0.8432276309668896,
'kevin.burke.dev': 0.8432276309668896,
'decisiondata.org': 0.8432276309668896,
'xenaproject.wordpress.com': 0.8432246085096772,
'benbernardblog.com': 0.8432189502435823,
'usefathom.com': 0.8432043711877606,
'www.pcpro.co.uk': 0.8431913701641113,
'progit.org': 0.8431775195017707,
'i.imgur.com': 0.8431547290624056,
'developer.okta.com': 0.8431277870072948,
'blog.medium.com': 0.8431066839897218,
'ffii.org': 0.8430969026172707,
'byrslf.co': 0.8430850673452497,
'sigops.org': 0.8430850673452497,
'blog.whoishiring.io': 0.8430812250163859,
'www.mcvuk.com': 0.8430765289027752,
'www.google.pl': 0.843042071197411,
'blog.untogether.co.uk': 0.843042071197411,
'www.uh.edu': 0.8430266348842538,
'fable.io': 0.8429358535210576,
'www.hizook.com': 0.8429332432339712,
'www.ovh.com': 0.8429034962421136,
'www.tedinski.com': 0.8428702241660795,
'entrepreneur.venturebeat.com': 0.8428673093425917,
'www.c-spanvideo.org': 0.8428673045430252,
'www.historyextra.com': 0.8428269409461818,
'www.defenseone.com': 0.8428181514777605,
'volument.com': 0.8428160789962235,
'www.imaging-resource.com': 0.8428098585582268,
'popupcity.net': 0.8427876823338736,
'www.skyandtelescope.com': 0.8427597458791004,
'www.viva64.com': 0.8426735071847921,
'existentialtype.wordpress.com': 0.8426727373097719,
'www.ikea.com': 0.8426410914015982,
'earnestcapital.com': 0.8426131831456453,
'jmarbach.com': 0.8425928309863708,
'chrisharrison.net': 0.8425324675324676,
'www.jsonline.com': 0.8425180713598075,
'parenting.nytimes.com': 0.8425153384707086,
'tesla-cdn.thron.com': 0.8424060866368837,
'www.neuroscientistnews.com': 0.8424034611022545,
'www.wondersofphysics.com': 0.842390899623404,
'www.lanl.gov': 0.8423253496487528,
'skeptics.stackexchange.com': 0.8423210116773696,
'www.rfc-editor.org': 0.8422724006868584,
'www.industrialempathy.com': 0.8422674053917267,
'performancejs.com': 0.8422235444245709,
'infiniteundo.com': 0.8422087232147684,
'www.heroku.com': 0.8422087232147684,
'blog.lse.epita.fr': 0.8421833530068131,
'lacker.io': 0.8421778420117882,
'timeline.com': 0.8421633036155716,
'wiki.blender.org': 0.8420940430794808,
'www.freakonomics.com': 0.8420407955613295,
'blog.fugue.co': 0.8420361620441877,
'mediadecoder.blogs.nytimes.com': 0.8419912813368542,
'www.irs.gov': 0.8419451366935324,
'www.golem.de': 0.8419385789535662,
'pingcap.com': 0.8419161509277815,
'john.onolan.org': 0.8418365869145599,
'blog.mocality.co.ke': 0.8417877979157168,
'status.coinbase.com': 0.8417606283759475,
'www.attejuvonen.fi': 0.8417377670517942,
'serversforhackers.com': 0.8417376768816427,
'work.caltech.edu': 0.8417166624718854,
'fuelyourcoding.com': 0.841709160679554,
'dyn.com': 0.8416196735446702,
'blog.tabini.ca': 0.8415950088038011,
'ieeexplore.ieee.org': 0.8415919239489138,
'www.israel21c.org': 0.8415492534155486,
'rubyonrails.org': 0.8415351989418526,
'www.designweek.co.uk': 0.8415188999809792,
'collison.ie': 0.8415120389596948,
'www.evilmadscientist.com': 0.8414856456254624,
'muddylemon.com': 0.8414700703965355,
'agateau.com': 0.8414619982013722,
'blog.miguelgrinberg.com': 0.8414611167027505,
'nmap.org': 0.8414349271123382,
'www.theblockcrypto.com': 0.8414118140696276,
'www.haaretz.com': 0.8413639156892987,
'codecapsule.com': 0.8413373129936518,
'ethz.ch': 0.84131835872283,
'blog.bloomca.me': 0.8413144015517519,
'blog.snapeda.com': 0.8413079629673079,
'evonomics.com': 0.8413074174618784,
'www2.macleans.ca': 0.8412847684511929,
'composition.al': 0.841279463048803,
'js13kgames.com': 0.8412786874884526,
'lists.basho.com': 0.8412692653556527,
'people.redhat.com': 0.8412692653556527,
'oceanservice.noaa.gov': 0.8412678170260773,
'sciencenorway.no': 0.8412438625204582,
'speaking.io': 0.8412438625204582,
'www.exceptionnotfound.net': 0.8412220203835995,
'www.2ality.com': 0.8411429398560686,
'www.eecs.harvard.edu': 0.841119342861029,
'bldgblog.blogspot.com': 0.8411187765808322,
'thisismynext.com': 0.8410701280879982,
'unchartedterritories.tomaspueyo.com': 0.8410441941601818,
'www.skilled.io': 0.8410441941601818,
'tenfourfox.blogspot.com': 0.8410170033847231,
'thebreakfastpost.com': 0.8410159372229994,
'blog.bumblebeelabs.com': 0.8409937866242836,
'mmikowski.github.io': 0.8409753773821584,
'paultyma.blogspot.com': 0.8409535720717558,
'materialize.io': 0.8409502033267174,
'www.ncdc.noaa.gov': 0.840942212674096,
'www.haystacksoftware.com': 0.8409422126740959,
'techiedelight.quora.com': 0.8409422126740959,
'blog.modernmechanix.com': 0.8408958299079167,
'www.math3ma.com': 0.8408873008599574,
'retractionwatch.com': 0.8408852959781978,
'www.appleoutsider.com': 0.8408203431401681,
'www.dreamsongs.com': 0.8408038061650092,
'blogs.cisco.com': 0.8407980680062469,
'www.research.ibm.com': 0.8407779430526359,
'gohugo.io': 0.8407728272201884,
'nicolodavis.com': 0.8407632567723468,
'www.andreykurenkov.com': 0.8407450453082168,
'conversionxl.com': 0.8407343949968893,
'www.qemu-advent-calendar.org': 0.8407224958949097,
'enfranchisedmind.com': 0.8407038581523874,
'ozkatz.github.com': 0.8406930647478468,
'ahrefs.com': 0.8406628331444651,
'paulmck.livejournal.com': 0.8406466273816793,
'spf13.com': 0.8406252111185971,
'adamschwartz.co': 0.8406212712895198,
'scikit-learn.org': 0.8406196399338859,
'www.lessonsoffailure.com': 0.8406081585039813,
'mathenchant.wordpress.com': 0.8406026505628841,
'charlesleifer.com': 0.8405838838741042,
'zeux.io': 0.8405743435687888,
'bicycledutch.wordpress.com': 0.8405743435687888,
'ai.stanford.edu': 0.8405537358599618,
'iximiuz.com': 0.8405441526018748,
'kerkour.com': 0.8405182576561551,
'www.racked.com': 0.8404972480623266,
'news.morningstar.com': 0.8404838510660001,
'hfet.org': 0.8404775873220368,
'newcome.wordpress.com': 0.8404605263157895,
'enterprisecraftsmanship.com': 0.8404567783699838,
'dailynous.com': 0.8404286033548175,
'usersknow.blogspot.com': 0.8402966516996352,
'www.sandiegouniontribune.com': 0.8402646673456328,
'www.playframework.com': 0.8402476003167625,
'betterhumans.coach.me': 0.840238289918768,
'thomaspark.co': 0.8402340830446674,
'deadlockempire.github.io': 0.8401976935749588,
'mimimatthews.com': 0.8401762295218094,
'chris.beams.io': 0.8401743605753639,
'www.whoismcafee.com': 0.8401743605753639,
'www.stlouisfed.org': 0.8401103027746665,
'fredrikdeboer.com': 0.8400945631732266,
'blog.cerebralab.com': 0.8400945631732265,
'weworkremotely.com': 0.8400536676760625,
'english.yonhapnews.co.kr': 0.840015150270215,
'blog.ffwll.ch': 0.8399806882990191,
'www.kaspersky.com': 0.8399734036718564,
'webmasters.stackexchange.com': 0.8399733386768045,
'taoofmac.com': 0.8399516851214591,
'webapps.stackexchange.com': 0.8399426081446101,
'nenadseo.com': 0.839939548717707,
'our.status.im': 0.8399358470807224,
'justinhileman.info': 0.839916636052546,
'www.financemagnates.com': 0.83991429571072,
'srlabs.de': 0.8399137310646634,
'emacsredux.com': 0.8399137310646634,
'www.colorforth.com': 0.8398830755348644,
'tech.trivago.com': 0.8398706352344396,
'rusty.ozlabs.org': 0.8398201966467941,
'www.unionleader.com': 0.8398201966467941,
'tomverbeure.github.io': 0.8397956166500623,
'www.huckmag.com': 0.8397408630849317,
'www.3quarksdaily.com': 0.8397300193715044,
'snarky.ca': 0.8397285354135811,
'basicattentiontoken.org': 0.8396694214876033,
'0patch.blogspot.com': 0.8396569050806407,
'www.stackfield.com': 0.8396561419057332,
'laktek.com': 0.8395904864603353,
'www.elasticsearch.com': 0.8395809014444943,
'thomashunter.name': 0.8395739505236344,
'tapbots.com': 0.8395257149612579,
'nlp.stanford.edu': 0.8395127943859381,
'shime.sh': 0.8395015090274327,
'news.consumerreports.org': 0.8395015090274327,
'www.alwaysrightinstitute.com': 0.8395015090274327,
'plover.net': 0.8394877101221607,
'oleb.net': 0.8394619474972026,
'www.mymoneyblog.com': 0.8394511334746066,
'blog.clarifai.com': 0.8394429821320054,
'technodrone.blogspot.com': 0.8394039735099338,
'labs.earthpeople.se': 0.8394039735099338,
'blog.greweb.fr': 0.8394025195959639,
'truthout.org': 0.8393907588504428,
'www.rowetel.com': 0.8393890552964602,
'acha.ninja': 0.8393871342567361,
'blog.carlmjohnson.net': 0.8393408901985292,
'programming-journal.org': 0.8393406966484743,
'www.news18.com': 0.8392713736775148,
'sam.bluwiki.com': 0.8392570453444638,
'www.monbiot.com': 0.8392563174673074,
'rawgit.com': 0.8392483684869478,
'www.foxla.com': 0.8392193702289447,
'dvt.name': 0.839218830660763,
'blog.facebook.com': 0.8392045415231093,
'freddiedeboer.substack.com': 0.8391840290729782,
'developer.marvel.com': 0.8391376451077944,
'www.lowendtalk.com': 0.8391326019880078,
'docs.celeryproject.org': 0.8391326019880078,
'blog.trendmicro.com': 0.8391258095534089,
'poststatus.com': 0.8390975027931495,
'blockgeeks.com': 0.8390450127680085,
'itnext.io': 0.8390022169180034,
'blog.existentialize.com': 0.8389858546214449,
'semiengineering.com': 0.8389442107551947,
'www.digibarn.com': 0.8389114239700469,
'mikemcquaid.com': 0.8388881112064441,
'hackflow.com': 0.838888111206444,
'lists.opensuse.org': 0.8388704318936877,
'weblog.rogueamoeba.com': 0.8387522024258429,
'pubs.acs.org': 0.838739869285439,
'www.webdesignmuseum.org': 0.8387125258530213,
'www.garron.blog': 0.8387125258530211,
'instadeq.com': 0.8387105083280972,
'www.brisbanetimes.com.au': 0.8386591269613846,
'leanpub.com': 0.8386465804494829,
'fendrich.se': 0.8386023294509152,
'dl.dropboxusercontent.com': 0.8385995778248918,
'lettier.github.io': 0.8385983762141392,
'www.edn.com': 0.8384880352277337,
'www.synthtopia.com': 0.8384691370060022,
'www.michaelbromley.co.uk': 0.8384548675731243,
'www.refsmmat.com': 0.8384548675731243,
'jonbischke.com': 0.8384341839739798,
'warzel.substack.com': 0.8384049720140551,
'blogs.kqed.org': 0.8383873467097201,
'www.xconomy.com': 0.8383614008234231,
'www.timeslive.co.za': 0.838354450918394,
'keithp.com': 0.8383347979461153,
'psycnet.apa.org': 0.8383190430040641,
'blog.online.net': 0.8383007137970923,
'www.atariarchives.org': 0.8382952495330802,
'www.fivethirtyeight.com': 0.8382374049516083,
'www.wearethemighty.com': 0.8381915632322696,
'www.arstechnica.com': 0.8381868933365827,
'heatst.com': 0.8381531996802586,
'blog.wekeroad.com': 0.838151852850549,
'www.dnainfo.com': 0.8381308827626469,
'wufoo.com': 0.8381135600270289,
'www.zompist.com': 0.8381135600270289,
'www.nbr.co.nz': 0.8380751982509018,
'tomtunguz.com': 0.8380729812475479,
'www.scripting.com': 0.8380697700961443,
'blog.twilio.com': 0.8380613391841799,
'www.nhplace.com': 0.8380324040052766,
'pypi.python.org': 0.8380184467463045,
'www.the-digital-reader.com': 0.837988405339907,
'www.vpri.org': 0.8379878140892437,
'www.bailis.org': 0.8379777722457122,
'www.maximise.dk': 0.8379766121747607,
'99designs.com': 0.8379580943514392,
'www.crummy.com': 0.8379490514739079,
'www.ibtimes.co.uk': 0.8379434303523721,
'vmware.github.io': 0.8379411472179317,
'www.6sqft.com': 0.8379298178385524,
'symkat.com': 0.8379298178385524,
'survata.com': 0.8378674714439094,
'www.greenwichtime.com': 0.8378674714439094,
'www.leandomainsearch.com': 0.8378176898959185,
'penguindreams.org': 0.8377993988882627,
'www.cs.ox.ac.uk': 0.8377993988882627,
'windowsteamblog.com': 0.8377936722512888,
'www.more-magic.net': 0.8377744800755205,
'wundergraph.com': 0.8377666240744057,
'www.stereogum.com': 0.8377124380298397,
'media.defcon.org': 0.8376483665038642,
'andrewsullivan.theatlantic.com': 0.8376455603446734,
'physics.aps.org': 0.8376321975113651,
'www.addedbytes.com': 0.8376314906854643,
'theconversation.edu.au': 0.8376217948668653,
'carwoo.com': 0.8376024057546919,
'www.swiftbysundell.com': 0.8375837630694467,
'bigthink.com': 0.837574858033942,
'www.nltk.org': 0.8375336447475157,
'pron.github.io': 0.8375247229364985,
'atlas.cern': 0.8375209380234506,
'www.nyse.com': 0.8375209380234506,
'icyphox.sh': 0.8374905510071093,
'hypothesis.works': 0.8374147657218561,
'jacobobryant.com': 0.8373876028525398,
'blogs.zappos.com': 0.8373722614685953,
'blog.joellehman.com': 0.8373621224652893,
'conal.net': 0.8373554518624804,
'www.dadhacker.com': 0.837354651092982,
'stars.chromeexperiments.com': 0.8372729777719929,
'blog.thousandeyes.com': 0.8372663541460196,
'www.drbunsen.org': 0.8372619145116595,
'blog.tinfoilsecurity.com': 0.8372571688925584,
'www.landley.net': 0.837248322147651,
'jjbeshara.com': 0.837248322147651,
'www.searchenginejournal.com': 0.8372190177985843,
'samgentle.com': 0.8371607111617112,
'www.mywinamp.com': 0.8371394760092047,
'www.acm.uiuc.edu': 0.8371394760092047,
'blogs.sfweekly.com': 0.8371151654308293,
'theaviationist.com': 0.8370871690994972,
'www.mailchimp.com': 0.8370705380744307,
'incubator.apache.org': 0.8370705380744307,
'www.mikealrogers.com': 0.8370700724965489,
'www.unofficialgoogledatascience.com': 0.8370621601099476,
'www.wildbunny.co.uk': 0.8370316066816674,
'mega.co.nz': 0.8370228944247635,
'vimcasts.org': 0.8370122300530931,
'home.web.cern.ch': 0.8370107312738333,
'news.virginia.edu': 0.8370064211588852,
'blog.zawodny.com': 0.8370064211588852,
'www.mdpi.com': 0.8370060708284108,
'blog.mousereeve.com': 0.8370038016134355,
'hackernewsbooks.com': 0.8369747899159664,
'detexify.kirelabs.org': 0.8369455416288776,
'blip.tv': 0.8369060266959822,
'goodstuff.im': 0.8369038418527343,
'opensignal.com': 0.8368950022832518,
'the-pastry-box-project.net': 0.836851759413345,
'parkerhiggins.net': 0.8368481032461187,
'www.roughtype.com': 0.8367664523195075,
'www.ecosia.org': 0.8367628729886891,
'dougmccune.com': 0.8367628729886891,
'lambdaisland.com': 0.8367557134451551,
'blog.8thlight.com': 0.8367523868737199,
'mml-book.github.io': 0.8367337334597409,
'akaptur.com': 0.8367337334597409,
'www.appneta.com': 0.8367099124724148,
'eos.org': 0.836703598387807,
'www.eia.gov': 0.836692477143149,
'www.freesoftwaremagazine.com': 0.8366844191693678,
'blog.isquaredsoftware.com': 0.8366450083184942,
'www.effectgames.com': 0.8366086074590122,
'danbirken.com': 0.8366082851176284,
'www.georgesaines.com': 0.8365836924813026,
'projects.nytimes.com': 0.8365185959005205,
'cilium.io': 0.8364900150066928,
'luisant.ca': 0.8364627688697146,
'kde.org': 0.8364506685642159,
'gengo.ai': 0.8364249578414841,
'gigamonkeys.com': 0.836424957841484,
'ask.slashdot.org': 0.8364157461593826,
'www.inf.ethz.ch': 0.8364037125832664,
'bufr.tumblr.com': 0.8363012314554059,
'go.bloomberg.com': 0.8362895418561958,
'www.mv-voice.com': 0.8362735866170948,
'www.visual6502.org': 0.8362643122145562,
'codeofrob.com': 0.8362576149569876,
'www.ludumdare.com': 0.8362566416598247,
'www.bradford-delong.com': 0.8361844302416513,
'sidsite.com': 0.8361486486486487,
'persquaremile.com': 0.8361417911687711,
'boss.blogs.nytimes.com': 0.8361416007600664,
'pathdependent.com': 0.8361365464427095,
'www.marketplacepulse.com': 0.8361110009915448,
'javascriptweblog.wordpress.com': 0.8361100871908929,
'blog.steve.fi': 0.8360870917809617,
'www.stefankendall.com': 0.8360443900320628,
'machinelearningmastery.com': 0.8360325276933736,
'tomcritchlow.com': 0.8360317433744368,
'www.vqronline.org': 0.8359910491447964,
'pmarchive.com': 0.8359774363850249,
'blog.exodusintel.com': 0.8359613952379756,
'icelandmonitor.mbl.is': 0.8359181324679799,
'learnui.design': 0.8359165073352933,
'www.wisdomandwonder.com': 0.8359037098645399,
'gnupg.org': 0.8359037098645399,
'www.otherhand.org': 0.8359037098645399,
'andyswan.com': 0.8358651631273738,
'vlaaad.github.io': 0.8358605537700576,
'unikernel.org': 0.8358276165245331,
'forums.adobe.com': 0.8358049697804542,
'theappeal.org': 0.8358049697804542,
'people.mozilla.org': 0.8357546553679396,
'www.multivax.com': 0.8357546553679396,
'www.hilarymason.com': 0.8357367418987102,
'metro.co.uk': 0.8357169083210791,
'www.fxguide.com': 0.8356995496213405,
'mail.gnome.org': 0.8356930267331182,
'www.ripe.net': 0.8356373665440678,
'polygraph.cool': 0.8355932203389831,
'newsinteractives.cbc.ca': 0.8355673184633825,
'oonwoye.com': 0.8355085503926205,
'www.centernetworks.com': 0.83550554167365,
'blog.kenperlin.com': 0.8354986352909005,
'codelabs.developers.google.com': 0.8354986352909004,
'ericasadun.com': 0.8354812545307914,
'voyager.jpl.nasa.gov': 0.8354770866238536,
'glinden.blogspot.com': 0.8354248687848085,
'www.jeffknupp.com': 0.8354226452254534,
'0xax.blogspot.com': 0.8354177064677306,
'home.comcast.net': 0.8354135974937222,
'www.scrapingbee.com': 0.8353947775852016,
'mail-index.netbsd.org': 0.8353813194329008,
'mikeindustries.com': 0.835369856283584,
'failory.com': 0.8353532155275709,
'www.ritsumei.ac.jp': 0.835322744014879,
'typing.io': 0.8353140916808149,
'blog.britishmuseum.org': 0.835306193677583,
'mailbox.org': 0.8352951727923229,
'hackety.org': 0.8352886313345826,
'invisible-island.net': 0.8352861612933625,
'blog.minio.io': 0.8352821552252798,
'www.etymonline.com': 0.8352819967060172,
'founderdating.com': 0.8352750374933166,
'users.softlab.ece.ntua.gr': 0.8352418158169491,
'read.acloud.guru': 0.8352400454348197,
'julian.digital': 0.8352206968521787,
'ben.balter.com': 0.8352023832422831,
'osc.universityofcalifornia.edu': 0.8351585723230902,
'futureoflife.org': 0.8351040617937822,
'www.multicians.org': 0.8350943317132041,
'allthatsinteresting.com': 0.8350698686955953,
'machinethink.net': 0.8350562425214455,
'blogs.denverpost.com': 0.8350421047806545,
'mishadoff.com': 0.8350340136054422,
'getfirebug.com': 0.8350340136054422,
'www.dxomark.com': 0.8349344540359945,
'www.mycpu.org': 0.8349193509952372,
'www.raphkoster.com': 0.8349185464697736,
'bps-research-digest.blogspot.com': 0.8349065337932573,
'cs229.stanford.edu': 0.834894479023565,
'sadanand-singh.github.io': 0.834894479023565,
'blog.doordash.com': 0.834891053529973,
'paul.kinlan.me': 0.834891053529973,
'sliderulemuseum.com': 0.8347529812606473,
'www.hpcwire.com': 0.8347353682368306,
'flaviocopes.com': 0.8347098860524725,
'nocomplexity.com': 0.8346402286365225,
'designbygravity.wordpress.com': 0.8346002566095803,
'what-if.xkcd.com': 0.8345955143488654,
'www.healthcareitnews.com': 0.834479735234925,
'turingpi.com': 0.834470989761092,
'pfrazee.github.io': 0.834470989761092,
'csrc.nist.gov': 0.8344705397477052,
'taskwarrior.org': 0.8344665233968941,
'grepular.com': 0.8344590445370506,
'nuclearsecrecy.com': 0.8344590445370506,
'maxkle.in': 0.8344294161546372,
'www.csd.uwo.ca': 0.8344273945101189,
'alanhollis.com': 0.8343851939878961,
'www.splitgraph.com': 0.8343399060316927,
'www.instapainting.com': 0.834303164841344,
'taxfoundation.org': 0.834303164841344,
'delta.chat': 0.8342867534842825,
'appft1.uspto.gov': 0.8342760797997661,
'leerob.io': 0.8342207211363054,
'chrisshort.net': 0.8342193690645514,
'blog.venanti.us': 0.8342064686003738,
'news.cnrs.fr': 0.8341880341880342,
'500px.com': 0.8341880341880342,
'krugman.blogs.nytimes.com': 0.8341820699049042,
'fearlesssalarynegotiation.com': 0.8341803150678754,
'www.frankmcsherry.org': 0.8341803150678754,
'nts.strzibny.name': 0.8341550107611373,
'developers.memsql.com': 0.8341526449980673,
'sr.ht': 0.8341334625060556,
'ebiquity.umbc.edu': 0.8340723188782359,
'news.ubc.ca': 0.8340723188782359,
'blog.wired.com': 0.8340173587394528,
'www.dailyherald.com': 0.8340105184549977,
'artur-martsinkovskyi.github.io': 0.8339880299244583,
'felixge.de': 0.8339880299244583,
'blog.ayoungprogrammer.com': 0.833983000311883,
'inhabitat.com': 0.833970800806039,
'www.lifehacker.com.au': 0.8339654277683552,
'science.howstuffworks.com': 0.8338999781361612,
'cancer.nautil.us': 0.8338999781361612,
'appcubby.com': 0.8338994925108514,
'www.math.utah.edu': 0.8338684428078268,
'tooslowexception.com': 0.8338684428078268,
'www.patrickcraig.co.uk': 0.8338645320250127,
'perfectionkills.com': 0.8338619239102345,
'ncona.com': 0.8337668211914532,
'blogs.aws.amazon.com': 0.8337525489654461,
'www.cnil.fr': 0.83372186106167,
'guidovranken.wordpress.com': 0.8337106909166525,
'littlebitofcode.com': 0.8337106909166525,
'crpgaddict.blogspot.com': 0.8337028204720323,
'zerodha.tech': 0.833697181290741,
'medicine.wustl.edu': 0.8336839373019279,
'blog.socialcops.com': 0.8336751280750007,
'ifttt.com': 0.8336678541410665,
'blog.scalyr.com': 0.8335986795297188,
'www.psypost.org': 0.8335783172647908,
'www.abcactionnews.com': 0.8335754162799685,
'one.google.com': 0.8335754162799685,
'blog.marksweep.com': 0.8335690009897165,
'basecamp.com': 0.8334897559746374,
'www.rdegges.com': 0.8334643316971914,
'michael.otacoo.com': 0.8334458765497461,
'toucharcade.com': 0.8334457758738516,
'adage.com': 0.8334270713794664,
'getbootstrap.com': 0.8334213784102003,
'blog.klipse.tech': 0.8334108661664996,
'classics.mit.edu': 0.8334006037794777,
'dev.to': 0.8333163721881794,
'deadline.com': 0.833286267441488,
'www.expressvpn.com': 0.8332852925186139,
'zverovich.net': 0.8332852925186139,
'www.bnl.gov': 0.8332751437457405,
'discuss.linuxcontainers.org': 0.8332368170670513,
'codesketch.com': 0.8332263874970826,
'www.leveragedsellout.com': 0.8331869545898674,
'www.imore.com': 0.8331827749737907,
'people.inf.ethz.ch': 0.8331734250731533,
'blog.smartbear.com': 0.8331510235912066,
'ryanflorence.com': 0.8331396140981399,
'stephanpastis.wordpress.com': 0.8331396140981399,
'static1.squarespace.com': 0.8331073989685633,
'createdigitalmusic.com': 0.8330970061467121,
'bbc.co.uk': 0.8330970061467121,
'www.nhm.ac.uk': 0.8330761814047353,
'press.princeton.edu': 0.8330608346746611,
'blog.hartleybrody.com': 0.8329386129816697,
'futureboy.us': 0.8329182087040424,
'www.windowslatest.com': 0.8329104232060686,
'ewontfix.com': 0.8328938411174854,
'developer.yahoo.com': 0.8328635993581618,
'jng.imagine27.com': 0.8328607138329057,
'gogs.io': 0.8328565779067133,
'geoffboeing.com': 0.8328409513000423,
'www.phillymag.com': 0.8328409513000423,
'graphthinking.blogspot.com': 0.8328383114974501,
'blog.authy.com': 0.8327906592352505,
'earthengine.google.com': 0.8327019997862043,
'adamwathan.me': 0.8326583014571802,
'distributionhacks.com': 0.8326485954892083,
'www.thinkster.io': 0.832628817491836,
'terminusdb.com': 0.832628817491836,
'www.technoblogy.com': 0.8326183396075918,
'30hourjobs.com': 0.8326092279817636,
'infoproc.blogspot.com': 0.8326089175295298,
'www.ftrain.com': 0.8326051014052062,
'foundation.zurb.com': 0.8326012313075719,
'blog.viraptor.info': 0.8325776839853443,
'bennesvig.com': 0.8325751054908778,
'www.expressionsofchange.org': 0.832572579886013,
'mrl.nyu.edu': 0.8325111338579982,
'blog.urfix.com': 0.8325111338579982,
'lists.x.org': 0.8324697754749568,
'huggingface.co': 0.8324625696123126,
'zeuxcg.org': 0.8324088201420566,
'vmssoftware.com': 0.8323629518973035,
'jimplush.com': 0.8323629518973035,
'thereboot.com': 0.8323490243960362,
'antimatter15.com': 0.8323430057057912,
'japhr.blogspot.com': 0.8323430057057912,
'blog.davidecoppola.com': 0.8323083996896996,
'www.kotaku.com.au': 0.832301047102968,
'3perf.com': 0.8322876151248816,
'ryxcommar.com': 0.8322876151248816,
'www.mysql.com': 0.8322876151248816,
'scitechdaily.com': 0.8322404618162444,
'rjzaworski.com': 0.8321799307958477,
'www.dolthub.com': 0.8321769413258827,
'www.jperla.com': 0.8321740485435664,
'roberto.open-lab.com': 0.8321725984943663,
'www.tofugu.com': 0.8321694358359643,
'www.preprints.org': 0.8321312662446396,
'ntrs.nasa.gov': 0.8321306733695456,
'blog.kaggle.com': 0.8321136466040312,
'getcarina.com': 0.8321067502232443,
'thebarentsobserver.com': 0.8320962403151547,
'www.improbable.com': 0.8320824782702279,
'type.method.ac': 0.8320667644923518,
'www.aleph.se': 0.8320667644923518,
'lists.macosforge.org': 0.8320667644923518,
'zguide.zeromq.org': 0.8320667644923518,
'onethingwell.org': 0.8320539832149718,
'www.gittip.com': 0.8320539832149718,
'electronics.stackexchange.com': 0.8320387304939284,
'www.newsobserver.com': 0.8319542731572184,
'voat.co': 0.8319340573214568,
'www.viemu.com': 0.8319229644593481,
'www.embeddeduse.com': 0.8318890814558059,
'ideolalia.com': 0.831795173528973,
'andrewdumont.me': 0.8317884623663075,
'sriramk.com': 0.8317884108541596,
'www.chenhuijing.com': 0.8317790598787409,
'www.yegor256.com': 0.8317697748343722,
'www.foxnews.com': 0.8317694725922751,
'www.cs.nott.ac.uk': 0.8317650230135005,
'www.trustedreviews.com': 0.8317627240153616,
'greatergood.berkeley.edu': 0.8316786529650598,
'mitxela.com': 0.8315323559022839,
'mail.haskell.org': 0.8315275148175146,
'cooking.stackexchange.com': 0.8315230832695719,
'www.apnews.com': 0.831500326107284,
'www.belfasttelegraph.co.uk': 0.8314967853302809,
'zackshapiro.com': 0.8314155294643537,
'www.collectspace.com': 0.831412959256786,
'www.alchemistowl.org': 0.8313334448247652,
'www.interfluidity.com': 0.8313143760797148,
'mithril.js.org': 0.8313043478260869,
'sunboxlabs.com': 0.8313043478260869,
'css.csail.mit.edu': 0.83128825969837,
'mattermark.com': 0.8312791828318407,
'ballotpedia.org': 0.8312622600459071,
'r2c.dev': 0.8312622600459071,
'goaccess.io': 0.8312325519112657,
'www.michaelwest.com.au': 0.8312257534350801,
'redsymbol.net': 0.8312144135803776,
'journals.aps.org': 0.8312128897645714,
'blog.gingerlime.com': 0.8312016352070194,
'bugsplat.info': 0.8311332389568854,
'projects.aljazeera.com': 0.8310960202717635,
'jlouisramblings.blogspot.com': 0.8310960202717635,
'www.datastuff.tech': 0.8310748264197387,
'www.cliffc.org': 0.8310367707450238,
'www.merriam-webster.com': 0.8310169102957669,
'washingtonpost.com': 0.831005263332199,
'www.leapmotion.com': 0.8309966909104267,
'dom.as': 0.8309526362308279,
'dotnet.github.io': 0.8309273722835752,
'rickcarlino.com': 0.8309273722835752,
'p-cos.blogspot.com': 0.8309273722835752,
'www.electronicsweekly.com': 0.8308205769875683,
'blog.mythic-beasts.com': 0.8308141567715057,
'blog.codiqa.com': 0.830787883150054,
'www.urbandictionary.com': 0.8307663266042938,
'blog.ericgoldman.org': 0.8307656053312806,
'thatcomputerscientist.com': 0.8307366610510188,
'shankman.com': 0.830715532286213,
'blog.uservoice.com': 0.830715532286213,
'shafyy.com': 0.8306912467549404,
'www.23andme.com': 0.8306912467549404,
'thebulletin.org': 0.8306726473588931,
'sekao.net': 0.8306717904781525,
'om.co': 0.8306518197608382,
'reacttraining.com': 0.8306279342292152,
'phili.pe': 0.8306279342292152,
'kamerontanseli.ghost.io': 0.8306279342292152,
'www.deccanherald.com': 0.830617757462069,
'recipes.hypotheses.org': 0.8306075161940402,
'www.scotthyoung.com': 0.8304676761207413,
'countercomplex.blogspot.com': 0.8304599554901336,
'blog.kraken.com': 0.8304102297264507,
'ello.co': 0.8303950986221522,
'asiatimes.com': 0.8303938402412525,
'blog.tindie.com': 0.8303900100100166,
'www.plagiarismtoday.com': 0.8303836069407874,
'moviecode.tumblr.com': 0.8303351870840227,
'www.the74million.org': 0.8303274336465168,
'analytics.usa.gov': 0.8303274336465168,
'blog.jpl-consulting.com': 0.8303274336465168,
'archive.nytimes.com': 0.8303195537893885,
'blogs.zdnet.com': 0.8302477853461946,
'www.gosquared.com': 0.8302108084053713,
'www.unite.ai': 0.8301971117330208,
'pastie.org': 0.8301429578448887,
'www.bluetooth.com': 0.830122591943958,
'heartbeat.fritz.ai': 0.8300819862294267,
'blog.liw.fi': 0.8300810779081896,
'gilesbowkett.blogspot.com': 0.8300257142091928,
'placesjournal.org': 0.8299577016504602,
'interfacelab.com': 0.8299038213762403,
'programming-motherfucker.com': 0.8298839065342439,
'www.thespacereview.com': 0.8298526044878857,
'www.zippia.com': 0.8298245614035087,
'alumnit.ca': 0.8298220650253975,
'pne.people.si.umich.edu': 0.8297232221938442,
'blog.scienceexchange.com': 0.8296882900501562,
'danielallendeutsch.com': 0.8296786018063418,
'veyronb.wordpress.com': 0.8296786018063417,
'www.learningclojure.com': 0.829663604832329,
'android.stackexchange.com': 0.8296351227828602,
'neo4j.com': 0.8296343605300973,
'www.insentricity.com': 0.8295390134146845,
'johnhawks.net': 0.8295371897944813,
'gist.io': 0.829536713446392,
'www.irishexaminer.com': 0.8295100407790094,
'blog.runnable.com': 0.8294913771537328,
'www.dmv.ca.gov': 0.8294663396318804,
'technology.stitchfix.com': 0.8293836920149785,
'anyasq.com': 0.8293440267345259,
'eigenfoo.xyz': 0.8293416074439832,
'sunlightfoundation.com': 0.8293282465454183,
'developer.rackspace.com': 0.829323262325674,
'rwmj.wordpress.com': 0.8293063560783351,
'www.makegameswith.us': 0.8292959190283576,
'wesleytansey.com': 0.8292534470294121,
'startupljackson.com': 0.8292534470294121,
'www.xprize.org': 0.8292534470294121,
'www.mongodb.org': 0.829231224668068,
'www.drive.com.au': 0.8292253521126759,
'crate.io': 0.8291626909612831,
'voice.kadira.io': 0.8291212925208525,
'blog.drewolson.org': 0.8291191739941484,
'slog.thestranger.com': 0.8291191739941484,
'remarkedly.com': 0.8291054023142576,
'artem.krylysov.com': 0.8290877591855479,
'www.collabora.com': 0.8290844858349273,
'www.handelsblatt.com': 0.8290824405762608,
'www.dailycamera.com': 0.8289847063149993,
'jozefg.bitbucket.org': 0.8289755956167212,
'baus.net': 0.8289755956167212,
'www.davisr.me': 0.8289669222429755,
'ghc.haskell.org': 0.8289669222429755,
'akkshaya.blog': 0.8289669222429755,
'www.boristhebrave.com': 0.8289241622574957,
'ootbcomp.com': 0.8289241622574957,
'www.puzzlescript.net': 0.8289241622574957,
'mattjibson.com': 0.828841660397834,
'knowledge.wharton.upenn.edu': 0.8288260425307535,
'simpleflying.com': 0.8288152386804701,
'houseoflawandorder.com': 0.828808793096398,
'abstractions.nautil.us': 0.8288076574107112,
'www.scalescale.com': 0.8288034171529194,
'www.ostechnix.com': 0.8288034171529194,
'www.londonreconnections.com': 0.8288034171529193,
'sandro.volpee.de': 0.8287903687677997,
'www.reghardware.com': 0.8287468808977058,
'www.shellcheck.net': 0.8287381145882802,
'm.youtube.com': 0.8287237652975437,
'www.freep.com': 0.828700914495102,
'www.cs.hmc.edu': 0.8286794342224705,
'www.post-gazette.com': 0.8286406919749909,
'ecorner.stanford.edu': 0.8285837667040196,
'olimex.wordpress.com': 0.8285738879971565,
'englishrussia.com': 0.8285720380368012,
'2ton.com.au': 0.8285684110847549,
'debugandrelease.blogspot.com': 0.8285505526982948,
'news.err.ee': 0.8285383023940318,
'academictimes.com': 0.8285283281973459,
'beyondgrep.com': 0.8285017969437128,
'www.lesspenguiny.com': 0.8285017969437128,
'tinyhack.com': 0.8285017969437128,
'www.mailgun.com': 0.8284973918318256,
'homepage.divms.uiowa.edu': 0.8284973918318256,
'grimoire.ca': 0.82849280304493,
'unicornfree.com': 0.8284746725029384,
'www.afr.com': 0.8284438035910985,
'www.routefifty.com': 0.8284215095159949,
'www.nintendolife.com': 0.8284071908632212,
'www.wolfram.com': 0.8283831975537147,
'www.rfi.fr': 0.8282942879074087,
'talkweb.eu': 0.8282723807936619,
'drdobbs.com': 0.8282680775958894,
'elidourado.com': 0.8282580762478886,
'www.math.uwaterloo.ca': 0.8282472101761524,
'www.industryweek.com': 0.8282421484626746,
'tipjoys2cents.blogspot.com': 0.8282133103461377,
'www.prerender.cloud': 0.8282133103461377,
'www.thezdi.com': 0.8282032202463764,
'www.ableton.com': 0.8281936977442643,
'kylerush.net': 0.8281936977442643,
'hbswk.hbs.edu': 0.8281850054053886,
'www.hopesandfears.com': 0.8281077808394027,
'mappingignorance.org': 0.8281015489846159,
'code.jsoftware.com': 0.8281015489846159,
'thefrailestthing.com': 0.828083220780241,
'blog.gopheracademy.com': 0.8280826702520344,
'threedots.tech': 0.8280375017895765,
'robnapier.net': 0.8280259986658036,
'blogs.nasa.gov': 0.8280019811960149,
'blog.indutny.com': 0.8279912351566244,
'cloudonaut.io': 0.8279487990535627,
'boston.craigslist.org': 0.827928286178696,
'www.techprowd.com': 0.827928286178696,
'smalldata.tech': 0.8279186302269647,
'notebook.drmaciver.com': 0.8278936902761703,
'www.getpostman.com': 0.8278936902761703,
'newcriterion.com': 0.82785314452846,
'www.nationalaffairs.com': 0.8278141715678232,
'roarmag.org': 0.8278111419376111,
'blog.ezliu.com': 0.82779905891824,
'engineering.heroku.com': 0.82779905891824,
'sulami.github.io': 0.8277548514011955,
'techdirt.com': 0.8277129992323573,
'www.automobilemag.com': 0.827708703374778,
'lihautan.com': 0.8275741663396405,
'aschmann.net': 0.8275741663396404,
'blog.apps.npr.org': 0.8275741663396404,
'ionicframework.com': 0.8275048014166265,
'google-engtools.blogspot.com': 0.8274021352313168,
'blog.davidsingleton.org': 0.8274021352313168,
'blog.jessitron.com': 0.8273903856669967,
'www.orau.org': 0.8272273741633687,
'digesttime.com': 0.8272248458013906,
'kswizz.com': 0.8272248458013906,
'wmpoweruser.com': 0.8272247723472747,
'nimbleindustries.io': 0.8272187229694837,
'jolieodell.wordpress.com': 0.8272099836896732,
'www.bookforum.com': 0.8271999089538162,
'genode.org': 0.8271844999040995,
'decorrespondent.nl': 0.82714602901724,
'theorangeduck.com': 0.8271367134087778,
'blog.redfin.com': 0.8271155479281936,
'threema.ch': 0.8270778121654588,
'blog.gaborcselle.com': 0.8270675104289051,
'www.eastbayexpress.com': 0.8270623051198946,
'www.nilsjuenemann.de': 0.8270167583183482,
'www.ausgamers.com': 0.8269871316556712,
'www.nickkolenda.com': 0.8269846287715326,
'freakonomics.blogs.nytimes.com': 0.8269510780284808,
'datadome.co': 0.8269501507198006,
'www.matusiak.eu': 0.8269501507198006,
'marksaroufim.substack.com': 0.8269362550900476,
'kernel.org': 0.8269362550900476,
'foaas.com': 0.8268624447770888,
'www.quiverquant.com': 0.8268223600931086,
'www.pennmedicine.org': 0.8267857142857142,
'panamapapers.icij.org': 0.8267786682199767,
'rare-technologies.com': 0.8267627529251013,
'www.wusa9.com': 0.826717577802525,
'rustwasm.github.io': 0.8266828193880259,
'exupero.org': 0.8266797238058068,
'nodesource.com': 0.8266756851479922,
'tryapl.org': 0.8266364460873337,
'insights.sei.cmu.edu': 0.8266146306841327,
'zeynep.substack.com': 0.8265133053820785,
'adamcaudill.com': 0.8265133053820785,
'flexport.engineering': 0.8265065205596224,
'theneurosphere.com': 0.8264888595136076,
'news.rice.edu': 0.8264758779932855,
'blog.ninlabs.com': 0.8264758497316637,
'blog.snapchat.com': 0.8264601098042835,
'twistedoakstudios.com': 0.8264580233843033,
'qvault.io': 0.8264470673072151,
'zachaysan.tumblr.com': 0.826417360607573,
'investigations.nbcnews.com': 0.8264128301922102,
'ourmachinery.com': 0.8263907059824316,
'www.fs.blog': 0.8263508838430117,
'www.linuxatemyram.com': 0.8263506013533878,
'blog.kraih.com': 0.8263506013533877,
'blog.nyaruka.com': 0.8263442628141022,
'jeffdechambeau.com': 0.8263442628141021,
'opensoul.org': 0.8263238944048071,
'www.cortes.design': 0.8263165312463789,
'www.pawfal.org': 0.8263046319865304,
'blog.typesafe.com': 0.8262863127448572,
'blog.andrewcantino.com': 0.8262623863759415,
'www.boeing.com': 0.8262623863759415,
'www.ted.com': 0.8262579048280261,
'blog.custora.com': 0.8262292214279253,
'www.naftaliharris.com': 0.8261791079056829,
'javascriptkicks.com': 0.8261791079056829,
'the-toast.net': 0.8261672942679482,
'www.cs.rit.edu': 0.8261648745519713,
'jliszka.github.io': 0.8261587974287634,
'm.economictimes.com': 0.8261099137432224,
'vessenes.com': 0.8260709719869568,
'www.ineteconomics.org': 0.826029111542426,
'blogs.unity3d.com': 0.8260137334026378,
'0fps.net': 0.8260060814624561,
'justcramer.com': 0.8260060814624561,
'www.nest.com': 0.8260056123053906,
'qualiacomputing.com': 0.8259333270861942,
'improbable.io': 0.8259007499406081,
'blog.dilbert.com': 0.8258693250944908,
'standpointmag.co.uk': 0.8258674439632754,
'nelsonslog.wordpress.com': 0.8258527827648116,
'www.pixelmator.com': 0.8258442158455263,
'www.c2.com': 0.8258100134548192,
'www.heraldscotland.com': 0.8257871536416392,
'yuzu-emu.org': 0.8257634254559968,
'www.physicscentral.com': 0.8257504892739511,
'www.overclock3d.net': 0.8257367279398189,
'www.crainsnewyork.com': 0.8257322176521333,
'www.dtc.umn.edu': 0.8256884706720989,
'livefreeordichotomize.com': 0.8256884706720989,
'graphics.wsj.com': 0.8256623456620759,
'lostechies.com': 0.8256485195117381,
'weblog.bocoup.com': 0.8256267961489369,
'www.rechargenews.com': 0.8256135856550654,
'weatherspark.com': 0.8256135856550654,
'wiki.apache.org': 0.8256007740509997,
'ceylon-lang.org': 0.8256007740509996,
'androidrookies.com': 0.8255966724327264,
'nwn.blogs.com': 0.8255570752578437,
'nathany.com': 0.8255395683453237,
'www.deepdotweb.com': 0.8255335258309922,
'jonathanischwartz.wordpress.com': 0.8254547893724321,
'www.pitheringabout.com': 0.8254520771980598,
'steve.yegge.googlepages.com': 0.8254511675613948,
'www.useronboard.com': 0.8253939552044646,
'www.kjmagnetics.com': 0.8253701708120639,
'ftp.openbsd.org': 0.8253198620399047,
'www.bogleheads.org': 0.8253198620399047,
'v5.chriskrycho.com': 0.8253173977170429,
'jordanmechner.com': 0.8253173977170429,
'fare.tunes.org': 0.8253125699065577,
'www.dangermouse.net': 0.8252739167424069,
'pdf.yt': 0.82525521399989,
'robbreport.com': 0.8252506344809964,
'insideevs.com': 0.8252342355806511,
'www.logolounge.com': 0.8252252252252251,
'ftp.arl.army.mil': 0.8252200375406061,
'www.linux.com': 0.8252089691299165,
'www.csoonline.com': 0.8251776577952398,
'catapult.co': 0.8251768351081376,
'blog.webkid.io': 0.8251768351081376,
'www.uwgb.edu': 0.8251526412738834,
'girinstud.io': 0.8251526412738834,
'adilmoujahid.com': 0.8251526412738834,
'www.rntz.net': 0.8251450579360111,
'www.smore.com': 0.8251318689390142,
'news.rthk.hk': 0.8251206340702621,
'www.c-span.org': 0.825114692730766,
'm.stopa.io': 0.8250423585067803,
'www.opb.org': 0.8250415548188467,
'artofmanliness.com': 0.825002300673692,
'opendylan.org': 0.825001402919603,
'tailwindcss.com': 0.8249955991764671,
'youtube-creators.googleblog.com': 0.8249955991764671,
'www.plosone.org': 0.824994642431906,
'la.curbed.com': 0.8249814854575737,
'gfycat.com': 0.8249792022510777,
'pressfreedomfoundation.org': 0.8249170036751421,
'www.tested.com': 0.8249099846980117,
'blog.gruntwork.io': 0.8248984430563132,
'www.innoq.com': 0.8248966445207588,
'granolashotgun.com': 0.8248948672452628,
'articles.moneycentral.msn.com': 0.8248521762231498,
'www.bu.edu': 0.8247468947911291,
'mozillalabs.com': 0.8247395166074101,
'thinkgrowth.org': 0.8247377515142629,
'soledadpenades.com': 0.8247300709391115,
'snips.ai': 0.8247300709391115,
'insightdataengineering.com': 0.8247228989993084,
'altdevblogaday.com': 0.8247202423028511,
'bugs.webkit.org': 0.8246896410441447,
'www.serpentine.com': 0.8246896410441446,
'crimereads.com': 0.8246832358182735,
'www.chadfowler.com': 0.8246492171080979,
'flow.microsoft.com': 0.8245931283905966,
'muut.com': 0.8245391490982966,
'deque.blog': 0.8245391490982966,
'www.nearform.com': 0.8245170505651781,
'www.us-cert.gov': 0.82449055422799,
'regexcrossword.com': 0.8244429611663081,
'chris-lamb.co.uk': 0.824442961166308,
'unretrofied.com': 0.8244331063039999,
'www.rsaconference.com': 0.8244327243272294,
'www.nhregister.com': 0.8244082580729432,
'cryoshon.co': 0.8244082580729432,
'www.dpmms.cam.ac.uk': 0.8243716227838693,
'www.patreon.com': 0.8243522520227725,
'salon.thefamily.co': 0.8243182801287786,
'www.leakedsource.com': 0.8242753623188406,
'www.clicktorelease.com': 0.824270209428051,
'www.chipworks.com': 0.824267686738341,
'www.hardkernel.com': 0.824267686738341,
'news.slashdot.org': 0.8242672968534664,
'developer.atlassian.com': 0.8242608280205023,
'www.top500.org': 0.8242567265672777,
'technicshistory.com': 0.8242340010424594,
'segment.io': 0.8242217790014779,
'mitsloan.mit.edu': 0.8242217790014779,
'thebuild.com': 0.8242049759936478,
'thistooshallgrow.com': 0.8241685536402895,
'alliance.nautil.us': 0.8241562516385773,
'www.slashgeek.net': 0.8241560987787111,
'www.zeroequalsfalse.press': 0.8241315599752687,
'poincare101.blogspot.com': 0.8240928369315027,
'blog.zarfhome.com': 0.8240852612726679,
'www.ryanpickren.com': 0.8240852612726679,
'design.canonical.com': 0.8240707368381324,
'www.theengineer.co.uk': 0.8240609124764349,
'the-magazine.org': 0.8239745714397421,
'dennybritz.com': 0.8239663447195944,
'ben.casnocha.com': 0.8239663447195944,
'www.alexirpan.com': 0.8239445529818444,
'www.siliconvalleywatcher.com': 0.8239203009772859,
'chill.com': 0.8239075286809479,
'kanoki.org': 0.8238968676022924,
'knowablemagazine.org': 0.8238968676022923,
'www.fun-mooc.fr': 0.8238950612054784,
'www.seas.upenn.edu': 0.823854197776325,
'longbets.org': 0.8238504937824159,
'papers.nips.cc': 0.8238165389819486,
'dyske.com': 0.8237859610530819,
'i.gizmodo.com': 0.8237724868134527,
'www.kth.se': 0.8237610739927905,
'www.tawheedkader.com': 0.8237541123759499,
'www.gov.uk': 0.8237306525116473,
'chsasank.github.io': 0.8237179544778666,
'plasmasturm.org': 0.8236804766955088,
'blog.backtype.com': 0.8236678521506691,
'feaneron.com': 0.8236399179105635,
'tech.backtype.com': 0.8236399179105635,
'www.annashipman.co.uk': 0.8236363636363635,
'sive.rs': 0.8235995425409803,
'keen.io': 0.8235778391019275,
'wensing.tumblr.com': 0.8234963171167604,
'www1.informatik.uni-erlangen.de': 0.8234356896394766,
'www.ewanleith.com': 0.8233853975881945,
'tech.puredanger.com': 0.8233619779868596,
'freshman.tech': 0.823356964994172,
'weakdh.org': 0.8233151183970855,
'boundary.com': 0.8233084145427331,
'taitems.tumblr.com': 0.8232972027089648,
'www2.eecs.berkeley.edu': 0.823294055901832,
'arvindn.livejournal.com': 0.8232708263906926,
'pcmonk.me': 0.8232633368501505,
'www.wfla.com': 0.8232633368501505,
'blog.adamnash.com': 0.8232369516538023,
'english.martinvarsavsky.net': 0.8232086570141259,
'www.usv.com': 0.8231252352764534,
'federalnewsnetwork.com': 0.8231197830462293,
'www.theatlas.com': 0.8231091015701053,
'www.riknieu.com': 0.8231091015701053,
'www.iterm2.com': 0.8231091015701053,
'www.itwire.com': 0.8231086108233077,
'www.standalone-sysadmin.com': 0.8231078625216017,
'www.autoblog.com': 0.8230993668082046,
'pv-magazine-usa.com': 0.8230980746736871,
'blog.andrewvc.com': 0.8230893291673793,
'www.mappingthejourney.com': 0.8230893291673792,
'blogs.s-osg.org': 0.8230893291673792,
'www.erica.biz': 0.8230405980636032,
'digitallife.today.msnbc.msn.com': 0.822992700729927,
'humantransit.org': 0.822978047436078,
'anaxi.com': 0.8229457720254757,
'bedrocklinux.org': 0.8229141632870333,
'sina.is': 0.8229141632870332,
'www.joachim-breitner.de': 0.8229052530564972,
'o.canada.com': 0.8228104673631248,
'michaelrbernste.in': 0.8228054773077562,
'status.digitalocean.com': 0.8227743399626116,
'www.computerworlduk.com': 0.8227710906704893,
'tech.newstatesman.com': 0.8227353406126373,
'abc7chicago.com': 0.8227318903356561,
'bytes.yingw787.com': 0.8226344289193692,
'www.mikekarnj.com': 0.822628692450413,
'blogarithms.github.io': 0.8226216439253611,
'designinstruct.com': 0.8226216439253611,
'fsecurify.com': 0.8225830523375955,
'deepsource.io': 0.8225746395058767,
'n-gate.com': 0.8225612808865542,
'www.25hoursaday.com': 0.8225607212300213,
'global.handelsblatt.com': 0.8225208403926864,
'pragli.com': 0.8225208403926864,
'manuel.bernhardt.io': 0.8225011685480498,
'blog.jooq.org': 0.8224250424416801,
'www.portfolio.com': 0.8224243429262191,
'web.nvd.nist.gov': 0.8224221426834478,
'www.hn-books.com': 0.8223443223443224,
'felixpalmer.github.io': 0.8223443223443223,
'weev.livejournal.com': 0.8223443223443223,
'www.harpersbazaar.com': 0.822307206378181,
'snapframework.com': 0.822307206378181,
'www.hussmanfunds.com': 0.822307206378181,
'benlog.com': 0.8222776118268347,
'www.fluentcpp.com': 0.8222033505775027,
'www.toptal.com': 0.8221835594409774,
'www.openssh.com': 0.8221823092541788,
'blog.avast.com': 0.8221738853904804,
'www.timesofisrael.com': 0.822154529331881,
'www.tomdalling.com': 0.8221220478981254,
'matloff.wordpress.com': 0.8221023761491743,
'www.stuff.co.nz': 0.8220743929973073,
'heather.cs.ucdavis.edu': 0.8220584263105031,
'theincidentaleconomist.com': 0.8220326740109917,
'radiapp.com': 0.8220183486238531,
'blog.ramen.is': 0.8219861931408673,
'geti2p.net': 0.8219861931408673,
'bgr.com': 0.8218680813488343,
'hey.com': 0.821866426984301,
'33bits.org': 0.8218275173853028,
'www.pressherald.com': 0.8218104832596405,
'www.staycaffeinated.com': 0.8217905775466726,
'www.igorkromin.net': 0.8217742285549731,
'zackarymorris.tumblr.com': 0.8217716107859097,
'seattletimes.com': 0.8217659740535453,
'news.northwestern.edu': 0.8217641342619865,
'canny.io': 0.8217480223498653,
'www.modernhealthcare.com': 0.8217164092599815,
'about.google': 0.8217164092599815,
'www.ifweassume.com': 0.8216730309542566,
'mobile.abc.net.au': 0.8216657761831399,
'blog.readme.io': 0.8216640179430849,
'www.bschaatsbergen.com': 0.8216640179430849,
'bendyworks.com': 0.8216400745937131,
'randywestergren.com': 0.8215693482108614,
'www.crn.com.au': 0.8215655177172531,
'yinwang0.wordpress.com': 0.8215439095876336,
'substack.net': 0.8215269428046106,
'www.toothycat.net': 0.8214683749692366,
'vega.github.io': 0.8214578695182713,
'pixelscommander.com': 0.8214578695182713,
'www.thinq.co.uk': 0.8214038185099727,
'www.huffingtonpost.com.au': 0.8213863864440609,
'www.depesz.com': 0.8213734717251845,
'www.w2lessons.com': 0.8213669213667488,
'www.mostlymaths.net': 0.8213616628443653,
'putanumonit.com': 0.8213391337481347,
'www.executeprogram.com': 0.8213391337481347,
'www.computing.co.uk': 0.8213046552380585,
'eblong.com': 0.8212918616104282,
'asylum.madhouse-project.org': 0.8212918616104282,
'www.granolashotgun.com': 0.8212918616104282,
'www.redditblog.com': 0.8212710455497659,
'www.causal.app': 0.821219763084809,
'valerieaurora.org': 0.8212075934194915,
'www.owasp.org': 0.8211882087756882,
'petition.parliament.uk': 0.8211640347006609,
'engineering.columbia.edu': 0.821142906452761,
'fastml.com': 0.8211297895691131,
'nerderati.com': 0.8211239168679134,
'www.nutsvolts.com': 0.8211239168679134,
'vine.co': 0.821112630700933,
'www.grobmeier.de': 0.8211091773217756,
'html.adobe.com': 0.8211013298757308,
'coursetro.com': 0.8210338632520735,
'www.rleonardi.com': 0.8210332103321032,
'make.wordpress.org': 0.8210118109358497,
'iq.opengenus.org': 0.8209763562067588,
'tech.slashdot.org': 0.820954477342236,
'www.stanforddaily.com': 0.8209221578417347,
'www.mapd.com': 0.8208973498293703,
'campaignstops.blogs.nytimes.com': 0.8208298386512924,
'www.theblaze.com': 0.8208160487478949,
'codingkaiser.blog': 0.8208103239764403,
'cushychicken.github.io': 0.8208101603109114,
'techblog.badoo.com': 0.8207887125985329,
'arcanesentiment.blogspot.com': 0.8207861411585611,
'www.lawfareblog.com': 0.8207748819367903,
'taskandpurpose.com': 0.8207638339548695,
'programmingdesignsystems.com': 0.8207024029574862,
'www.robots.ox.ac.uk': 0.8206399949682861,
'us.pycon.org': 0.820630666964755,
'market.android.com': 0.8206208283828789,
'blog.kirigin.com': 0.8205932101360961,
'www.pragprog.com': 0.820579551832526,
'blog.perfectaudience.com': 0.8205466327449291,
'mreinhold.org': 0.820546632744929,
'www.thoughtworks.com': 0.8205459808249582,
'blog.frankel.ch': 0.8205171785173707,
'philipnilsson.github.io': 0.8205088545817216,
'www.deathandtaxesmag.com': 0.8203995498364007,
'7news.com.au': 0.8203793607701599,
'keminglabs.com': 0.8203793607701598,
'research.securitum.com': 0.8203793607701598,
'blog.nanonets.com': 0.8203793607701598,
'www.roguelynn.com': 0.8203703703703703,
'zenhack.net': 0.8203703703703703,
'corpgov.law.harvard.edu': 0.8203690845488739,
'thelongandshort.org': 0.8203387274635711,
'bramcohen.livejournal.com': 0.820330588464305,
'www.nyu.edu': 0.82033015672031,
'howtosplitanatom.com': 0.8203248926942794,
'losangeles.cbslocal.com': 0.8202840702703775,
'www.freebsdfoundation.org': 0.8202742664017543,
'pqina.nl': 0.8202180020912505,
'ramblemaps.com': 0.820186718359544,
'blog.httpwatch.com': 0.8201707118020548,
'aroberge.blogspot.com': 0.8201707118020547,
'joanwestenberg.medium.com': 0.8201707118020546,
'www.texastribune.org': 0.820127140783887,
'vagosec.org': 0.8201145209512104,
'blog.michielborkent.nl': 0.8200731194854671,
'www.desmogblog.com': 0.8200731194854671,
'informationarchitects.net': 0.8200687812178955,
'www.mpg.de': 0.8200224577197879,
'devcenter.heroku.com': 0.8200215315397114,
'blog.hackerrank.com': 0.8199539191124647,
'www.mailboxapp.com': 0.8199115962474446,
'libregraphicsworld.org': 0.8198809873858414,
'www.netmarketshare.com': 0.8198634237702345,
'www.honeycomb.io': 0.8198446827660495,
'wwwnc.cdc.gov': 0.8197995558341037,
'davidhauser.com': 0.8197995558341036,
'www.philanthropy.com': 0.8197755193144188,
'www.armstrongeconomics.com': 0.8197658321751845,
'www.techpowerup.com': 0.8197581725949339,
'fourweekmba.com': 0.8197401870662347,
'runwayml.com': 0.8197062084117198,
'cs.uwaterloo.ca': 0.8197032651831856,
'jeffarchibald.ca': 0.8196469792108286,
'monograph.io': 0.8196424057593232,
'jaredforsyth.com': 0.8196424057593232,
'rework.withgoogle.com': 0.8196231659693813,
'www.butwhatfor.com': 0.8195815182778377,
'webglfundamentals.org': 0.8195389645544945,
'skepchick.org': 0.8195389645544945,
't37.net': 0.8195018817046663,
'blockstack.org': 0.8195018817046663,
'www.ticalc.org': 0.8195018248001847,
'konghq.com': 0.8194820842106035,
'cseweb.ucsd.edu': 0.8194210882358477,
'www.orlandosentinel.com': 0.8194144516611968,
'ansuz.sooke.bc.ca': 0.8194061161585031,
'platform.uno': 0.8193808271817485,
'svs.gsfc.nasa.gov': 0.8193808271817484,
'i-kh.net': 0.819366852886406,
'www.t3x.org': 0.8193537789546194,
'thinkingagriculture.io': 0.8193537789546194,
'www.colorado.edu': 0.8192948515552042,
'philipkiely.com': 0.8192790918819292,
'theultralinx.com': 0.8192615833622455,
'the-witness.net': 0.8192531710522758,
'www.woot.com': 0.8192133344078468,
'blog.crowdtilt.com': 0.8192133344078468,
'docs.python-guide.org': 0.8192031083028783,
'blog.indextank.com': 0.819186830384332,
'bfi.uchicago.edu': 0.8191137736578458,
'theirrelevantinvestor.com': 0.8190345082551709,
'herman.asia': 0.8190298507462687,
'quartzy.qz.com': 0.8189961869849737,
'www.peterbe.com': 0.8189793251113663,
'shakycode.com': 0.8189532189180286,
'jeffreykegler.github.io': 0.8189309404959616,
'programmer.97things.oreilly.com': 0.8189307834762161,
'spectrum.chat': 0.8189034014413787,
'alugy.com': 0.8189014793607566,
'www.modern.ie': 0.8188594320310705,
'ilovetypography.com': 0.8188542303813234,
'parametric.press': 0.8188376402212918,
'blog.staffjoy.com': 0.8187965320542665,
'grahamcluley.com': 0.8187959495636524,
'benediktmeurer.de': 0.8187751653221643,
'www.loggly.com': 0.818750224724795,
'typophile.com': 0.8187110947432705,
'themorningnews.com': 0.8187110947432705,
'blog.estimote.com': 0.8187092985024379,
'www.nola.com': 0.8187092491447127,
'blog.fanout.io': 0.8187064944602657,
'theoreticalminimum.com': 0.8186915887850467,
'snap.stanford.edu': 0.8186423752137325,
'serhack.me': 0.8186312306298296,
'digital.nhs.uk': 0.8186026992823302,
'investor.yahoo.net': 0.8185979779164885,
'www.gamingonlinux.com': 0.818574619301555,
'interactions.acm.org': 0.8185585358755773,
'htmlpreview.github.io': 0.8185193011459522,
'codeberg.org': 0.8185193011459521,
'www.arm.com': 0.8184839406823776,
'www.fedscoop.com': 0.8184720810751717,
'video.google.com': 0.8184638061300447,
'eyeondesign.aiga.org': 0.8184578256865039,
'vihart.com': 0.8184565521589099,
'history.nasa.gov': 0.8184538842053415,
'www.webmonkey.com': 0.8184424171038894,
'www.pacifict.com': 0.8184091497388042,
'www.anaconda.com': 0.8184074729195554,
'jamesclear.com': 0.8183928113741262,
'images.guide': 0.8183520599250935,
'sandofsky.com': 0.818333099426741,
'tessrinearson.com': 0.8182949341417263,
'mysqlserverteam.com': 0.8182817858828835,
'www.virology.ws': 0.8182801242351887,
'transparencyreport.google.com': 0.8182801242351887,
'b.lesseverything.com': 0.8182544712962565,
'forklog.media': 0.8182524048349719,
'mega.nz': 0.8182293546514141,
'postcards.blogs.fortune.cnn.com': 0.8181713824074244,
'kevinrose.com': 0.8181713824074243,
'www.sankalpjonna.com': 0.8181713824074243,
'nick.typepad.com': 0.8181333642385573,
'mathbabe.org': 0.8180898841225986,
'hacksoflife.blogspot.com': 0.8180862740536983,
'nickmchardy.com': 0.8180611629393052,
'www.spectrumnews.org': 0.8180586243621752,
'www.fictiv.com': 0.8180206950485711,
'www.riskbasedsecurity.com': 0.8179622866340774,
'juliagalef.com': 0.8178989768184225,
'wheels.blogs.nytimes.com': 0.8178917680568788,
'medium.learningbyshipping.com': 0.8178382028024892,
'travel.stackexchange.com': 0.8178270639852792,
'developer.yahoo.net': 0.8177152660164336,
'www.lostbookofsales.com': 0.8177143703947889,
'blog.rainforestqa.com': 0.8176913530816636,
'www.zeldman.com': 0.8176675642405572,
'www.shogan.co.uk': 0.8176447363707375,
'blog.friendfeed.com': 0.8176243006331482,
'tech.instacart.com': 0.8176187126924949,
'www.well-typed.com': 0.8176129455942598,
'madebynathan.com': 0.8175850766542176,
'www.dam.brown.edu': 0.8175673958400209,
'www.electrive.com': 0.8175415128966471,
'open.bekk.no': 0.8174977146526387,
'launch.is': 0.8174957585928975,
'mindyourdecisions.com': 0.8174283379047065,
'avc.blogs.com': 0.8174249012744215,
'underscore.io': 0.8174038787709598,
'beakerbrowser.com': 0.8173898378490387,
'tibleiz.net': 0.8173670622724108,
'learnlayout.com': 0.8173662532870822,
'statescoop.com': 0.8173342969093829,
'pcottle.github.io': 0.8173258003766477,
'mkaz.blog': 0.8173258003766477,
'davmac.wordpress.com': 0.817306487680897,
'khalilstemmler.com': 0.817286266376623,
'mobile.slate.com': 0.817274377154757,
'pepijndevos.nl': 0.8172690940621161,
'en.arguman.org': 0.8172440600460996,
'dougseven.com': 0.8172346051319319,
'www.hsj.co.uk': 0.8172346051319318,
'www.washingtoncitypaper.com': 0.8172227432631624,
'www.stefanjudis.com': 0.8172227432631624,
'remotehabits.com': 0.8170840840435508,
'pawelurbanek.com': 0.8170758370632349,
'www.linode.com': 0.8170518536342932,
'mattrogish.com': 0.8170279345181233,
'www.dataorienteddesign.com': 0.8170279345181233,
'vtorosyan.github.io': 0.8170279345181232,
'www.adapteva.com': 0.8170168040129737,
'waldo.jaquith.org': 0.8170168040129737,
'www.sentient.ai': 0.8169811320754717,
'cheeaun.com': 0.8169811320754716,
'adam.chlipala.net': 0.8169734054322659,
'www.fastmail.com': 0.8169005980617469,
'law.duke.edu': 0.8169005980617468,
'uglyduck.ca': 0.8167837603451541,
'labs.sentinelone.com': 0.8167337370996636,
'whoisnnamdi.com': 0.8166875449806325,
'eagain.net': 0.8166739998002963,
'news.columbia.edu': 0.8166739998002963,
'bibwild.wordpress.com': 0.8166660149109494,
'docs.fedoraproject.org': 0.8166606678373952,
'branchfree.org': 0.8166606678373952,
'www.howmanypeopleareinspacerightnow.com': 0.8166606678373952,
'devhints.io': 0.8166486478801723,
'http.cat': 0.8166351606805292,
'www.aspi.org.au': 0.8166351606805292,
'lalo.li': 0.8166351606805292,
'forum.linode.com': 0.8166342281215091,
'www.cyanogenmod.com': 0.8166342281215091,
'www.math.brown.edu': 0.8166342281215091,
'embracethered.com': 0.8166342281215091,
'insights.stackoverflow.com': 0.8166112476989804,
'codemanship.co.uk': 0.816610401586209,
'bombthrower.com': 0.8165653679484848,
'open.dgraph.io': 0.8165613796925617,
'dlewis.net': 0.8165199873064167,
'www.infoworld.com': 0.8165194544092945,
'simonsfoundation.org': 0.8164711754414633,
'blog.evanweaver.com': 0.8164587254389714,
'webdocs.cs.ualberta.ca': 0.8164587254389714,
'runnable.com': 0.8164587254389714,
'scienceblog.com': 0.816442930863004,
'theinvisiblethings.blogspot.com': 0.8164118858386505,
'caterina.net': 0.8163684777302764,
'www.cockeyed.com': 0.8163493619189814,
'garry.tv': 0.8163458866047911,
'tech.mattmillman.com': 0.8163458866047911,
'pavelfatin.com': 0.8163391912303228,
'blog.testmunk.com': 0.8163326429746816,
'righteousruminations.blogspot.com': 0.8163302805339576,
'playnice.ly': 0.8163302805339576,
'thedriven.io': 0.8162915628964128,
'liorpachter.wordpress.com': 0.8162914041012925,
'rr-project.org': 0.8162878787878787,
'vis4.net': 0.8162878787878787,
'purelyfunctional.tv': 0.8162705368577864,
'cachestocaches.com': 0.816247477572823,
'www.novelr.com': 0.8162474775728229,
'www.myasmine.com': 0.8162289080621473,
'www.hanshq.net': 0.8162230743510456,
'urbankchoze.blogspot.com': 0.8160645118710772,
'powazek.com': 0.8160565844295424,
'www.mjt.me.uk': 0.8160235722456507,
'www.loc.gov': 0.8160198829440672,
'www.businessinsider.fr': 0.8160110236421557,
'benmetcalfe.com': 0.816002952282756,
'cloudant.com': 0.8159956502678918,
'buddycloud.com': 0.8159660889619967,
'devinhelton.com': 0.8159603862822139,
'blog.onedrive.com': 0.8159603862822139,
'hackoftheday.securitytube.net': 0.8159603862822139,
'internetintel.oracle.com': 0.8159603862822139,
'news.rpi.edu': 0.8159324993062729,
'kyso.io': 0.8157856217718672,
'blog.cryptoaustralia.org.au': 0.8157807107640108,
'www.pixelmonkey.org': 0.8157110214426208,
'gbracha.blogspot.com': 0.8156837220936173,
'pandoralive.info': 0.8156756405717832,
'www.lessmilk.com': 0.8156734149395104,
'blog.everyblock.com': 0.815665920751951,
'moyix.blogspot.com': 0.815665920751951,
'blog.keep.network': 0.815665920751951,
'www.dpreview.com': 0.8156119793953887,
'blog.calbucci.com': 0.8156110920088071,
'kowainik.github.io': 0.8156055311358869,
'blogs.discovery.com': 0.8156055311358867,
'flic.kr': 0.8155522718024564,
'ledgersmbdev.blogspot.com': 0.8155503540881356,
'www.bostonmagazine.com': 0.8155484449427585,
'www.haydenplanetarium.org': 0.815539827164826,
'lukesmith.xyz': 0.815539827164826,
'blog.djy.io': 0.815539827164826,
'www.treasury.gov': 0.8154535511908948,
'abovethecrowd.com': 0.815439760100364,
'yogthos.net': 0.8154331327745159,
'www.adn.com': 0.8153813333062851,
'www.eveonline.com': 0.8153685129167869,
'blog.metamorphium.com': 0.8153509023913048,
'dannyvankooten.com': 0.8152380952380952,
'www.creativereview.co.uk': 0.8152232350076929,
'blog.hasura.io': 0.8152023189276763,
'artsbeat.blogs.nytimes.com': 0.8151914351876436,
'dosync.posterous.com': 0.8151891543888505,
'openreview.net': 0.8151568208322646,
'blog.serverdensity.com': 0.8151341964704665,
'manybutfinite.com': 0.8151251257545465,
'mars.nasa.gov': 0.8150920162802555,
'www.rbth.com': 0.8150730163714182,
'www.fefe.de': 0.8150624797096998,
'mcla.ug': 0.8150272592774294,
'chem.tufts.edu': 0.8150272592774294,
'code.jquery.com': 0.8149967373525299,
'tantek.com': 0.8149534354896242,
'www.grahamcluley.com': 0.8149532812632393,
'query.nytimes.com': 0.814942807315685,
'greig.cc': 0.8149045173924782,
'www.xaprb.com': 0.8149019506115348,
'badssl.com': 0.814885496183206,
'bbirnbaum.com': 0.8148706344450946,
'officialandroid.blogspot.com': 0.8147398134191081,
'ruudvanasseldonk.com': 0.8147309156410547,
'www.softwareatscale.dev': 0.8146603203498196,
'heap.engineering': 0.8146603203498196,
'videlalvaro.github.io': 0.8146574608017414,
'norswap.com': 0.8146006935103474,
'reactionwheel.net': 0.8145996500684587,
'bitslog.wordpress.com': 0.8145723063065177,
'rasterize.io': 0.8145327073791039,
'adlervermillion.com': 0.8145327073791038,
'timeless.judofyr.net': 0.81453154875717,
'haskell.org': 0.8145294619307533,
'githut.info': 0.8145279229959786,
'www.gadgette.com': 0.8145279229959786,
'frantic.im': 0.8145279229959786,
'www.creativeapplications.net': 0.8145169207397831,
'googlecommerce.blogspot.com': 0.8144066006321738,
'mikeash.com': 0.8144025820168224,
'wecode.wepay.com': 0.8143827137547628,
'www.cs.ucdavis.edu': 0.8143765574898738,
'data.heapanalytics.com': 0.8143226776111164,
'blog.mixergy.com': 0.8143177092759811,
'www.sudo.ws': 0.8142855202937566,
'simplystatistics.org': 0.814273938401625,
'pbpython.com': 0.81426991310501,
'www.andrew.cmu.edu': 0.814256651162313,
'www.techinasia.com': 0.8142474634536047,
'www.scirra.com': 0.8142144867968754,
'mrmacintosh.com': 0.8142070774434265,
'omarabid.com': 0.8142001608362681,
'db.cs.cmu.edu': 0.8141839403388533,
'reactkungfu.com': 0.8141839403388533,
'paulbridger.com': 0.8141723199425726,
'openbazaar.org': 0.8141723199425726,
'www.globalpost.com': 0.8141091737204239,
'david-smith.org': 0.8140528179043202,
'www.philipotoole.com': 0.8140231382115423,
'20bits.com': 0.8139880326617631,
'www.kickscondor.com': 0.813983802425237,
'blog.zdsmith.com': 0.813920131333506,
'www.iclarified.com': 0.8139139706368135,
'www.x-plane.com': 0.8138853765840854,
'mirage.io': 0.8138853765840854,
'blog.expensify.com': 0.8138812059202145,
'www.iso.org': 0.8138437732374225,
'cryptoservices.github.io': 0.8138386793877036,
'hedgehogreview.com': 0.8138372159261252,
'www.simple.com': 0.8138043652042445,
'ftp.cs.ucla.edu': 0.8137212614300459,
'blog.laptopmag.com': 0.8137088732053118,
'cliffmass.blogspot.com': 0.8136763300597577,
'kubuszok.com': 0.8136519481838238,
'whatblag.com': 0.813578419801053,
'insight.kellogg.northwestern.edu': 0.8135739399890168,
'skerritt.blog': 0.8135660771798076,
'www.shacknews.com': 0.8135400926338027,
'www.hcn.org': 0.8134966474909895,
'www.desmoinesregister.com': 0.8134966474909894,
'www.intelligentliving.co': 0.81349213300375,
'www.huffpost.com': 0.8134799974360535,
'nextbigfuture.com': 0.8134623168672729,
'www.arneswinnen.net': 0.8134615384615383,
'www.bellingcat.com': 0.8134267051774864,
'www.zverovich.net': 0.8133918753159222,
'blog.delicious.com': 0.8133799785061123,
'hackthesystem.com': 0.8133799785061122,
'blog.comae.io': 0.8133732585740193,
'altdevblogaday.org': 0.8133173831188817,
'w2.eff.org': 0.8132979261023464,
'news.fastcompany.com': 0.8132816312546639,
'blog.dominodatalab.com': 0.8132112268692228,
'www.burlingtonfreepress.com': 0.8131021194605009,
'betterwebtype.com': 0.8130859124005918,
'tech.yipit.com': 0.8130859124005917,
'www.capitolhillseattle.com': 0.8130859124005917,
'flyosity.com': 0.8130320794442989,
'juffalow.com': 0.8130140383642517,
'www.alexstjohn.com': 0.8130023652493408,
'awardwinningfjords.com': 0.8129597143337574,
'humanbioscience.org': 0.8129551673058468,
'programmersatwork.wordpress.com': 0.8129191502536695,
'www.gao.gov': 0.8129073172130272,
'www.sci-news.com': 0.8129054883394705,
'www.teslarati.com': 0.8128722926988359,
'www.creativedeletion.com': 0.8128708612164178,
'hushkit.net': 0.8128656957185499,
'www.sketch.com': 0.8127222898502466,
'0xfe.blogspot.com': 0.8127222898502466,
'www.governmentattic.org': 0.8127166548773207,
'www.iflscience.com': 0.8127104066471765,
'www.bmwblog.com': 0.8127009540598622,
'thephd.github.io': 0.8126088980839451,
'papercups.io': 0.8125137829701504,
'deeplearning4j.org': 0.8125126453331284,
'www.kaiostech.com': 0.8125018296149806,
'www.kdab.com': 0.8124430720535263,
'communities.intel.com': 0.8123620795777623,
'www.knowablemagazine.org': 0.8123571766847992,
'sci-hub.tw': 0.8122620383047426,
'flink.apache.org': 0.8122287873617684,
'www.mobilecrunch.com': 0.8120912628571249,
'blog.liangzan.net': 0.8120625991864463,
'paulcavallaro.com': 0.8120625991864463,
'www.freebsdnews.net': 0.8120155038759689,
'www.cs.dartmouth.edu': 0.8120120009628422,
'slashdot.org': 0.8120103261102206,
'beta.theglobeandmail.com': 0.8119800311478199,
'www.franzoni.eu': 0.8119518701628052,
'www.unz.com': 0.8119446655347015,
'launch.gust.com': 0.8119242878514453,
'mattlakeman.org': 0.8119242878514452,
'us.battle.net': 0.8119135744478586,
'vitess.io': 0.8119046881511972,
'www.parsonsmatt.org': 0.8118880860804719,
'www.sciam.com': 0.8118415645513156,
'blog.inklingmarkets.com': 0.8118015537952591,
'darknetdiaries.com': 0.8117495396709252,
'blog.mixpanel.com': 0.8117419261805633,
'blog.kippt.com': 0.8117273158415922,
'danielrapp.github.io': 0.8117273158415922,
'semantic-domain.blogspot.com': 0.8117250261807765,
'blog.jayfields.com': 0.8117106131262737,
'dronegarageblog.wordpress.com': 0.8116970637651796,
'm.cnet.com': 0.8116871715030578,
'bionic.fullstory.com': 0.8116504854368931,
'blog.reybango.com': 0.8116192339099034,
'baesystemsai.blogspot.com': 0.8116098514216743,
'tasvideos.org': 0.811592254952584,
'www.nathancunn.com': 0.8115765934419358,
'theguardian.com': 0.8115466314959758,
'www.bls.gov': 0.8114586352368218,
'www.slashgear.com': 0.8114130074807454,
'get21stnight.com': 0.8113811919256128,
'wadefoster.net': 0.8113380585952003,
'backstage.soundcloud.com': 0.8113341783615344,
'blog.mongohq.com': 0.8113084373038951,
'blog.samwhited.com': 0.8113018683008136,
'lukethomas.com': 0.8112900298646131,
'johnmurray.io': 0.811284046692607,
'www.scpr.org': 0.8111931888951548,
'serverless.com': 0.8111853186864186,
'www.mintpressnews.com': 0.8111566169738434,
'dev.hubspot.com': 0.8111491117505176,
'uxdesign.smashingmagazine.com': 0.8110875951026609,
'www.dcscience.net': 0.8110637872870197,
'www.badlogicgames.com': 0.8110531473954321,
'www.imdb.com': 0.8109585365005275,
'www.kevinwilliampang.com': 0.810916179337232,
'benjaminmayo.co.uk': 0.8109161793372319,
'www.tonymacx86.com': 0.8109161793372319,
'news.firedoglake.com': 0.8109161793372319,
'gitorious.org': 0.810892038104006,
'www.indystar.com': 0.810892038104006,
'www.nextbigwhat.com': 0.810871453085769,
'vectorized.io': 0.8108646859647678,
'www.fakesteve.net': 0.8108603160330469,
'theness.com': 0.8108582902726974,
'blog.zenkit.com': 0.8108296140970532,
'consequenceofsound.net': 0.8108001038237775,
'dealbook.blogs.nytimes.com': 0.810782053555595,
'raccoon.onyxbits.de': 0.8107413934848161,
'www.linaro.org': 0.8107386685587079,
'insights.hpe.com': 0.8107316248294252,
'webflow.com': 0.8107174188583165,
'insidesearch.blogspot.com': 0.8107157266121297,
'www.hhs.gov': 0.8107124152561004,
'blog.folyo.me': 0.8107124152561003,
'blog.playcanvas.com': 0.8107042145283994,
'oldblog.antirez.com': 0.8106726118909099,
'computinged.wordpress.com': 0.8106676355919645,
'www.beekeeperstudio.io': 0.8106491457950028,
'www.meetup.com': 0.8106205529212165,
'capiche.com': 0.8105820355692206,
'www.cbp.gov': 0.810546875,
'stackexchange.com': 0.8105386561469894,
'kapeli.com': 0.8105257538297364,
'blog.sigfpe.com': 0.8105035346007113,
'nianticlabs.com': 0.8104898882641829,
'econlog.econlib.org': 0.8104880381821702,
'bitsofco.de': 0.8104598795589895,
'blog.gyrosco.pe': 0.8104246070876013,
'blog.n01se.net': 0.8104076949009623,
'mozakai.blogspot.com': 0.8104076949009623,
'iamwil.posterous.com': 0.8103954830327119,
'www.minyanville.com': 0.8103877873546691,
'jcjc-dev.com': 0.8103364623358701,
'decryptmedia.com': 0.8103177363469551,
'alexvermeer.com': 0.8103141664126979,
'reddragdiva.dreamwidth.org': 0.8103141664126978,
'anarchogeek.com': 0.8103141664126978,
'truthonthemarket.com': 0.810302799232476,
'numenta.com': 0.8102602204436639,
'economictimes.indiatimes.com': 0.8102491144464327,
'www.perceptualedge.com': 0.8102043958285661,
'www.garykessler.net': 0.8101761252446185,
'loopj.com': 0.8101761252446184,
'newteevee.com': 0.8101251805166595,
'morningconsult.com': 0.8101075221559366,
'mrmrs.io': 0.810089337360651,
'www.springboard.com': 0.8100885519221457,
'www.computinghistory.org.uk': 0.8100885519221456,
'ownlocal.com': 0.810072359778719,
'wellcomecollection.org': 0.810072359778719,
'owasp.org': 0.810072359778719,
'linuxreviews.org': 0.8100650707969348,
'monadical.com': 0.8100469597795646,
'gamesfromwithin.com': 0.810042091205529,
'www.geeksforgeeks.org': 0.8100336615814749,
'www.universal-sci.com': 0.810032793083701,
'esoteric.codes': 0.8099527252346412,
'www.petermichaud.com': 0.8099337587386388,
'www.freedom-to-tinker.com': 0.8099164380793796,
'www.laptopmag.com': 0.809846769190124,
'www.3news.co.nz': 0.8098352603019351,
'sel4.systems': 0.8098186580704161,
'www.accountingtoday.com': 0.8098186580704161,
'host.madison.com': 0.8097889409717836,
'www.nokia.com': 0.8097879208159734,
'insanecoding.blogspot.com': 0.8097776259075163,
'kdenlive.org': 0.8097696203436352,
'www.bitstamp.net': 0.8097578373340156,
'stonemaiergames.com': 0.809735836328836,
'ceronman.com': 0.809735818851372,
'airmail.news': 0.809735818851372,
'nickwignall.com': 0.8097127006730107,
'www.develop-online.net': 0.8096968183358267,
'brunocalza.me': 0.809690221614063,
'gadgetopia.com': 0.8096902216140629,
'seekingalpha.com': 0.8096704138265437,
'www.jofreeman.com': 0.8095931882025065,
'muckandbrass.com': 0.8095931882025065,
'nobelprize.org': 0.8095392106610299,
'www.vg247.com': 0.8095156620373399,
'chetansurpur.com': 0.8094671848396732,
'haxiomic.github.io': 0.8094631273918713,
'deislabs.io': 0.8094631273918712,
'jmc.stanford.edu': 0.8094565130568238,
'philpapers.org': 0.8094356549031609,
'youtube-global.blogspot.com': 0.8094272365570714,
'rubygems.org': 0.8093966911744022,
'www.laquadrature.net': 0.8093644943397358,
'dheera.net': 0.80935652930136,
'use.perl.org': 0.8093278422294106,
'dannorth.net': 0.8093207850460444,
'orbitalindex.com': 0.8093095444635583,
'historyofyesterday.com': 0.8092882173909588,
'www.citymetric.com': 0.809287154904464,
'www.codecommit.com': 0.8092420697380182,
'danlec.com': 0.8092306399056751,
'endlessparentheses.com': 0.8092053968063221,
'www.vrfinal.com': 0.809205396806322,
'www.the-american-interest.com': 0.8091888493515856,
'cycle-gap.blogspot.com': 0.8091399291892002,
'elinux.org': 0.8091062649443477,
'agtb.wordpress.com': 0.8091062649443477,
'addyosmani.github.com': 0.8090929406134104,
'www.immadsnewworld.com': 0.8090591990898237,
'interplayoflight.wordpress.com': 0.8090551181102362,
'kylecordes.com': 0.8090551181102361,
'www.ndss-symposium.org': 0.8090548860707052,
'www.grantland.com': 0.808953355501278,
'accu.org': 0.8089301368051649,
'americans.org': 0.8089046493675752,
'languagengine.co': 0.8088694832763874,
'www.al-monitor.com': 0.8088667083388086,
'www.marksimonson.com': 0.8088667083388086,
'danilocampos.com': 0.8088667083388086,
'money.yahoo.com': 0.8088648827181377,
'www.mathpages.com': 0.8088432529287259,
'nakamotoinstitute.org': 0.8088145420012367,
'blog.ionelmc.ro': 0.8087651491972426,
'www.strawpoll.me': 0.8087480632308696,
'www.theweek.co.uk': 0.8087282943708363,
'www.morningbrew.com': 0.8086875731228074,
'ksj.mit.edu': 0.8086785009861932,
'ozar.me': 0.808655053694382,
'scheme2006.cs.uchicago.edu': 0.8085933953841323,
'9fans.net': 0.8085933953841322,
'pythonhosted.org': 0.8085610188346061,
'lite.cnn.com': 0.8085535287872542,
'www.fark.com': 0.808501385570164,
'www.techendo.co': 0.808501106941333,
'downlode.org': 0.8084934128064108,
'upstart.me': 0.8084934128064108,
'www.extendslogic.com': 0.8084934128064108,
'theundefeated.com': 0.8084934128064108,
'blog.appsignal.com': 0.8084690154513154,
'www.informationarchitects.jp': 0.8084214332325731,
'www.intrade.com': 0.8083885146980836,
'lukasz.langa.pl': 0.8083885146980836,
'www.energy.gov': 0.8083804906443909,
'talkpython.fm': 0.8083525540804629,
'yourbasic.org': 0.8083074906530393,
'www.sachinagarwal.com': 0.8083003952569169,
'openshift.redhat.com': 0.8082657786610448,
'tech.adroll.com': 0.8082443572105479,
'paleofuture.gizmodo.com': 0.8082365397238809,
'tgceec.tumblr.com': 0.808209532747702,
'bowery.io': 0.8081978171238972,
'blog.sourcing.io': 0.8081838340004474,
'www.9to5mac.com': 0.8081802724018405,
'bostinno.com': 0.8081755973845505,
'jatan.space': 0.8081563211953493,
'www-inst.eecs.berkeley.edu': 0.8081204961840847,
'evrone.com': 0.8080872755977898,
'labs.domipheus.com': 0.8080351711070707,
'www.thrillist.com': 0.8079971777160608,
'www.gizmag.com': 0.8079969849430149,
'landley.net': 0.807958662672192,
'danweinreb.org': 0.8078992257179908,
'tyk.io': 0.807867976487508,
'mymicrocontroller.com': 0.8078630033918272,
'www.samharris.org': 0.8078422754090425,
'epetitions.direct.gov.uk': 0.8078284811604556,
'blog.tildeloop.com': 0.8078284811604556,
'discuss.emberjs.com': 0.8078241273575438,
'blog.superuser.com': 0.8077665343961988,
'oliveremberton.com': 0.807743372835657,
'vpnpro.com': 0.8077402589309914,
'www.lastwordonnothing.com': 0.8076974958304752,
'stevenf.com': 0.8076913832175415,
'www.timeanddate.com': 0.8076143611045982,
'cornellsun.com': 0.8075830492659702,
'letoverlambda.com': 0.8075830492659702,
'snapcraft.io': 0.8075475712589127,
'www.observer.com': 0.8074847637579465,
'unit42.paloaltonetworks.com': 0.8074697991147913,
'www.fastcocreate.com': 0.8074665167071722,
'mixmax.com': 0.8074624950006974,
'blog.goodaudience.com': 0.8074490566695363,
'wikitravel.org': 0.8074371698943806,
'elzr.com': 0.8074371698943806,
'gerger.co': 0.8074101861989343,
'developer.amazonwebservices.com': 0.8073933592944413,
'blog.postmarkapp.com': 0.8073171397946091,
'adereth.github.io': 0.8073171397946091,
'www.gamepro.com': 0.8073017117935154,
'www.sencha.com': 0.8072902381119292,
'christianheilmann.com': 0.8072902381119292,
'mg.co.za': 0.8072653978420726,
'python-history.blogspot.com': 0.8072408370446265,
'www.warhistoryonline.com': 0.807215873070334,
'theatlantic.com': 0.8072127719392547,
'sakurity.com': 0.8071790086703337,
'blog.yalabot.com': 0.8071726013094649,
'softwareengineering.stackexchange.com': 0.8071626259596408,
'vnhacker.blogspot.com': 0.8071372108129627,
'www.securityweek.com': 0.8070728244513765,
'www.myce.com': 0.8070371527879234,
'progressiveapp.store': 0.8070260217041482,
'teamblog.supportbee.com': 0.8070260217041482,
'arogozhnikov.github.io': 0.8070036402591226,
'mechanical-sympathy.blogspot.com': 0.8069161416177959,
'strangelyconsistent.org': 0.8069045196582395,
'www.themarysue.com': 0.8068976247535697,
'www.attendly.com': 0.8068580186201768,
'www.snell-pym.org.uk': 0.8068326941714653,
'tools.cisco.com': 0.8068196782315848,
'influxdb.com': 0.8068173283593989,
'worksinprogress.co': 0.8067896834805538,
'www.scripps.edu': 0.8067572864866587,
'dissenter.firedoglake.com': 0.806652591404526,
'crackstation.net': 0.8066141922329747,
'www.afar.com': 0.8065122291007617,
'hasura.io': 0.8064237491553132,
'www.cs.nyu.edu': 0.8063872255489021,
'www.mtholyoke.edu': 0.8063872255489021,
'blog.startifact.com': 0.8063862184736535,
'www.colourlovers.com': 0.8063834547609555,
'www.gettingmoreawesome.com': 0.8063764986530164,
'www.themorningnews.org': 0.8063013263733552,
'hardcoresoftware.learningbyshipping.com': 0.8062797912662047,
'bramlambrecht.com': 0.8062668907413534,
'www.spacesafetymagazine.com': 0.8062339137045579,
'newscenter.berkeley.edu': 0.8062219609407214,
'suckless.org': 0.8062207668887819,
'www.staticgen.com': 0.8062025100696604,
'www.eecs.umich.edu': 0.8062025100696603,
'www.thefader.com': 0.8061953047624655,
'www.independent.ie': 0.806109251382116,
'pivotallabs.com': 0.8060806747033802,
'www.opsian.com': 0.8060621471224426,
'www.taigeair.com': 0.8060621471224426,
'talkingpointsmemo.com': 0.8060591184167385,
'williampross.com': 0.8060455947863886,
'popmotion.io': 0.8060424719294661,
'robb.weblaws.org': 0.806042471929466,
'globaldev.co.uk': 0.806011328959103,
'pencerw.com': 0.806,
'michelenasti.com': 0.8059842345365967,
'sourcesort.com': 0.8059044933660855,
'www.monash.edu': 0.8059044933660855,
'haydenjames.io': 0.8058768485324959,
'stjepang.github.io': 0.805873630277718,
'blog.webb.page': 0.805873630277718,
'tcfast.com': 0.8058521366516195,
'blog.figma.com': 0.8058473182210699,
'blog.fikesfarm.com': 0.8058387400788318,
'www.telecompaper.com': 0.8058387400788318,
'mikegerwitz.com': 0.8058387400788318,
'micheleincalifornia.blogspot.com': 0.8058373166902851,
'cyber.harvard.edu': 0.8058229699826737,
'content.time.com': 0.8057897338858303,
'catless.ncl.ac.uk': 0.8057745286288139,
'www.techflash.com': 0.8057273188584126,
'omis.me': 0.805697502622664,
'anildash.com': 0.8056599652743452,
'www.authy.com': 0.8056112224448897,
'bitterwinter.org': 0.8056112224448897,
'news.illinois.edu': 0.8055525618659666,
'trilema.com': 0.8055277480254573,
'nshipster.com': 0.8055021399615895,
'uppy.io': 0.8054909385894538,
'probmods.org': 0.805478770001181,
'labs.umbrella.com': 0.805478770001181,
'www.lawyersgunsmoneyblog.com': 0.8054692502161738,
'paulspontifications.blogspot.com': 0.8054688521990506,
'wiki.python.org': 0.8054642788538674,
'metaphysicaldeveloper.wordpress.com': 0.8054364779428177,
'tech.okcupid.com': 0.805426569995148,
'leejo.github.io': 0.8054207149429934,
'scholar.google.com': 0.8053870344669396,
'jseliger.com': 0.8053643358322942,
'www.openrightsgroup.org': 0.8053456370357672,
'zendev.com': 0.8053227485814811,
'deepsense.ai': 0.8053174136210413,
'googlepublicpolicy.blogspot.com': 0.8052966184857965,
'reading.supply': 0.8052571979264512,
'www.fullstackpython.com': 0.8052571979264512,
'making.duolingo.com': 0.8052499518418211,
'cloudrail.com': 0.8052345044507019,
'crt.sh': 0.8052208835341366,
'charity.wtf': 0.8052128017402501,
'gettingreal.37signals.com': 0.8052119053726816,
'products.office.com': 0.805195056791058,
'blog.spiderlabs.com': 0.8051909890788225,
'javascript.christmas': 0.8051778217183054,
'www.ruby-forum.com': 0.8051778217183053,
'webkul.github.io': 0.8051216600975734,
'crypt.codemancers.com': 0.8050979962669772,
'aomedia.org': 0.805084051401519,
'www.physics-astronomy.org': 0.8050038695306542,
'dothemath.ucsd.edu': 0.8049264940426683,
'www.av8n.com': 0.8049169449005759,
'signalsandthreads.com': 0.8049169449005759,
'www.apptorium.com': 0.8048961063441398,
'smarterware.org': 0.8048459862314374,
'www.playstation.com': 0.8048413390083552,
'toys.lerdorf.com': 0.8048289738430583,
'blog.handcraft.com': 0.8047698525990352,
'www.alex-ionescu.com': 0.8047611301824745,
'team.adzerk.com': 0.8047527420132489,
'www.trottercashion.com': 0.8047172511803042,
'www.dylanpaulus.com': 0.8047172511803041,
'shiroyasha.io': 0.8046842108009492,
'sunlightlabs.com': 0.8046689741257063,
'johnbender.us': 0.8046551925330852,
'www.orionmagazine.org': 0.8046409236882213,
'500ish.com': 0.8045980649418749,
'mbostock.github.com': 0.8045837440928983,
'www.eclogiselle.com': 0.8045630694400487,
'j00ru.vexillium.org': 0.8045036919905235,
'www.partow.net': 0.8045036919905235,
'www.zpub.com': 0.8045036919905235,
'www.hq.nasa.gov': 0.8043661879315178,
'news.usni.org': 0.804355836696792,
'charemza.name': 0.8043090754458252,
'gondor.io': 0.8043052663473409,
'gamelab.mit.edu': 0.8042844920724259,
'starcraft.com': 0.8042844920724259,
'sorentwo.com': 0.804261705269134,
'thebrowser.com': 0.8042503564660758,
'ericsink.com': 0.804229811533429,
'blogs.loc.gov': 0.8041942737849643,
'sslmate.com': 0.8041845438238772,
'googlereader.blogspot.com': 0.8041760226307137,
'pingcap.github.io': 0.8041760226307136,
'blog.nearlyfreespeech.net': 0.8041648746111758,
'blog.peer5.com': 0.8041599344447704,
'understandlegacycode.com': 0.8041241866706869,
'www3.imperial.ac.uk': 0.8040897254741548,
'ovid.github.io': 0.8040060267347496,
'perell.com': 0.8039678014067241,
'marketingland.com': 0.8039660516562604,
'maierfelix.github.io': 0.803954084248265,
'www.plainsite.org': 0.803949941652702,
'www.fabcapo.com': 0.8039204604178055,
'www.osohq.com': 0.8039193847705445,
'marc.merlins.org': 0.8038831339196976,
'satyajit.ranjeev.in': 0.8038831339196976,
'kirindave.tumblr.com': 0.8038831339196976,
'transcend.io': 0.8038268126556799,
'pozorvlak.livejournal.com': 0.8038248807746377,
'elweb.co': 0.8038124805380112,
'www.arqbackup.com': 0.8037337901625023,
'www.turnkeylinux.org': 0.803725088601539,
'www.box.com': 0.8037180799167778,
'preciousplastic.com': 0.8036437246963563,
'www.stonetemple.com': 0.8036218680934559,
'oeis.org': 0.8036018584338912,
'www.bath.ac.uk': 0.8036016394623412,
'services.google.com': 0.803590777077633,
'theodi.org': 0.8035882804728938,
'saucelabs.com': 0.8035344497963749,
'greydanus.github.io': 0.8035295033581054,
'www.educative.io': 0.8035052925700027,
'librarianshipwreck.wordpress.com': 0.8034837043428301,
'sway.com': 0.803480126236027,
'datacolada.org': 0.803472812615791,
'convox.com': 0.8033973749313138,
'www.tokbox.com': 0.8033848854367407,
'review.firstround.com': 0.8033799488939632,
'www.netmagazine.com': 0.8033798192718756,
'www.stanfordalumni.org': 0.8033701218466093,
'www.fox5ny.com': 0.8033480309814771,
'phx.corporate-ir.net': 0.8033227542756777,
'christophjanz.blogspot.com': 0.8033072428565003,
'grugq.github.io': 0.8032867126150643,
'gist.githubusercontent.com': 0.8032867126150643,
'www.welivesecurity.com': 0.8032782303750765,
'www.linfo.org': 0.8032479251672556,
'www.vikingcodeschool.com': 0.8032454361054767,
'talsraviv.com': 0.8032454361054766,
'wyss.harvard.edu': 0.8032452848408873,
'brainspl.at': 0.8032452848408873,
'simblob.blogspot.com': 0.8032046395373721,
'sefsar.com': 0.8031660842268623,
'www.anmolsarma.in': 0.803136984148784,
'www.semiaccurate.com': 0.8031240460985013,
'www.mikekohn.net': 0.8031074147500573,
'm.wsj.com': 0.8030996898658181,
'doc-ok.org': 0.8030754588314308,
'duncandavidson.com': 0.8030754588314308,
'www.redbullstratos.com': 0.8030754588314308,
'lodev.org': 0.8030754588314308,
'www.sqreen.io': 0.8030754588314308,
'weburbanist.com': 0.8030719460684711,
'www.homelandsecuritynewswire.com': 0.8030383963381544,
'programming.oreilly.com': 0.8029986620998482,
'www.yuiblog.com': 0.8029870560713339,
'mgalligan.com': 0.802938284099674,
'startupsanonymous.com': 0.802896517417948,
'genomebiology.biomedcentral.com': 0.8028333073906616,
'www.commentary.org': 0.8028283532567575,
'paidcontent.org': 0.8027935252684901,
'www.entrepreneur.com': 0.8027573121471558,
'opensource.google.com': 0.8027457335537694,
'gus-massa.blogspot.com': 0.8027457335537693,
'attractivechaos.wordpress.com': 0.8027428934082615,
'timothyfitz.wordpress.com': 0.8027428934082615,
'news.voyage.auto': 0.8027425585809037,
'blog.safia.rocks': 0.802725147613407,
'pagely.com': 0.802725147613407,
'pedestrianobservations.wordpress.com': 0.8027251476134069,
'www.huckmagazine.com': 0.8027158674008132,
'amplitude.com': 0.8027111401704623,
'academicvc.com': 0.8026691214318213,
'zurb.com': 0.802669002882095,
'www.fogcreek.com': 0.8026427258093481,
'pubs.aeaweb.org': 0.8025918806108637,
'ochronus.online': 0.8025797396414792,
'mukulrathi.co.uk': 0.8025366605137177,
'www.tinfoilsecurity.com': 0.8025110355010865,
'blog.intercom.io': 0.802438081407932,
'www.counterpointresearch.com': 0.802349177742619,
'caines.ca': 0.8023472216796856,
'bitmessage.org': 0.8023472216796855,
'fare.livejournal.com': 0.802344964463485,
'sameroom.io': 0.8023231254672252,
'grapheneos.org': 0.8022611036781359,
'ryanhoover.me': 0.8022215636421977,
'ghost.org': 0.8021756677926977,
'ably.com': 0.8021756677926974,
'dgraph.io': 0.8021756677926974,
'www.pcg-random.org': 0.8021607965447277,
'www.electronicdesign.com': 0.802117143232502,
'blog.benjamin-encz.de': 0.8021070483963555,
'blog.apollographql.com': 0.8021070483963553,
'interaktiv.morgenpost.de': 0.8021068976672623,
'www.neogaf.com': 0.8020794141131092,
'jasongullickson.posterous.com': 0.802070055097611,
'www.lvguowei.me': 0.8020465972489538,
'theapeiron.co.uk': 0.8020408163265306,
'insideairbnb.com': 0.8020408163265306,
'www.business-standard.com': 0.8020316557325065,
'blog.chiariglione.org': 0.8020183701272139,
'news.byu.edu': 0.8019177911188085,
'ico.org.uk': 0.8019103627766435,
'www.asktog.com': 0.8018750459311643,
'www.thelightphone.com': 0.8018513951254583,
'weblog.muledesign.com': 0.8018281461815683,
'functional.christmas': 0.8017833933313524,
'dendory.net': 0.8017566800074669,
'css-irl.info': 0.8016782621735088,
'm.spiegel.de': 0.8016676297112596,
'blog.rubybestpractices.com': 0.8016652051765923,
'buzzsumo.com': 0.8016526731161184,
'edinburghhacklab.com': 0.8016359918200409,
'www.uc.edu': 0.8016359918200409,
'deprogrammaticaipsum.com': 0.8016284628618231,
'blog.teamtreehouse.com': 0.8016183213966954,
'www.poynter.org': 0.8015928533864275,
'blog.thirdyearmba.com': 0.8015659486074044,
'cmr.github.io': 0.8015566134648262,
'www.hackingwithswift.com': 0.8015552594719607,
'www.43folders.com': 0.801514257843383,
'www.whispersystems.org': 0.801509946546816,
'www.bravenewgeek.com': 0.8014600688018096,
'blog.siliconpublishing.com': 0.8014553154641547,
'psiloveyou.xyz': 0.8014114620819677,
'social.msdn.microsoft.com': 0.8014114620819677,
'thenewinquiry.com': 0.8014082764643022,
'www.grc.com': 0.8014045918631542,
'yle.fi': 0.8013958725561011,
'www.buildzoom.com': 0.8013406095826441,
'www.kmjn.org': 0.8013048425718535,
'www.miaminewtimes.com': 0.8012606895008024,
'pipedream.com': 0.8012606895008023,
'xoofx.com': 0.8012606895008023,
'9elements.com': 0.8012357442611142,
'www.safetyresearch.net': 0.8012295081967213,
'memcached.org': 0.8012295081967213,
'letterstoanewdeveloper.com': 0.801226878568775,
'viewfromthewing.boardingarea.com': 0.8012126203272316,
'nanxiao.me': 0.8011961282535711,
'ferrucc.io': 0.8011506248520655,
'blog.mrmeyer.com': 0.8011434789943767,
'www.dailycal.org': 0.8011077312639568,
'tldp.org': 0.8011077312639568,
'www.nbcphiladelphia.com': 0.8010985913944949,
'ww2.kqed.org': 0.8010518269035773,
'nitter.net': 0.8010512756640218,
'www.mathieupassenaud.fr': 0.8010059231728947,
'uploadvr.com': 0.8009858227111218,
'lincolnloop.com': 0.800978810280933,
'gaming.icrontic.com': 0.8009712545655284,
'www.techvibes.com': 0.8009550432546098,
'www.news.gatech.edu': 0.8009318982607027,
'leastauthority.com': 0.8008308575748613,
'find.xyz': 0.8008120556386431,
'rhodesmill.org': 0.8007877241325981,
'authzed.com': 0.8007485330706217,
'www.agweb.com': 0.8007485330706217,
'allegro.tech': 0.8007425067937589,
'nex-3.com': 0.8007357953179356,
'macaw.co': 0.8007357953179356,
'nic.ferrier.me.uk': 0.8007109685993726,
'www.planetscale.com': 0.8007054295824444,
'arshadchowdhury.com': 0.8006990158865002,
'www.adbusters.org': 0.8006901908979573,
'techno-anthropology.blogspot.com': 0.8006290310742641,
'noteflakes.com': 0.8006120189176902,
'blog.ellenchisa.com': 0.8006120189176901,
'ajaxian.com': 0.8005654734164377,
'getchip.com': 0.8004935213803819,
'www.dutchnews.nl': 0.8004918240352497,
'www.efavdb.com': 0.8004115226337449,
'mabez.dev': 0.8004115226337449,
'www.azul.com': 0.80040857295631,
'www.bonappetit.com': 0.8003933326639253,
'microconf.com': 0.8003448118609695,
'mxb.at': 0.8003448118609695,
'rancher.com': 0.8003102115898207,
'antoniocangiano.com': 0.8002939423538775,
'www.modernretail.co': 0.8002715044965122,
'gigamonkeys.wordpress.com': 0.8001954409713545,
'blog.veitheller.de': 0.8001954409713545,
'www.thehistoryblog.com': 0.8001885843714048,
'www.javacodegeeks.com': 0.80016657689339,
'blog.stevensanderson.com': 0.8001363248010883,
'neilmitchell.blogspot.com': 0.8001226738254366,
'www.foundingfuel.com': 0.800121664181143,
'simpleprogrammer.com': 0.8001130650615119,
'eclipse.org': 0.8000856208636091,
'www.acs.org': 0.8000105725594235,
'sigusr2.net': 0.8000105725594235}
def export_top_domains_to_json():
with open(TOP_DOMAINS_JSON_PATH, 'w') as output_file:
json.dump(DOMAINS, output_file, indent=2) | null |
179,745 | from mwmbl.tinysearchengine import TinyIndex, Document
def run():
urls = set()
with TinyIndex(Document, 'data/index.tinysearch') as index:
for i in range(index.num_pages):
print("Page", i)
page = index.get_page(i)
new_urls = {doc.url for doc in page}
urls |= new_urls
print("URLs", len(urls)) | null |
179,746 | import sqlite3
from mwmbl.indexer import URLS_PATH
from mwmbl.app import get_config_and_index
def create_database():
with sqlite3.connect(URLS_PATH) as connection:
connection.execute("""
CREATE TABLE urls (url TEXT PRIMARY KEY)
""")
def get_url_batches():
config, index = get_config_and_index()
for page_num in range(config.index_config.num_pages):
if page_num % 1000 == 0:
print("Processing page", page_num)
page = index.get_page(page_num)
if page is None:
continue
yield [url for title, url, extract in page]
def run():
create_database()
url_batches = get_url_batches()
with sqlite3.connect(URLS_PATH) as connection:
for url_batch in url_batches:
parameters = [(url,) for url in url_batch]
connection.executemany("""
INSERT OR IGNORE INTO urls VALUES (?)
""", parameters) | null |
179,747 | import os
from pathlib import Path
from random import Random
import numpy as np
from scipy.stats import sem
from mwmbl.tinysearchengine.indexer import TinyIndex, Document, _trim_items_to_page, astuple
from zstandard import ZstdCompressor
from mwmbl.utils import add_term_info
random = Random(1)
INDEX_PATH = Path(__file__).parent.parent / "devdata" / "index-v2.tinysearch"
def astuple(dc):
"""
Convert a type to a tuple - values at the end that are None can be truncated.
"""
value = tuple(dc.__dict__.values())
while value[-1] is None:
value = value[:-1]
return value
class Document:
title: str
url: str
extract: str
score: float
term: Optional[str] = None
state: Optional[int] = None
def __init__(self, title, url, extract, score, term=None, state=None):
# Sometimes the title or extract may be None, probably because of user generated content
# It's not allowed to be None though, or things will break
self.title = title if title is not None else ''
self.url = url
self.extract = extract if extract is not None else ''
self.score = score
self.term = term
self.state = state
def _trim_items_to_page(compressor: ZstdCompressor, page_size: int, items:list[T]):
# Find max number of items that fit on a page
return _binary_search_fitting_size(compressor, page_size, items, 0, len(items))
class TinyIndex(Generic[T]):
def __init__(self, item_factory: Callable[..., T], index_path, mode='r'):
if mode not in {'r', 'w'}:
raise ValueError(f"Mode should be one of 'r' or 'w', got {mode}")
with open(index_path, 'rb') as index_file:
metadata_page = index_file.read(METADATA_SIZE)
metadata_bytes = metadata_page.rstrip(b'\x00')
metadata = TinyIndexMetadata.from_bytes(metadata_bytes)
if metadata.item_factory != item_factory.__name__:
raise ValueError(f"Metadata item factory '{metadata.item_factory}' in the index "
f"does not match the passed item factory: '{item_factory.__name__}'")
self.item_factory = item_factory
self.index_path = index_path
self.mode = mode
self.num_pages = metadata.num_pages
self.page_size = metadata.page_size
logger.info(f"Loaded index with {self.num_pages} pages and {self.page_size} page size")
self.index_file = None
self.mmap = None
def __enter__(self):
self.index_file = open(self.index_path, 'r+b')
prot = PROT_READ if self.mode == 'r' else PROT_READ | PROT_WRITE
self.mmap = mmap(self.index_file.fileno(), 0, prot=prot)
return self
def __exit__(self, exc_type, exc_val, exc_tb):
self.mmap.close()
self.index_file.close()
def retrieve(self, key: str) -> List[T]:
index = self.get_key_page_index(key)
logger.debug(f"Retrieving index {index}")
page = self.get_page(index)
return [item for item in page if item.term is None or item.term == key]
def get_key_page_index(self, key) -> int:
key_hash = mmh3.hash(key, signed=False)
return key_hash % self.num_pages
def get_page(self, i) -> list[T]:
"""
Get the page at index i, decompress and deserialise it using JSON
"""
results = self._get_page_tuples(i)
return [self.item_factory(*item) for item in results]
def _get_page_tuples(self, i):
page_data = self.mmap[i * self.page_size + METADATA_SIZE:(i + 1) * self.page_size + METADATA_SIZE]
decompressor = ZstdDecompressor()
try:
decompressed_data = decompressor.decompress(page_data)
except ZstdError as e:
logger.exception(f"Error decompressing page {i}: {e}")
return []
return json.loads(decompressed_data.decode('utf8'))
def store_in_page(self, page_index: int, values: list[T]):
value_tuples = [astuple(value) for value in values]
self._write_page(value_tuples, page_index)
def _write_page(self, data, i: int):
"""
Serialise the data using JSON, compress it and store it at index i.
If the data is too big, it will store the first items in the list and discard the rest.
"""
if self.mode != 'w':
raise UnsupportedOperation("The file is open in read mode, you cannot write")
page_data = _get_page_data(self.page_size, data)
logger.debug(f"Got page data of length {len(page_data)}")
self.mmap[i * self.page_size + METADATA_SIZE:(i+1) * self.page_size + METADATA_SIZE] = page_data
def create(item_factory: Callable[..., T], index_path: str, num_pages: int, page_size: int):
if os.path.isfile(index_path):
raise FileExistsError(f"Index file '{index_path}' already exists")
metadata = TinyIndexMetadata(VERSION, page_size, num_pages, item_factory.__name__)
metadata_bytes = metadata.to_bytes()
metadata_padded = _pad_to_page_size(metadata_bytes, METADATA_SIZE)
page_bytes = _get_page_data(page_size, [])
with open(index_path, 'wb') as index_file:
index_file.write(metadata_padded)
for i in range(num_pages):
index_file.write(page_bytes)
return TinyIndex(item_factory, index_path=index_path)
def add_term_info(document: Document, index: TinyIndex, page_index: int):
tokenized = tokenize_document(document.url, document.title, document.extract, document.score)
for token in tokenized.tokens:
token_page_index = index.get_key_page_index(token)
if token_page_index == page_index:
return Document(document.title, document.url, document.extract, document.score, token)
raise ValueError("Could not find token in page index")
def run():
compressor = ZstdCompressor()
with TinyIndex(Document, INDEX_PATH) as index:
# Get some random integers between 0 and index.num_pages:
pages = random.sample(range(index.num_pages), 10000)
old_sizes = []
new_sizes = []
for i in pages:
page = index.get_page(i)
term_documents = []
for document in page:
term_document = add_term_info(document, index, i)
term_documents.append(term_document)
value_tuples = [astuple(value) for value in term_documents]
num_fitting, compressed = _trim_items_to_page(compressor, index.page_size, value_tuples)
new_sizes.append(num_fitting)
old_sizes.append(len(page))
print("Old sizes mean", np.mean(old_sizes), sem(old_sizes))
print("New sizes mean", np.mean(new_sizes), sem(new_sizes)) | null |
179,748 | import requests
from mwmbl.crawler import Batch, Item, ItemContent
URL = 'http://localhost:5000/crawler/batches/'
def run():
batch = Batch(user_id='test_user_id111111111111111111111111', items=[Item(
url='https://www.theguardian.com/stage/2007/nov/18/theatre',
content=ItemContent(
title='A nation in search of the new black | Theatre | The Guardian',
extract="Topic-stuffed and talk-filled, Kwame Kwei-Armah's new play proves that issue-driven drama is (despite reports of its death) still being written and staged…",
links=[]),
timestamp=123456,
status=200,
)])
result = requests.post(URL, data=batch.json())
print("Result", result.content) | null |
179,749 | import glob
import gzip
import json
import requests
from mwmbl.indexer import CRAWL_GLOB
API_ENDPOINT = "http://95.216.215.29/batches/historical"
def total_num_batches():
return len(glob.glob(CRAWL_GLOB))
def get_batches():
for path in sorted(glob.glob(CRAWL_GLOB)):
hashed_batch = json.load(gzip.open(path))
yield hashed_batch
def convert_item(item):
return {
'url': item['url'],
'status': 200,
'timestamp': item['timestamp'],
'content': {
'title': item['title'],
'extract': item['extract'],
'links': item['links'],
}
}
def run():
total_batches = total_num_batches()
batches = get_batches()
for i, hashed_batch in enumerate(batches):
new_batch = {
'user_id_hash': hashed_batch['user_id_hash'],
'timestamp': hashed_batch['timestamp'],
'items': [convert_item(item) for item in hashed_batch['items']]
}
response = requests.post(API_ENDPOINT, json=new_batch)
print(f"Response {i} of {total_batches}", response) | null |
179,750 | from setuptools import find_packages, setup
from typing import List
def parse_requirements(file_name: str) -> List[str]:
with open(file_name) as f:
return [
require.strip() for require in f
if require.strip() and not require.startswith('#')
] | null |
179,751 | from setuptools import find_packages, setup
from typing import List
def readme():
with open('README.md', encoding='utf-8') as f:
content = f.read()
return content | null |
179,752 | from setuptools import find_packages, setup
from typing import List
version_file = 'modelscope_agent/version.py'
def get_version():
with open(version_file, 'r', encoding='utf-8') as f:
exec(compile(f.read(), version_file, 'exec'))
return locals()['__version__'] | null |
179,753 | from __future__ import annotations
import os
import sys
from functools import partial
import gradio as gr
from dotenv import load_dotenv
from gradio_chatbot import ChatBot
from modelscope_agent.agent import AgentExecutor
from modelscope_agent.llm import LLMFactory
from predict import stream_predict, upload_image
from modelscope.utils.config import Config
with gr.Blocks(css=MAIN_CSS_CODE, theme=gr.themes.Soft()) as demo:
upload_image_url = gr.State('')
# agent 对象
tool_cfg_file = os.getenv('TOOL_CONFIG_FILE')
model_cfg_file = os.getenv('MODEL_CONFIG_FILE')
model_cfg = Config.from_file(model_cfg_file)
tool_cfg = Config.from_file(tool_cfg_file)
model_name = 'modelscope-agent-qwen-7b'
llm = LLMFactory.build_llm(model_name, model_cfg)
agent = AgentExecutor(llm, tool_cfg)
stream_predict_p = partial(stream_predict, agent=agent)
with gr.Row():
gr.HTML(
"""<h1 align="left" style="min-width:200px; margin-top:0;">ModelScopeGPT</h1>"""
)
status_display = gr.HTML(
'', elem_id='status_display', visible=False, show_label=False)
with gr.Row(elem_id='container_row').style(equal_height=True):
with gr.Column(
scale=8,
elem_classes=['chatInterface', 'chatDialog', 'chatContent']):
with gr.Row(elem_id='chat-container'):
chatbot = ChatBot(
elem_id='chatbot',
elem_classes=['markdown-body'],
show_label=False)
chatbot_classic = gr.Textbox(
lines=20,
visible=False,
interactive=False,
label='classic_chatbot',
elem_id='chatbot_classic')
with gr.Row(elem_id='chat-bottom-container'):
with gr.Column(min_width=70, scale=1):
clear_session_button = gr.Button(
'清除', elem_id='clear_session_button')
with gr.Column(min_width=100, scale=1):
upload_button = gr.UploadButton(
'上传图片', file_types=['image'])
with gr.Column(scale=12):
user_input = gr.Textbox(
show_label=False,
placeholder='和我聊聊吧～',
elem_id='chat-input').style(container=False)
uploaded_image_box = gr.HTML(
'', visible=False, show_label=False)
with gr.Column(min_width=70, scale=1):
submitBtn = gr.Button('发送', variant='primary')
with gr.Column(min_width=110, scale=1):
regenerate_button = gr.Button(
'重新生成', elem_id='regenerate_button')
with gr.Column(min_width=470, scale=4, elem_id='settings'):
icon_path = 'https://img.alicdn.com/imgextra/i4/O1CN01kpkVcX1wSCO362MH4_!!6000000006306-1-tps-805-805.gif'
info_text = '我是ModelScopeGPT（魔搭GPT）， 是一个大小模型协同的agent系统。\
我具备多种能力，可以通过大模型做中枢（controller），来控制魔搭社区的各种多模态模型api回复用户的问题。\
除此之外，我还集成了知识库检索引擎，可以解答用户在魔搭社区使用模型遇到的问题以及模型知识相关问答。'
gr.HTML(f"""
<div class="robot-info">
<img src="{icon_path}"</img>
<div class="robot-info-text">
"{info_text}"
</div>
</div>
""")
gr.Examples(
examples=[
'写一首简短的夏天落日的诗',
'讲一个小男孩的故事，20字左右',
'用男声读出来',
'生成个图片看看',
'从下面的地址，找到省市区等元素，地址：浙江杭州市江干区九堡镇三村村一区',
],
inputs=[user_input],
examples_per_page=20,
label='示例',
elem_id='chat-examples')
stream_predict_input = [chatbot, user_input, upload_image_url]
stream_predict_output = [chatbot, status_display]
clean_outputs = [gr.update(value=''), '']
clean_outputs_target = [user_input, uploaded_image_box]
user_input.submit(
stream_predict_p,
stream_predict_input,
stream_predict_output,
show_progress=True)
user_input.submit(
fn=lambda: clean_outputs, inputs=[], outputs=clean_outputs_target)
submitBtn.click(
stream_predict_p,
stream_predict_input,
stream_predict_output,
show_progress=True)
submitBtn.click(
fn=lambda: clean_outputs, inputs=[], outputs=clean_outputs_target)
regenerate_button.click(
fn=lambda: clean_outputs, inputs=[], outputs=clean_outputs_target)
regenerate_button.click(
stream_predict_p,
stream_predict_input,
stream_predict_output,
show_progress=True)
upload_button.upload(upload_image, upload_button,
[uploaded_image_box, upload_image_url])
clear_session_button.click(
fn=clear_session,
inputs=[],
outputs=[chatbot, uploaded_image_box, upload_image_url])
demo.title = 'ModelScopeGPT 🎁'
demo.queue(concurrency_count=10, status_update_rate='auto', api_open=False)
demo.launch(show_api=False, share=False)
def clear_session():
agent.reset()
return {
chatbot: gr.update(value=[]),
uploaded_image_box: '',
upload_image_url: '',
} | null |
179,754 | from __future__ import annotations
import os
import re
import traceback
import uuid
from copy import deepcopy
import gradio as gr
import json
def stream_predict(
chatbot, # ChatBot
user_input, # Textbox
upload_image_url, # imageUrl
agent # agent
):
print(f'upload_image_url: {upload_image_url}')
if not user_input:
if len(history) == 0:
yield chatbot, '请输入问题……'
return
else:
user_input = chatbot[-1][0]
chatbot = chatbot[:-1]
yield chatbot, '重新生成回答……'
else:
print(upload_image_url)
if upload_image_url:
user_input = f'\n' + user_input
print('user_input:', user_input)
yield chatbot, '开始生成回答……'
chatbot.append((user_input, '处理中...'))
yield chatbot, '开始实时传输回答……'
response = ''
try:
for frame in agent.stream_run(user_input, remote=False):
is_final = frame.get('is_final')
llm_result = frame.get('llm_text', '')
exec_result = frame.get('exec_result', '')
error_message = frame.get('error')
if is_final:
chatbot[-1] = (chatbot[-1][0], response)
yield chatbot, '完成回答'
break
elif error_message:
chatbot[-1] = (chatbot[-1][0], error_message)
yield chatbot, ''
else:
if len(exec_result) != 0:
# llm_result
frame_text = f'\n\n<|startofexec|>{exec_result}<|endofexec|>\n'
else:
# action_exec_result
frame_text = llm_result
response = f'{response}{frame_text}'
chatbot[-1] = (chatbot[-1][0], response)
yield chatbot, ''
except Exception:
traceback.print_exc()
chatbot[-1] = (chatbot[-1][0], 'chat_async error.')
yield chatbot, '' | null |
179,755 | from __future__ import annotations
import os
import re
import traceback
import uuid
from copy import deepcopy
import gradio as gr
import json
def upload_image(file):
gr_file_path = f'./file={file.name}'
return [
gr.HTML.update(
f"<div class=\"uploaded-image-box\"><img src=\"{gr_file_path}\"></img><div>",
visible=True), gr_file_path
] | null |
179,756 | import os
import random
import shutil
import traceback
import gradio as gr
import modelscope_gradio_components as mgr
from config_utils import get_avatar_image, get_ci_dir, parse_configuration
from gradio_utils import format_cover_html
from modelscope_agent.schemas import Message
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_gradio_components.components.Chatbot.llm_thinking_presets import \
qwen
from user_core import init_user_chatbot_agent
uuid_str = 'local_user'
def check_uuid(uuid_str):
if not uuid_str or uuid_str == '':
if os.getenv('MODELSCOPE_ENVIRONMENT') == 'studio':
raise gr.Error('请登陆后使用! (Please login first)')
else:
uuid_str = 'local_user'
return uuid_str
def init_user(state):
try:
seed = state.get('session_seed', random.randint(0, 1000000000))
user_agent, user_memory = init_user_chatbot_agent(uuid_str)
user_agent.seed = seed
state['user_agent'] = user_agent
state['user_memory'] = user_memory
except Exception as e:
logger.query_error(
uuid=uuid_str,
error=str(e),
details={'error_traceback': traceback.format_exc()})
raise Exception(e)
return state
def get_ci_dir():
return DEFAULT_CODE_INTERPRETER_DIR
class Message(BaseModel):
"""
Message: message information
"""
role: str = ''
content: str = ''
def send_message(chatbot, input, _state):
# 将发送的消息添加到聊天历史
if 'user_agent' not in _state:
init_user(_state)
# 将发送的消息添加到聊天历史
_uuid_str = check_uuid(uuid_str)
user_agent = _state['user_agent']
user_memory = _state['user_memory']
append_files = []
for file in input.files:
file_name = os.path.basename(file.path)
# covert xxx.json to xxx_uuid_str.json
file_name = file_name.replace('.', f'_{_uuid_str}.')
file_path = os.path.join(get_ci_dir(), file_name)
if not os.path.exists(file_path):
# make sure file path's directory exists
os.makedirs(os.path.dirname(file_path), exist_ok=True)
shutil.copy(file.path, file_path)
append_files.append(file_path)
chatbot.append([{'text': input.text, 'files': input.files}, None])
yield {
user_chatbot: chatbot,
user_chatbot_input: None,
}
# get short term memory history
history = user_memory.get_history()
# get long term memory knowledge, currently get one file
uploaded_file = None
if len(append_files) > 0:
uploaded_file = append_files[0]
ref_doc = user_memory.run(
query=input.text, url=uploaded_file, checked=True)
response = ''
try:
for frame in user_agent.run(
input.text,
history=history,
ref_doc=ref_doc,
append_files=append_files):
# important! do not change this
response += frame
chatbot[-1][1] = response
yield {
user_chatbot: chatbot,
}
if len(history) == 0:
user_memory.update_history(
Message(role='system', content=user_agent.system_prompt))
user_memory.update_history([
Message(role='user', content=input.text),
Message(role='assistant', content=response),
])
except Exception as e:
if 'dashscope.common.error.AuthenticationError' in str(e):
msg = 'DASHSCOPE_API_KEY should be set via environment variable. You can acquire this in ' \
'https://help.aliyun.com/zh/dashscope/developer-reference/activate-dashscope-and-create-an-api-key'
elif 'rate limit' in str(e):
msg = 'Too many people are calling, please try again later.'
else:
msg = str(e)
chatbot[-1][1] = msg
yield {user_chatbot: chatbot} | null |
179,757 | import os
import random
import shutil
import traceback
import gradio as gr
import json
import modelscope_gradio_components as mgr
import yaml
from builder_core import (beauty_output, gen_response_and_process,
init_builder_chatbot_agent)
from config_utils import (DEFAULT_AGENT_DIR, Config, get_avatar_image,
get_ci_dir, get_user_dir,
is_valid_plugin_configuration, parse_configuration,
save_avatar_image, save_builder_configuration,
save_plugin_configuration)
from gradio_utils import format_cover_html, format_goto_publish_html
from i18n import I18n
from modelscope_agent.schemas import Message
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_gradio_components.components.Chatbot.llm_thinking_presets import \
qwen
from publish_util import (pop_user_info_from_config, prepare_agent_zip,
reload_agent_zip)
from user_core import init_user_chatbot_agent
def check_uuid(uuid_str):
if not uuid_str or uuid_str == '':
if os.getenv('MODELSCOPE_ENVIRONMENT') == 'studio':
raise gr.Error('请登陆后使用! (Please login first)')
else:
uuid_str = 'local_user'
return uuid_str
with demo:
uuid_str = gr.Textbox(label='modelscope_uuid', visible=False)
draw_seed = random.randint(0, 1000000000)
state = gr.State({'session_seed': draw_seed})
i18n = I18n('zh-cn')
with gr.Row():
with gr.Column(scale=5):
header = gr.Markdown(i18n.get('header'))
with gr.Column(scale=1):
language = gr.Dropdown(
choices=[('中文', 'zh-cn'), ('English', 'en')],
show_label=False,
container=False,
value='zh-cn',
interactive=True)
with gr.Row():
with gr.Column():
with gr.Tabs() as tabs:
with gr.Tab(i18n.get_whole('create'), id=0) as create_tab:
with gr.Column():
# "Create" 标签页的 Chatbot 组件
start_text = '欢迎使用agent创建助手。我可以帮助您创建一个定制agent。'\
'您希望您的agent主要用于什么领域或任务？比如，您可以说，我想做一个RPG游戏agent'
create_chatbot = mgr.Chatbot(
show_label=False,
value=[[None, start_text]],
flushing=False,
show_copy_button=True,
llm_thinking_presets=[
qwen(
action_input_title='调用 <Action>',
action_output_title='完成调用')
])
create_chat_input = mgr.MultimodalInput(
label=i18n.get('message'),
placeholder=i18n.get('message_placeholder'),
interactive=False,
upload_button_props=dict(visible=False),
submit_button_props=dict(
label=i18n.get('sendOnLoading')))
configure_tab = gr.Tab(i18n.get_whole('configure'), id=1)
with configure_tab:
with gr.Column():
# "Configure" 标签页的配置输入字段
with gr.Row():
bot_avatar_comp = gr.Image(
label=i18n.get('form_avatar'),
sources=['upload'],
interactive=True,
type='filepath',
scale=1,
width=182,
height=182,
)
with gr.Column(scale=4):
name_input = gr.Textbox(
label=i18n.get('form_name'),
placeholder=i18n.get(
'form_name_placeholder'))
description_input = gr.Textbox(
label=i18n.get('form_description'),
placeholder=i18n.get(
'form_description_placeholder'))
instructions_input = gr.Textbox(
label=i18n.get('form_instructions'),
placeholder=i18n.get(
'form_instructions_placeholder'),
lines=3)
model_selector = gr.Dropdown(
label=i18n.get('form_model'))
agent_language_selector = gr.Dropdown(
label=i18n.get('form_agent_language'),
choices=['zh', 'en'],
value='zh')
suggestion_input = gr.Dataframe(
show_label=False,
value=[['']],
datatype=['str'],
headers=[i18n.get_whole('form_prompt_suggestion')],
type='array',
col_count=(1, 'fixed'),
interactive=True)
gr.Markdown(
'*注意：知识库上传的文本文档默认按照\\n\\n切分，pdf默认按照页切分。如果片段'
'对应的字符大于[配置文件](https://github.com/modelscope/modelscope-agent/'
'blob/master/apps/agentfabric/config/model_config.json)中指定模型的'
'knowledge限制，则在被召回时有可能会被截断。*')
knowledge_input = gr.File(
label=i18n.get('form_knowledge'),
file_count='multiple',
file_types=[
'text', '.json', '.csv', '.pdf', '.md'
])
capabilities_checkboxes = gr.CheckboxGroup(
label=i18n.get('form_capabilities'))
with gr.Accordion(
i18n.get('open_api_accordion'),
open=False) as open_api_accordion:
openapi_schema = gr.Textbox(
label='Schema',
placeholder=
'Enter your OpenAPI schema here, JSON or YAML format only'
)
with gr.Group():
openapi_auth_type = gr.Radio(
label='Authentication Type',
choices=['None', 'API Key'],
value='None')
openapi_auth_apikey = gr.Textbox(
label='API Key',
placeholder='Enter your API Key here')
openapi_auth_apikey_type = gr.Radio(
label='API Key type', choices=['Bearer'])
openapi_privacy_policy = gr.Textbox(
label='Privacy Policy',
placeholder='Enter privacy policy URL')
configure_button = gr.Button(
i18n.get('form_update_button'))
with gr.Accordion(
label=i18n.get('import_config'),
open=False) as update_accordion:
with gr.Column():
update_space = gr.Textbox(
label=i18n.get('space_addr'),
placeholder=i18n.get('input_space_addr'))
import_button = gr.Button(
i18n.get_whole('import_space'))
gr.Markdown(
f'#### {i18n.get_whole("import_hint")}')
with gr.Column():
# Preview
preview_header = gr.HTML(
f"""<div class="preview_header">{i18n.get('preview')}<div>""")
user_chat_bot_cover = gr.HTML(format_cover_html({}, None))
user_chatbot = mgr.Chatbot(
value=[[None, None]],
elem_id='user_chatbot',
elem_classes=['markdown-body'],
avatar_images=get_avatar_image('', uuid_str),
height=650,
latex_delimiters=[],
show_label=False,
visible=False,
show_copy_button=True,
llm_thinking_presets=[
qwen(
action_input_title='调用 <Action>',
action_output_title='完成调用')
])
preview_chat_input = mgr.MultimodalInput(
interactive=False,
label=i18n.get('message'),
placeholder=i18n.get('message_placeholder'),
upload_button_props=dict(
label=i18n.get('upload_btn'),
file_types=['file', 'image', 'audio', 'video', 'text'],
file_count='multiple'),
submit_button_props=dict(label=i18n.get('sendOnLoading')))
user_chat_bot_suggest = gr.Dataset(
label=i18n.get('prompt_suggestion'),
components=[preview_chat_input],
samples=[])
user_chat_bot_suggest.select(
lambda evt: evt[0],
inputs=[user_chat_bot_suggest],
outputs=[preview_chat_input])
with gr.Accordion(
label=i18n.get('publish'),
open=False) as publish_accordion:
publish_alert_md = gr.Markdown(f'{i18n.get("publish_alert")}')
with gr.Row():
with gr.Column():
publish_button = gr.Button(i18n.get_whole('build'))
build_hint_md = gr.Markdown(
f'#### 1.{i18n.get("build_hint")}')
with gr.Column():
publish_link = gr.HTML(
value=format_goto_publish_html(
i18n.get_whole('publish'), '', {}, True))
publish_hint_md = gr.Markdown(
f'#### 2.{i18n.get("publish_hint")}')
configure_updated_outputs = [
state,
# config form
bot_avatar_comp,
name_input,
description_input,
instructions_input,
model_selector,
agent_language_selector,
suggestion_input,
knowledge_input,
capabilities_checkboxes,
# bot
user_chat_bot_cover,
user_chat_bot_suggest,
preview_chat_input,
create_chat_input,
]
# 初始化表单
def init_ui_config(uuid_str, _state, builder_cfg, model_cfg, tool_cfg):
logger.query_info(
uuid=uuid_str,
message='builder_cfg',
details={'builder_cfg': str(builder_cfg)})
# available models
models = list(model_cfg.keys())
capabilities = [(tool_cfg[tool_key]['name'], tool_key)
for tool_key in tool_cfg.keys()
if tool_cfg[tool_key].get('is_active', False)]
_state['model_cfg'] = model_cfg
_state['tool_cfg'] = tool_cfg
_state['capabilities'] = capabilities
bot_avatar = get_avatar_image(builder_cfg.get('avatar', ''),
uuid_str)[1]
suggests = builder_cfg.get('prompt_recommend', [''])
return {
state:
_state,
bot_avatar_comp:
gr.Image(value=bot_avatar),
name_input:
builder_cfg.get('name', ''),
description_input:
builder_cfg.get('description'),
instructions_input:
builder_cfg.get('instruction'),
model_selector:
gr.Dropdown(
value=builder_cfg.get('model', models[0]), choices=models),
agent_language_selector:
builder_cfg.get('language') or 'zh',
suggestion_input:
[[str] for str in suggests] if len(suggests) > 0 else [['']],
knowledge_input:
builder_cfg.get('knowledge', [])
if len(builder_cfg['knowledge']) > 0 else None,
capabilities_checkboxes:
gr.CheckboxGroup(
value=[
tool for tool in builder_cfg.get('tools', {}).keys()
if builder_cfg.get('tools').get(tool).get('use', False)
],
choices=capabilities),
# bot
user_chat_bot_cover:
format_cover_html(builder_cfg, bot_avatar),
user_chat_bot_suggest:
gr.Dataset(
components=[preview_chat_input],
samples=[[item] for item in suggests]),
}
# tab 切换的事件处理
configure_tab.select(
on_congifure_tab_select,
inputs=[state, uuid_str],
outputs=configure_updated_outputs)
# 配置 "Create" 标签页的消息发送功能
create_chat_input.submit(
create_send_message,
inputs=[create_chatbot, create_chat_input, state, uuid_str],
outputs=[
create_chatbot, user_chat_bot_cover, user_chatbot,
user_chat_bot_suggest, create_chat_input
])
# 配置 "Configure" 标签页的提交按钮功能
configure_button.click(
process_configuration,
inputs=[
uuid_str, bot_avatar_comp, name_input, description_input,
instructions_input, model_selector, agent_language_selector,
suggestion_input, knowledge_input, capabilities_checkboxes,
openapi_schema, openapi_auth_type, openapi_auth_apikey,
openapi_auth_apikey_type, openapi_privacy_policy, state
],
outputs=[
user_chat_bot_cover, user_chatbot, user_chat_bot_suggest,
suggestion_input
])
# 配置 "Preview" 的消息发送功能
preview_chat_input.submit(
preview_send_message,
inputs=[user_chatbot, preview_chat_input, state, uuid_str],
outputs=[user_chatbot, user_chat_bot_cover, preview_chat_input])
# configuration for publish
publish_button.click(
publish_agent,
inputs=[name_input, uuid_str, state],
outputs=[publish_link],
)
import_button.click(
import_space,
inputs=[update_space, uuid_str, state],
outputs=configure_updated_outputs,
)
language.select(
change_lang,
inputs=[language],
outputs=configure_updated_outputs + [
configure_button, create_chat_input, open_api_accordion,
preview_header, preview_chat_input, publish_accordion, header,
publish_alert_md, build_hint_md, publish_hint_md
])
demo.load(
init_all, inputs=[uuid_str, state], outputs=configure_updated_outputs)
def parse_configuration(uuid_str=''):
"""parse configuration
Args:
Returns:
dict: parsed configuration
"""
model_cfg_file = os.getenv('MODEL_CONFIG_FILE', DEFAULT_MODEL_CONFIG_FILE)
builder_cfg_file = get_user_cfg_file(uuid_str)
# use default if not exists
if not os.path.exists(builder_cfg_file):
# create parents directory
os.makedirs(os.path.dirname(builder_cfg_file), exist_ok=True)
# copy the template to the address
builder_cfg_file_temp = './config/builder_config.json'
if builder_cfg_file_temp != builder_cfg_file:
shutil.copy(builder_cfg_file_temp, builder_cfg_file)
tool_cfg_file = os.getenv('TOOL_CONFIG_FILE', DEFAULT_TOOL_CONFIG_FILE)
builder_cfg = Config.from_file(builder_cfg_file)
model_cfg = Config.from_file(model_cfg_file)
tool_cfg = Config.from_file(tool_cfg_file)
tools_info = builder_cfg.tools
available_tool_list = []
for key, value in tools_info.items():
if value['use']:
available_tool_list.append(key)
tool_cfg[key]['use'] = value['use']
openapi_plugin_file = get_user_openapi_plugin_cfg_file(uuid_str)
plugin_cfg = {}
available_plugin_list = []
if os.path.exists(openapi_plugin_file):
openapi_plugin_cfg = Config.from_file(openapi_plugin_file)
try:
config_dict = openapi_schema_convert(
schema=openapi_plugin_cfg.schema,
auth=openapi_plugin_cfg.auth.to_dict())
plugin_cfg = Config(config_dict)
for name, config in config_dict.items():
available_plugin_list.append(name)
except Exception as e:
logger.query_error(
uuid=uuid_str,
error=str(e),
content={
'error_traceback':
traceback.format_exc(),
'error_details':
'The format of the plugin config file is incorrect.'
})
return builder_cfg, model_cfg, tool_cfg, available_tool_list, plugin_cfg, available_plugin_list
def on_congifure_tab_select(_state, uuid_str):
uuid_str = check_uuid(uuid_str)
configure_updated = _state.get('configure_updated', False)
if configure_updated:
builder_cfg, model_cfg, tool_cfg, available_tool_list, _, _ = parse_configuration(
uuid_str)
_state['configure_updated'] = False
return init_ui_config(uuid_str, _state, builder_cfg, model_cfg,
tool_cfg)
else:
return {state: _state} | null |
179,758 | import os
import random
import shutil
import traceback
import gradio as gr
import json
import modelscope_gradio_components as mgr
import yaml
from builder_core import (beauty_output, gen_response_and_process,
init_builder_chatbot_agent)
from config_utils import (DEFAULT_AGENT_DIR, Config, get_avatar_image,
get_ci_dir, get_user_dir,
is_valid_plugin_configuration, parse_configuration,
save_avatar_image, save_builder_configuration,
save_plugin_configuration)
from gradio_utils import format_cover_html, format_goto_publish_html
from i18n import I18n
from modelscope_agent.schemas import Message
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_gradio_components.components.Chatbot.llm_thinking_presets import \
qwen
from publish_util import (pop_user_info_from_config, prepare_agent_zip,
reload_agent_zip)
from user_core import init_user_chatbot_agent
def check_uuid(uuid_str):
with demo:
uuid_str = gr.Textbox(label='modelscope_uuid', visible=False)
draw_seed = random.randint(0, 1000000000)
state = gr.State({'session_seed': draw_seed})
i18n = I18n('zh-cn')
with gr.Row():
with gr.Column(scale=5):
header = gr.Markdown(i18n.get('header'))
with gr.Column(scale=1):
language = gr.Dropdown(
choices=[('中文', 'zh-cn'), ('English', 'en')],
show_label=False,
container=False,
value='zh-cn',
interactive=True)
with gr.Row():
with gr.Column():
with gr.Tabs() as tabs:
with gr.Tab(i18n.get_whole('create'), id=0) as create_tab:
with gr.Column():
# "Create" 标签页的 Chatbot 组件
start_text = '欢迎使用agent创建助手。我可以帮助您创建一个定制agent。'\
'您希望您的agent主要用于什么领域或任务？比如，您可以说，我想做一个RPG游戏agent'
create_chatbot = mgr.Chatbot(
show_label=False,
value=[[None, start_text]],
flushing=False,
show_copy_button=True,
llm_thinking_presets=[
qwen(
action_input_title='调用 <Action>',
action_output_title='完成调用')
])
create_chat_input = mgr.MultimodalInput(
label=i18n.get('message'),
placeholder=i18n.get('message_placeholder'),
interactive=False,
upload_button_props=dict(visible=False),
submit_button_props=dict(
label=i18n.get('sendOnLoading')))
configure_tab = gr.Tab(i18n.get_whole('configure'), id=1)
with configure_tab:
with gr.Column():
# "Configure" 标签页的配置输入字段
with gr.Row():
bot_avatar_comp = gr.Image(
label=i18n.get('form_avatar'),
sources=['upload'],
interactive=True,
type='filepath',
scale=1,
width=182,
height=182,
)
with gr.Column(scale=4):
name_input = gr.Textbox(
label=i18n.get('form_name'),
placeholder=i18n.get(
'form_name_placeholder'))
description_input = gr.Textbox(
label=i18n.get('form_description'),
placeholder=i18n.get(
'form_description_placeholder'))
instructions_input = gr.Textbox(
label=i18n.get('form_instructions'),
placeholder=i18n.get(
'form_instructions_placeholder'),
lines=3)
model_selector = gr.Dropdown(
label=i18n.get('form_model'))
agent_language_selector = gr.Dropdown(
label=i18n.get('form_agent_language'),
choices=['zh', 'en'],
value='zh')
suggestion_input = gr.Dataframe(
show_label=False,
value=[['']],
datatype=['str'],
headers=[i18n.get_whole('form_prompt_suggestion')],
type='array',
col_count=(1, 'fixed'),
interactive=True)
gr.Markdown(
'*注意：知识库上传的文本文档默认按照\\n\\n切分，pdf默认按照页切分。如果片段'
'对应的字符大于[配置文件](https://github.com/modelscope/modelscope-agent/'
'blob/master/apps/agentfabric/config/model_config.json)中指定模型的'
'knowledge限制，则在被召回时有可能会被截断。*')
knowledge_input = gr.File(
label=i18n.get('form_knowledge'),
file_count='multiple',
file_types=[
'text', '.json', '.csv', '.pdf', '.md'
])
capabilities_checkboxes = gr.CheckboxGroup(
label=i18n.get('form_capabilities'))
with gr.Accordion(
i18n.get('open_api_accordion'),
open=False) as open_api_accordion:
openapi_schema = gr.Textbox(
label='Schema',
placeholder=
'Enter your OpenAPI schema here, JSON or YAML format only'
)
with gr.Group():
openapi_auth_type = gr.Radio(
label='Authentication Type',
choices=['None', 'API Key'],
value='None')
openapi_auth_apikey = gr.Textbox(
label='API Key',
placeholder='Enter your API Key here')
openapi_auth_apikey_type = gr.Radio(
label='API Key type', choices=['Bearer'])
openapi_privacy_policy = gr.Textbox(
label='Privacy Policy',
placeholder='Enter privacy policy URL')
configure_button = gr.Button(
i18n.get('form_update_button'))
with gr.Accordion(
label=i18n.get('import_config'),
open=False) as update_accordion:
with gr.Column():
update_space = gr.Textbox(
label=i18n.get('space_addr'),
placeholder=i18n.get('input_space_addr'))
import_button = gr.Button(
i18n.get_whole('import_space'))
gr.Markdown(
f'#### {i18n.get_whole("import_hint")}')
with gr.Column():
# Preview
preview_header = gr.HTML(
f"""<div class="preview_header">{i18n.get('preview')}<div>""")
user_chat_bot_cover = gr.HTML(format_cover_html({}, None))
user_chatbot = mgr.Chatbot(
value=[[None, None]],
elem_id='user_chatbot',
elem_classes=['markdown-body'],
avatar_images=get_avatar_image('', uuid_str),
height=650,
latex_delimiters=[],
show_label=False,
visible=False,
show_copy_button=True,
llm_thinking_presets=[
qwen(
action_input_title='调用 <Action>',
action_output_title='完成调用')
])
preview_chat_input = mgr.MultimodalInput(
interactive=False,
label=i18n.get('message'),
placeholder=i18n.get('message_placeholder'),
upload_button_props=dict(
label=i18n.get('upload_btn'),
file_types=['file', 'image', 'audio', 'video', 'text'],
file_count='multiple'),
submit_button_props=dict(label=i18n.get('sendOnLoading')))
user_chat_bot_suggest = gr.Dataset(
label=i18n.get('prompt_suggestion'),
components=[preview_chat_input],
samples=[])
user_chat_bot_suggest.select(
lambda evt: evt[0],
inputs=[user_chat_bot_suggest],
outputs=[preview_chat_input])
with gr.Accordion(
label=i18n.get('publish'),
open=False) as publish_accordion:
publish_alert_md = gr.Markdown(f'{i18n.get("publish_alert")}')
with gr.Row():
with gr.Column():
publish_button = gr.Button(i18n.get_whole('build'))
build_hint_md = gr.Markdown(
f'#### 1.{i18n.get("build_hint")}')
with gr.Column():
publish_link = gr.HTML(
value=format_goto_publish_html(
i18n.get_whole('publish'), '', {}, True))
publish_hint_md = gr.Markdown(
f'#### 2.{i18n.get("publish_hint")}')
configure_updated_outputs = [
state,
# config form
bot_avatar_comp,
name_input,
description_input,
instructions_input,
model_selector,
agent_language_selector,
suggestion_input,
knowledge_input,
capabilities_checkboxes,
# bot
user_chat_bot_cover,
user_chat_bot_suggest,
preview_chat_input,
create_chat_input,
]
# 初始化表单
# tab 切换的事件处理
configure_tab.select(
on_congifure_tab_select,
inputs=[state, uuid_str],
outputs=configure_updated_outputs)
# 配置 "Create" 标签页的消息发送功能
def format_message_with_builder_cfg(_state, chatbot, builder_cfg,
uuid_str):
create_chat_input.submit(
create_send_message,
inputs=[create_chatbot, create_chat_input, state, uuid_str],
outputs=[
create_chatbot, user_chat_bot_cover, user_chatbot,
user_chat_bot_suggest, create_chat_input
])
# 配置 "Configure" 标签页的提交按钮功能
configure_button.click(
process_configuration,
inputs=[
uuid_str, bot_avatar_comp, name_input, description_input,
instructions_input, model_selector, agent_language_selector,
suggestion_input, knowledge_input, capabilities_checkboxes,
openapi_schema, openapi_auth_type, openapi_auth_apikey,
openapi_auth_apikey_type, openapi_privacy_policy, state
],
outputs=[
user_chat_bot_cover, user_chatbot, user_chat_bot_suggest,
suggestion_input
])
# 配置 "Preview" 的消息发送功能
preview_chat_input.submit(
preview_send_message,
inputs=[user_chatbot, preview_chat_input, state, uuid_str],
outputs=[user_chatbot, user_chat_bot_cover, preview_chat_input])
# configuration for publish
publish_button.click(
publish_agent,
inputs=[name_input, uuid_str, state],
outputs=[publish_link],
)
import_button.click(
import_space,
inputs=[update_space, uuid_str, state],
outputs=configure_updated_outputs,
)
language.select(
change_lang,
inputs=[language],
outputs=configure_updated_outputs + [
configure_button, create_chat_input, open_api_accordion,
preview_header, preview_chat_input, publish_accordion, header,
publish_alert_md, build_hint_md, publish_hint_md
])
demo.load(
init_all, inputs=[uuid_str, state], outputs=configure_updated_outputs)
def gen_response_and_process(agent,
query: str,
memory: MemoryWithRetrievalKnowledge,
uuid_str: str,
print_info: bool = False):
def beauty_output(response: str, step_result: str):
def create_send_message(chatbot, input, _state, uuid_str):
uuid_str = check_uuid(uuid_str)
# 将发送的消息添加到聊天历史
builder_agent = _state['builder_agent']
builder_memory = _state['builder_memory']
chatbot.append([{'text': input.text, 'files': input.files}, None])
yield {
create_chatbot: chatbot,
create_chat_input: None,
}
response = ''
for frame in gen_response_and_process(
builder_agent,
query=input.text,
memory=builder_memory,
print_info=True,
uuid_str=uuid_str):
llm_result = frame.get('llm_text', '')
exec_result = frame.get('exec_result', '')
step_result = frame.get('step', '')
logger.query_info(
uuid=uuid_str, message='frame', details={'frame': str(frame)})
if len(exec_result) != 0:
if isinstance(exec_result, dict):
exec_result = exec_result['result']
assert isinstance(exec_result, Config)
yield format_message_with_builder_cfg(
_state,
chatbot,
exec_result.to_dict(),
uuid_str=uuid_str)
else:
# llm result
if isinstance(llm_result, dict):
content = llm_result['content']
else:
content = llm_result
frame_text = content
response = beauty_output(f'{response}{frame_text}',
step_result)
chatbot[-1][1] = response
yield {
create_chatbot: chatbot,
} | null |
179,759 | import os
import random
import shutil
import traceback
import gradio as gr
import json
import modelscope_gradio_components as mgr
import yaml
from builder_core import (beauty_output, gen_response_and_process,
init_builder_chatbot_agent)
from config_utils import (DEFAULT_AGENT_DIR, Config, get_avatar_image,
get_ci_dir, get_user_dir,
is_valid_plugin_configuration, parse_configuration,
save_avatar_image, save_builder_configuration,
save_plugin_configuration)
from gradio_utils import format_cover_html, format_goto_publish_html
from i18n import I18n
from modelscope_agent.schemas import Message
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_gradio_components.components.Chatbot.llm_thinking_presets import \
qwen
from publish_util import (pop_user_info_from_config, prepare_agent_zip,
reload_agent_zip)
from user_core import init_user_chatbot_agent
def init_user(uuid_str, state):
try:
seed = state.get('session_seed', random.randint(0, 1000000000))
user_agent, user_memory = init_user_chatbot_agent(uuid_str)
user_agent.seed = seed
state['user_agent'] = user_agent
state['user_memory'] = user_memory
except Exception as e:
logger.query_error(
uuid=uuid_str,
error=str(e),
details={'error_traceback': traceback.format_exc()})
return state
def update_builder(uuid_str, state):
try:
builder_agent, builder_memory = init_builder_chatbot_agent(uuid_str)
state['builder_agent'] = builder_agent
state['builder_memory'] = builder_memory
except Exception as e:
logger.query_error(
uuid=uuid_str,
error=str(e),
details={'error_traceback': traceback.format_exc()})
return state
def check_uuid(uuid_str):
if not uuid_str or uuid_str == '':
if os.getenv('MODELSCOPE_ENVIRONMENT') == 'studio':
raise gr.Error('请登陆后使用! (Please login first)')
else:
uuid_str = 'local_user'
return uuid_str
gr.Blocks(css='assets/app.css')
def get_user_dir(uuid_str=''):
return os.path.join(DEFAULT_BUILDER_CONFIG_DIR, uuid_str)
def save_builder_configuration(builder_cfg, uuid_str=''):
builder_cfg_file = get_user_cfg_file(uuid_str)
if uuid_str != '' and not os.path.exists(
os.path.dirname(builder_cfg_file)):
os.makedirs(os.path.dirname(builder_cfg_file))
with open(builder_cfg_file, 'w', encoding='utf-8') as f:
f.write(json.dumps(builder_cfg, indent=2, ensure_ascii=False))
def is_valid_plugin_configuration(openapi_plugin_cfg):
if 'schema' in openapi_plugin_cfg:
schema = openapi_plugin_cfg['schema']
if isinstance(schema, dict):
return True
else:
return False
def save_plugin_configuration(openapi_plugin_cfg, uuid_str):
openapi_plugin_cfg_file = get_user_openapi_plugin_cfg_file(uuid_str)
if uuid_str != '' and not os.path.exists(
os.path.dirname(openapi_plugin_cfg_file)):
os.makedirs(os.path.dirname(openapi_plugin_cfg_file))
with open(openapi_plugin_cfg_file, 'w', encoding='utf-8') as f:
f.write(json.dumps(openapi_plugin_cfg, indent=2, ensure_ascii=False))
def get_avatar_image(bot_avatar, uuid_str=''):
user_avatar_path = os.path.join(
os.path.dirname(__file__), 'assets/user.jpg')
bot_avatar_path = os.path.join(os.path.dirname(__file__), 'assets/bot.jpg')
if len(bot_avatar) > 0:
bot_avatar_path = os.path.join(DEFAULT_BUILDER_CONFIG_DIR, uuid_str,
bot_avatar)
if uuid_str != '':
# use default if not exists
if not os.path.exists(bot_avatar_path):
# create parents directory
os.makedirs(os.path.dirname(bot_avatar_path), exist_ok=True)
# copy the template to the address
temp_bot_avatar_path = os.path.join(DEFAULT_BUILDER_CONFIG_DIR,
bot_avatar)
if not os.path.exists(temp_bot_avatar_path):
# fall back to default local avatar image
temp_bot_avatar_path = os.path.join('./config', bot_avatar)
if not os.path.exists(temp_bot_avatar_path):
temp_bot_avatar_path = os.path.join(
'./config', 'custom_bot_avatar.png')
shutil.copy(temp_bot_avatar_path, bot_avatar_path)
return [user_avatar_path, bot_avatar_path]
def save_avatar_image(image_path, uuid_str=''):
bot_avatar = os.path.basename(image_path)
bot_avatar_path = os.path.join(DEFAULT_BUILDER_CONFIG_DIR, uuid_str,
bot_avatar)
shutil.copy(image_path, bot_avatar_path)
return bot_avatar, bot_avatar_path
def process_configuration(uuid_str, bot_avatar, name, description,
instructions, model, agent_language, suggestions,
knowledge_files, capabilities_checkboxes,
openapi_schema, openapi_auth,
openapi_auth_apikey, openapi_auth_apikey_type,
openapi_privacy_policy, state):
uuid_str = check_uuid(uuid_str)
tool_cfg = state['tool_cfg']
capabilities = state['capabilities']
bot_avatar, bot_avatar_path = save_avatar_image(bot_avatar, uuid_str)
suggestions_filtered = [row for row in suggestions if row[0]]
if len(suggestions_filtered) == 0:
suggestions_filtered = [['']]
user_dir = get_user_dir(uuid_str)
if knowledge_files is not None:
new_knowledge_files = [
os.path.join(user_dir, os.path.basename((f.name)))
for f in knowledge_files
]
for src_file, dst_file in zip(knowledge_files,
new_knowledge_files):
if not os.path.exists(dst_file):
shutil.copy(src_file.name, dst_file)
else:
new_knowledge_files = []
builder_cfg = {
'name': name,
'avatar': bot_avatar,
'description': description,
'instruction': instructions,
'prompt_recommend': [row[0] for row in suggestions_filtered],
'knowledge': new_knowledge_files,
'tools': {
capability:
dict([(key, value if key != 'use' else
(capability in capabilities_checkboxes))
for key, value in tool_cfg[capability].items()])
for capability in map(lambda item: item[1], capabilities)
},
'model': model,
'language': agent_language,
}
try:
try:
schema_dict = json.loads(openapi_schema)
except json.decoder.JSONDecodeError:
schema_dict = yaml.safe_load(openapi_schema)
except Exception as e:
raise gr.Error(
f'OpenAPI schema format error, should be one of json and yaml: {e}'
)
openapi_plugin_cfg = {
'schema': schema_dict,
'auth': {
'type': openapi_auth,
'apikey': openapi_auth_apikey,
'apikey_type': openapi_auth_apikey_type
},
'privacy_policy': openapi_privacy_policy
}
if is_valid_plugin_configuration(openapi_plugin_cfg):
save_plugin_configuration(openapi_plugin_cfg, uuid_str)
except Exception as e:
logger.query_error(
uuid=uuid_str,
error=str(e),
details={'error_traceback': traceback.format_exc()})
save_builder_configuration(builder_cfg, uuid_str)
update_builder(uuid_str, state)
init_user(uuid_str, state)
return {
user_chat_bot_cover:
gr.HTML(
visible=True,
value=format_cover_html(builder_cfg, bot_avatar_path)),
user_chatbot:
gr.update(
visible=False,
avatar_images=get_avatar_image(bot_avatar, uuid_str)),
suggestion_input: [item[:] for item in suggestions_filtered],
user_chat_bot_suggest:
gr.Dataset(
components=[preview_chat_input], samples=suggestions_filtered),
} | null |
179,760 | import os
import random
import shutil
import traceback
import gradio as gr
import json
import modelscope_gradio_components as mgr
import yaml
from builder_core import (beauty_output, gen_response_and_process,
init_builder_chatbot_agent)
from config_utils import (DEFAULT_AGENT_DIR, Config, get_avatar_image,
get_ci_dir, get_user_dir,
is_valid_plugin_configuration, parse_configuration,
save_avatar_image, save_builder_configuration,
save_plugin_configuration)
from gradio_utils import format_cover_html, format_goto_publish_html
from i18n import I18n
from modelscope_agent.schemas import Message
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_gradio_components.components.Chatbot.llm_thinking_presets import \
qwen
from publish_util import (pop_user_info_from_config, prepare_agent_zip,
reload_agent_zip)
from user_core import init_user_chatbot_agent
gr.Blocks(css='assets/app.css')
def get_ci_dir():
return DEFAULT_CODE_INTERPRETER_DIR
class Message(BaseModel):
"""
Message: message information
"""
role: str = ''
content: str = ''
def preview_send_message(chatbot, input, _state, uuid_str):
# 将发送的消息添加到聊天历史
# _uuid_str = check_uuid(uuid_str)
user_agent = _state['user_agent']
append_files = []
for file in input.files:
file_name = os.path.basename(file.path)
# covert xxx.json to xxx_uuid_str.json
file_name = file_name.replace('.', f'_{uuid_str}.')
file_path = os.path.join(get_ci_dir(), file_name)
if not os.path.exists(file_path):
# make sure file path's directory exists
os.makedirs(os.path.dirname(file_path), exist_ok=True)
shutil.copy(file.path, file_path)
append_files.append(file_path)
user_memory = _state['user_memory']
chatbot.append([{'text': input.text, 'files': input.files}, None])
yield {
user_chatbot: mgr.Chatbot(visible=True, value=chatbot),
user_chat_bot_cover: gr.HTML(visible=False),
preview_chat_input: None
}
# get chat history from memory
history = user_memory.get_history()
# get knowledge from memory, currently get one file
uploaded_file = None
if len(append_files) > 0:
uploaded_file = append_files[0]
ref_doc = user_memory.run(
query=input.text,
url=uploaded_file,
max_token=4000,
top_k=2,
checked=True)
response = ''
try:
for frame in user_agent.run(
input.text,
history=history,
ref_doc=ref_doc,
append_files=append_files,
uuid_str=uuid_str):
# append_files=new_file_paths):
# important! do not change this
response += frame
chatbot[-1][1] = response
yield {user_chatbot: chatbot}
if len(history) == 0:
user_memory.update_history(
Message(role='system', content=user_agent.system_prompt))
user_memory.update_history([
Message(role='user', content=input.text),
Message(role='assistant', content=response),
])
except Exception as e:
if 'dashscope.common.error.AuthenticationError' in str(e):
msg = 'DASHSCOPE_API_KEY should be set via environment variable. You can acquire this in ' \
'https://help.aliyun.com/zh/dashscope/developer-reference/activate-dashscope-and-create-an-api-key'
elif 'rate limit' in str(e):
msg = 'Too many people are calling, please try again later.'
else:
msg = str(e)
chatbot[-1][1] = msg
yield {user_chatbot: chatbot} | null |
179,761 | import os
import random
import shutil
import traceback
import gradio as gr
import json
import modelscope_gradio_components as mgr
import yaml
from builder_core import (beauty_output, gen_response_and_process,
init_builder_chatbot_agent)
from config_utils import (DEFAULT_AGENT_DIR, Config, get_avatar_image,
get_ci_dir, get_user_dir,
is_valid_plugin_configuration, parse_configuration,
save_avatar_image, save_builder_configuration,
save_plugin_configuration)
from gradio_utils import format_cover_html, format_goto_publish_html
from i18n import I18n
from modelscope_agent.schemas import Message
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_gradio_components.components.Chatbot.llm_thinking_presets import \
qwen
from publish_util import (pop_user_info_from_config, prepare_agent_zip,
reload_agent_zip)
from user_core import init_user_chatbot_agent
def check_uuid(uuid_str):
if not uuid_str or uuid_str == '':
if os.getenv('MODELSCOPE_ENVIRONMENT') == 'studio':
raise gr.Error('请登陆后使用! (Please login first)')
else:
uuid_str = 'local_user'
return uuid_str
DEFAULT_AGENT_DIR = '/tmp/agentfabric'
def pop_user_info_from_config(src_dir, uuid_str):
""" Remove all personal information from the configuration files and return this data.
The purpose of this is to ensure that personal information is not stored in plain text
when releasing.
Args:
src_dir (str): config root path
uuid_str (str): user id
"""
user_info = {}
# deal with plugin cfg
plugin_config_path = f'{src_dir}/config/{uuid_str}/openapi_plugin_config.json'
if os.path.exists(plugin_config_path):
with open(plugin_config_path, 'r') as f:
plugin_config = json.load(f)
if 'auth' in plugin_config:
if plugin_config['auth']['type'] == 'API Key':
user_info['apikey'] = plugin_config['auth'].pop('apikey')
user_info['apikey_type'] = plugin_config['auth'].pop(
'apikey_type')
with open(plugin_config_path, 'w') as f:
json.dump(plugin_config, f, indent=2, ensure_ascii=False)
return user_info
def prepare_agent_zip(agent_name, src_dir, uuid_str, state):
# 设置阿里云OSS的认证信息
local_file = os.path.abspath(os.path.dirname(__file__))
ak_id, ak_secret, endpoint, bucket_name = get_oss_config()
auth = oss2.Auth(ak_id, ak_secret)
bucket = oss2.Bucket(auth, endpoint, bucket_name)
new_directory = f'{src_dir}/upload/{uuid_str}' # 新目录的路径
# 创建新目录
if os.path.exists(new_directory):
shutil.rmtree(new_directory)
os.makedirs(new_directory)
# 复制config下的uuid_str目录到new_directory下并改名为local_user
uuid_str_path = f'{src_dir}/config/{uuid_str}' # 指向uuid_str目录的路径
local_user_path = f'{new_directory}/config' # 新的目录路径
shutil.copytree(uuid_str_path, local_user_path, dirs_exist_ok=True)
target_conf = os.path.join(local_user_path, 'builder_config.json')
builder_cfg = Config.from_file(target_conf)
builder_cfg.knowledge = [
'config/' + f.split('/')[-1] for f in builder_cfg.knowledge
]
with open(target_conf, 'w') as f:
json.dump(builder_cfg.to_dict(), f, indent=2, ensure_ascii=False)
# 复制config目录下所有.json文件到new_directory/config
config_path = f'{local_file}/config'
new_config_path = f'{new_directory}/config'
def find_json_and_images(directory):
# 确保路径以斜杠结束
directory = os.path.join(directory, '')
# 找到所有的JSON文件
json_files = [
os.path.join(directory, 'model_config.json'),
os.path.join(directory, 'tool_config.json'),
]
# 找到所有的图片文件
image_files = glob.glob(directory + '*.png') + \
glob.glob(directory + '*.jpg') + \
glob.glob(directory + '*.jpeg') + \
glob.glob(directory + '*.gif') # 根据需要可以添加更多图片格式
return json_files + image_files
for f in find_json_and_images(config_path):
if not os.path.exists(
os.path.join(new_config_path, os.path.basename(f))):
shutil.copy(f, new_config_path)
# 复制assets目录到new_directory
assets_path = f'{local_file}/assets'
new_assets_path = f'{new_directory}/assets'
shutil.copytree(assets_path, new_assets_path, dirs_exist_ok=True)
# 在requirements.txt中添加新的行
requirements_file = f'{local_file}/requirements.txt'
new_requirements_file = f'{new_directory}/requirements.txt'
modelscope_agent_pkg = DEFAULT_MS_PKG.replace('version', MS_VERSION)
with open(requirements_file, 'r') as file:
content = file.readlines()
with open(new_requirements_file, 'w') as file:
file.write(modelscope_agent_pkg + '\n')
file.writelines(content)
# 复制.py文件到新目录
for file in os.listdir(local_file):
if file.endswith('.py'):
shutil.copy(f'{local_file}/{file}', new_directory)
# 打包新目录
archive_path = shutil.make_archive(new_directory, 'zip', new_directory)
# 使用抽象出的函数上传到OSS并设置权限
file_url = upload_to_oss(bucket, archive_path,
f'agents/user/{uuid_str}/{agent_name}.zip')
shutil.rmtree(new_directory)
# 获取必须设置的envs
envs_required = {}
for t, t_cfg in builder_cfg.tools.items():
if t == 'amap_weather' and t_cfg['is_active'] and t_cfg['use']:
envs_required['AMAP_TOKEN'] = 'Your-AMAP-TOKEN'
elif t in ['web_search'] and t_cfg['is_active'] and t_cfg['use']:
envs_required[
'BING_SEARCH_V7_SUBSCRIPTION_KEY'] = 'Yor-BingSearch-KEY'
elif t in ['speech-generation', 'video-generation'
] and t_cfg['is_active'] and t_cfg['use']:
envs_required['MODELSCOPE_API_TOKEN'] = 'Your-MODELSCOPE_API_TOKEN'
return file_url, envs_required
def publish_agent(name, uuid_str, state):
uuid_str = check_uuid(uuid_str)
env_params = {}
env_params.update(
pop_user_info_from_config(DEFAULT_AGENT_DIR, uuid_str))
output_url, envs_required = prepare_agent_zip(name, DEFAULT_AGENT_DIR,
uuid_str, state)
env_params.update(envs_required)
# output_url = "https://test.url"
return format_goto_publish_html(
i18n.get_whole('publish'), output_url, env_params) | null |
179,762 | import os
import random
import shutil
import traceback
import gradio as gr
import json
import modelscope_gradio_components as mgr
import yaml
from builder_core import (beauty_output, gen_response_and_process,
init_builder_chatbot_agent)
from config_utils import (DEFAULT_AGENT_DIR, Config, get_avatar_image,
get_ci_dir, get_user_dir,
is_valid_plugin_configuration, parse_configuration,
save_avatar_image, save_builder_configuration,
save_plugin_configuration)
from gradio_utils import format_cover_html, format_goto_publish_html
from i18n import I18n
from modelscope_agent.schemas import Message
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_gradio_components.components.Chatbot.llm_thinking_presets import \
qwen
from publish_util import (pop_user_info_from_config, prepare_agent_zip,
reload_agent_zip)
from user_core import init_user_chatbot_agent
def check_uuid(uuid_str):
if not uuid_str or uuid_str == '':
if os.getenv('MODELSCOPE_ENVIRONMENT') == 'studio':
raise gr.Error('请登陆后使用! (Please login first)')
else:
uuid_str = 'local_user'
return uuid_str
with demo:
uuid_str = gr.Textbox(label='modelscope_uuid', visible=False)
draw_seed = random.randint(0, 1000000000)
state = gr.State({'session_seed': draw_seed})
i18n = I18n('zh-cn')
with gr.Row():
with gr.Column(scale=5):
header = gr.Markdown(i18n.get('header'))
with gr.Column(scale=1):
language = gr.Dropdown(
choices=[('中文', 'zh-cn'), ('English', 'en')],
show_label=False,
container=False,
value='zh-cn',
interactive=True)
with gr.Row():
with gr.Column():
with gr.Tabs() as tabs:
with gr.Tab(i18n.get_whole('create'), id=0) as create_tab:
with gr.Column():
# "Create" 标签页的 Chatbot 组件
start_text = '欢迎使用agent创建助手。我可以帮助您创建一个定制agent。'\
'您希望您的agent主要用于什么领域或任务？比如，您可以说，我想做一个RPG游戏agent'
create_chatbot = mgr.Chatbot(
show_label=False,
value=[[None, start_text]],
flushing=False,
show_copy_button=True,
llm_thinking_presets=[
qwen(
action_input_title='调用 <Action>',
action_output_title='完成调用')
])
create_chat_input = mgr.MultimodalInput(
label=i18n.get('message'),
placeholder=i18n.get('message_placeholder'),
interactive=False,
upload_button_props=dict(visible=False),
submit_button_props=dict(
label=i18n.get('sendOnLoading')))
configure_tab = gr.Tab(i18n.get_whole('configure'), id=1)
with configure_tab:
with gr.Column():
# "Configure" 标签页的配置输入字段
with gr.Row():
bot_avatar_comp = gr.Image(
label=i18n.get('form_avatar'),
sources=['upload'],
interactive=True,
type='filepath',
scale=1,
width=182,
height=182,
)
with gr.Column(scale=4):
name_input = gr.Textbox(
label=i18n.get('form_name'),
placeholder=i18n.get(
'form_name_placeholder'))
description_input = gr.Textbox(
label=i18n.get('form_description'),
placeholder=i18n.get(
'form_description_placeholder'))
instructions_input = gr.Textbox(
label=i18n.get('form_instructions'),
placeholder=i18n.get(
'form_instructions_placeholder'),
lines=3)
model_selector = gr.Dropdown(
label=i18n.get('form_model'))
agent_language_selector = gr.Dropdown(
label=i18n.get('form_agent_language'),
choices=['zh', 'en'],
value='zh')
suggestion_input = gr.Dataframe(
show_label=False,
value=[['']],
datatype=['str'],
headers=[i18n.get_whole('form_prompt_suggestion')],
type='array',
col_count=(1, 'fixed'),
interactive=True)
gr.Markdown(
'*注意：知识库上传的文本文档默认按照\\n\\n切分，pdf默认按照页切分。如果片段'
'对应的字符大于[配置文件](https://github.com/modelscope/modelscope-agent/'
'blob/master/apps/agentfabric/config/model_config.json)中指定模型的'
'knowledge限制，则在被召回时有可能会被截断。*')
knowledge_input = gr.File(
label=i18n.get('form_knowledge'),
file_count='multiple',
file_types=[
'text', '.json', '.csv', '.pdf', '.md'
])
capabilities_checkboxes = gr.CheckboxGroup(
label=i18n.get('form_capabilities'))
with gr.Accordion(
i18n.get('open_api_accordion'),
open=False) as open_api_accordion:
openapi_schema = gr.Textbox(
label='Schema',
placeholder=
'Enter your OpenAPI schema here, JSON or YAML format only'
)
with gr.Group():
openapi_auth_type = gr.Radio(
label='Authentication Type',
choices=['None', 'API Key'],
value='None')
openapi_auth_apikey = gr.Textbox(
label='API Key',
placeholder='Enter your API Key here')
openapi_auth_apikey_type = gr.Radio(
label='API Key type', choices=['Bearer'])
openapi_privacy_policy = gr.Textbox(
label='Privacy Policy',
placeholder='Enter privacy policy URL')
configure_button = gr.Button(
i18n.get('form_update_button'))
with gr.Accordion(
label=i18n.get('import_config'),
open=False) as update_accordion:
with gr.Column():
update_space = gr.Textbox(
label=i18n.get('space_addr'),
placeholder=i18n.get('input_space_addr'))
import_button = gr.Button(
i18n.get_whole('import_space'))
gr.Markdown(
f'#### {i18n.get_whole("import_hint")}')
with gr.Column():
# Preview
preview_header = gr.HTML(
f"""<div class="preview_header">{i18n.get('preview')}<div>""")
user_chat_bot_cover = gr.HTML(format_cover_html({}, None))
user_chatbot = mgr.Chatbot(
value=[[None, None]],
elem_id='user_chatbot',
elem_classes=['markdown-body'],
avatar_images=get_avatar_image('', uuid_str),
height=650,
latex_delimiters=[],
show_label=False,
visible=False,
show_copy_button=True,
llm_thinking_presets=[
qwen(
action_input_title='调用 <Action>',
action_output_title='完成调用')
])
preview_chat_input = mgr.MultimodalInput(
interactive=False,
label=i18n.get('message'),
placeholder=i18n.get('message_placeholder'),
upload_button_props=dict(
label=i18n.get('upload_btn'),
file_types=['file', 'image', 'audio', 'video', 'text'],
file_count='multiple'),
submit_button_props=dict(label=i18n.get('sendOnLoading')))
user_chat_bot_suggest = gr.Dataset(
label=i18n.get('prompt_suggestion'),
components=[preview_chat_input],
samples=[])
user_chat_bot_suggest.select(
lambda evt: evt[0],
inputs=[user_chat_bot_suggest],
outputs=[preview_chat_input])
with gr.Accordion(
label=i18n.get('publish'),
open=False) as publish_accordion:
publish_alert_md = gr.Markdown(f'{i18n.get("publish_alert")}')
with gr.Row():
with gr.Column():
publish_button = gr.Button(i18n.get_whole('build'))
build_hint_md = gr.Markdown(
f'#### 1.{i18n.get("build_hint")}')
with gr.Column():
publish_link = gr.HTML(
value=format_goto_publish_html(
i18n.get_whole('publish'), '', {}, True))
publish_hint_md = gr.Markdown(
f'#### 2.{i18n.get("publish_hint")}')
configure_updated_outputs = [
state,
# config form
bot_avatar_comp,
name_input,
description_input,
instructions_input,
model_selector,
agent_language_selector,
suggestion_input,
knowledge_input,
capabilities_checkboxes,
# bot
user_chat_bot_cover,
user_chat_bot_suggest,
preview_chat_input,
create_chat_input,
]
# 初始化表单
def init_ui_config(uuid_str, _state, builder_cfg, model_cfg, tool_cfg):
logger.query_info(
uuid=uuid_str,
message='builder_cfg',
details={'builder_cfg': str(builder_cfg)})
# available models
models = list(model_cfg.keys())
capabilities = [(tool_cfg[tool_key]['name'], tool_key)
for tool_key in tool_cfg.keys()
if tool_cfg[tool_key].get('is_active', False)]
_state['model_cfg'] = model_cfg
_state['tool_cfg'] = tool_cfg
_state['capabilities'] = capabilities
bot_avatar = get_avatar_image(builder_cfg.get('avatar', ''),
uuid_str)[1]
suggests = builder_cfg.get('prompt_recommend', [''])
return {
state:
_state,
bot_avatar_comp:
gr.Image(value=bot_avatar),
name_input:
builder_cfg.get('name', ''),
description_input:
builder_cfg.get('description'),
instructions_input:
builder_cfg.get('instruction'),
model_selector:
gr.Dropdown(
value=builder_cfg.get('model', models[0]), choices=models),
agent_language_selector:
builder_cfg.get('language') or 'zh',
suggestion_input:
[[str] for str in suggests] if len(suggests) > 0 else [['']],
knowledge_input:
builder_cfg.get('knowledge', [])
if len(builder_cfg['knowledge']) > 0 else None,
capabilities_checkboxes:
gr.CheckboxGroup(
value=[
tool for tool in builder_cfg.get('tools', {}).keys()
if builder_cfg.get('tools').get(tool).get('use', False)
],
choices=capabilities),
# bot
user_chat_bot_cover:
format_cover_html(builder_cfg, bot_avatar),
user_chat_bot_suggest:
gr.Dataset(
components=[preview_chat_input],
samples=[[item] for item in suggests]),
}
# tab 切换的事件处理
configure_tab.select(
on_congifure_tab_select,
inputs=[state, uuid_str],
outputs=configure_updated_outputs)
# 配置 "Create" 标签页的消息发送功能
create_chat_input.submit(
create_send_message,
inputs=[create_chatbot, create_chat_input, state, uuid_str],
outputs=[
create_chatbot, user_chat_bot_cover, user_chatbot,
user_chat_bot_suggest, create_chat_input
])
# 配置 "Configure" 标签页的提交按钮功能
configure_button.click(
process_configuration,
inputs=[
uuid_str, bot_avatar_comp, name_input, description_input,
instructions_input, model_selector, agent_language_selector,
suggestion_input, knowledge_input, capabilities_checkboxes,
openapi_schema, openapi_auth_type, openapi_auth_apikey,
openapi_auth_apikey_type, openapi_privacy_policy, state
],
outputs=[
user_chat_bot_cover, user_chatbot, user_chat_bot_suggest,
suggestion_input
])
# 配置 "Preview" 的消息发送功能
preview_chat_input.submit(
preview_send_message,
inputs=[user_chatbot, preview_chat_input, state, uuid_str],
outputs=[user_chatbot, user_chat_bot_cover, preview_chat_input])
# configuration for publish
publish_button.click(
publish_agent,
inputs=[name_input, uuid_str, state],
outputs=[publish_link],
)
import_button.click(
import_space,
inputs=[update_space, uuid_str, state],
outputs=configure_updated_outputs,
)
language.select(
change_lang,
inputs=[language],
outputs=configure_updated_outputs + [
configure_button, create_chat_input, open_api_accordion,
preview_header, preview_chat_input, publish_accordion, header,
publish_alert_md, build_hint_md, publish_hint_md
])
demo.load(
init_all, inputs=[uuid_str, state], outputs=configure_updated_outputs)
DEFAULT_AGENT_DIR = '/tmp/agentfabric'
def parse_configuration(uuid_str=''):
"""parse configuration
Args:
Returns:
dict: parsed configuration
"""
model_cfg_file = os.getenv('MODEL_CONFIG_FILE', DEFAULT_MODEL_CONFIG_FILE)
builder_cfg_file = get_user_cfg_file(uuid_str)
# use default if not exists
if not os.path.exists(builder_cfg_file):
# create parents directory
os.makedirs(os.path.dirname(builder_cfg_file), exist_ok=True)
# copy the template to the address
builder_cfg_file_temp = './config/builder_config.json'
if builder_cfg_file_temp != builder_cfg_file:
shutil.copy(builder_cfg_file_temp, builder_cfg_file)
tool_cfg_file = os.getenv('TOOL_CONFIG_FILE', DEFAULT_TOOL_CONFIG_FILE)
builder_cfg = Config.from_file(builder_cfg_file)
model_cfg = Config.from_file(model_cfg_file)
tool_cfg = Config.from_file(tool_cfg_file)
tools_info = builder_cfg.tools
available_tool_list = []
for key, value in tools_info.items():
if value['use']:
available_tool_list.append(key)
tool_cfg[key]['use'] = value['use']
openapi_plugin_file = get_user_openapi_plugin_cfg_file(uuid_str)
plugin_cfg = {}
available_plugin_list = []
if os.path.exists(openapi_plugin_file):
openapi_plugin_cfg = Config.from_file(openapi_plugin_file)
try:
config_dict = openapi_schema_convert(
schema=openapi_plugin_cfg.schema,
auth=openapi_plugin_cfg.auth.to_dict())
plugin_cfg = Config(config_dict)
for name, config in config_dict.items():
available_plugin_list.append(name)
except Exception as e:
logger.query_error(
uuid=uuid_str,
error=str(e),
content={
'error_traceback':
traceback.format_exc(),
'error_details':
'The format of the plugin config file is incorrect.'
})
return builder_cfg, model_cfg, tool_cfg, available_tool_list, plugin_cfg, available_plugin_list
agent_zip(agent_url, dst_dir, uuid_str, state):
# download zip from agent_url, and unzip to dst_dir/uuid_str
# 从URL中解析出文件名
parsed_url = urlparse(agent_url)
filename = os.path.basename(parsed_url.path)
zip_path = os.path.join(dst_dir, filename)
# 提取agent_name（去掉'.zip'）
agent_name, _ = os.path.splitext(filename)
# 创建临时解压目录
temp_extract_dir = os.path.join(dst_dir, f'temp_{uuid_str}')
if os.path.exists(temp_extract_dir):
shutil.rmtree(temp_extract_dir)
os.makedirs(temp_extract_dir)
# 下载ZIP文件
response = requests.get(agent_url)
if response.status_code == 200:
with open(zip_path, 'wb') as file:
file.write(response.content)
else:
raise RuntimeError(
f'download file from {agent_url} error:\n {response.reason}')
# 解压ZIP文件到临时目录
with zipfile.ZipFile(zip_path, 'r') as zip_ref:
zip_ref.extractall(temp_extract_dir)
# 解析version信息
version = parse_version_from_file(
os.path.join(temp_extract_dir, '/version.py'))
print(f'agent fabric version: {version}')
# 创建目标config路径
target_config_path = os.path.join(dst_dir, 'config', uuid_str)
if os.path.exists(target_config_path):
shutil.rmtree(target_config_path)
os.makedirs(target_config_path)
# 复制config目录
# 兼容老版本配置放到local_user目录下，以及新版本直接放在config目录下
if os.path.exists(os.path.join(temp_extract_dir, 'config', 'local_user')):
config_source_path = os.path.join(temp_extract_dir, 'config',
'local_user')
elif os.path.exists(os.path.join(temp_extract_dir, 'config')):
config_source_path = os.path.join(temp_extract_dir, 'config')
else:
raise RuntimeError('未找到正确的配置文件信息')
if os.path.exists(config_source_path):
for item in os.listdir(config_source_path):
s = os.path.join(config_source_path, item)
d = os.path.join(target_config_path, item)
if os.path.isdir(s):
shutil.copytree(s, d, dirs_exist_ok=True)
else:
shutil.copy2(s, d)
# 清理：删除临时目录和下载的ZIP文件
shutil.rmtree(temp_extract_dir)
os.remove(zip_path)
# 修改知识库路径 config/xxx to /tmp/agentfabric/config/$uuid/xxx
target_conf = os.path.join(target_config_path, 'builder_config.json')
builder_cfg = Config.from_file(target_conf)
builder_cfg.knowledge = [
f'{target_config_path}/' + f.split('/')[-1]
for f in builder_cfg.knowledge
]
with open(target_conf, 'w') as f:
json.dump(builder_cfg.to_dict(), f, indent=2, ensure_ascii=False)
return agent_name
def import_space(agent_url, uuid_str, state):
uuid_str = check_uuid(uuid_str)
_ = reload_agent_zip(agent_url, DEFAULT_AGENT_DIR, uuid_str, state)
# update config
builder_cfg, model_cfg, tool_cfg, available_tool_list, _, _ = parse_configuration(
uuid_str)
return init_ui_config(uuid_str, state, builder_cfg, model_cfg,
tool_cfg) | null |
179,763 | import os
import random
import shutil
import traceback
import gradio as gr
import json
import modelscope_gradio_components as mgr
import yaml
from builder_core import (beauty_output, gen_response_and_process,
init_builder_chatbot_agent)
from config_utils import (DEFAULT_AGENT_DIR, Config, get_avatar_image,
get_ci_dir, get_user_dir,
is_valid_plugin_configuration, parse_configuration,
save_avatar_image, save_builder_configuration,
save_plugin_configuration)
from gradio_utils import format_cover_html, format_goto_publish_html
from i18n import I18n
from modelscope_agent.schemas import Message
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_gradio_components.components.Chatbot.llm_thinking_presets import \
qwen
from publish_util import (pop_user_info_from_config, prepare_agent_zip,
reload_agent_zip)
from user_core import init_user_chatbot_agent
gr.Blocks(css='assets/app.css')
class I18n():
def __init__(self, lang):
self.lang = lang
self.langIndex = support_lang.index(lang)
def get(self, field):
return i18n.get(field)[self.langIndex]
def get_whole(self, field):
return f'{i18n.get(field)[0]}({i18n.get(field)[1]})'
def change_lang(language):
i18n = I18n(language)
return {
bot_avatar_comp:
gr.Image(label=i18n.get('form_avatar')),
name_input:
gr.Textbox(
label=i18n.get('form_name'),
placeholder=i18n.get('form_name_placeholder')),
description_input:
gr.Textbox(
label=i18n.get('form_description'),
placeholder=i18n.get('form_description_placeholder')),
instructions_input:
gr.Textbox(
label=i18n.get('form_instructions'),
placeholder=i18n.get('form_instructions_placeholder')),
model_selector:
gr.Dropdown(label=i18n.get('form_model')),
agent_language_selector:
gr.Dropdown(label=i18n.get('form_agent_language')),
knowledge_input:
gr.File(label=i18n.get('form_knowledge')),
capabilities_checkboxes:
gr.CheckboxGroup(label=i18n.get('form_capabilities')),
open_api_accordion:
gr.Accordion(label=i18n.get('open_api_accordion')),
configure_button:
gr.Button(i18n.get('form_update_button')),
preview_header:
gr.HTML(
f"""<div class="preview_header">{i18n.get('preview')}<div>"""),
preview_chat_input:
mgr.MultimodalInput(
label=i18n.get('message'),
placeholder=i18n.get('message_placeholder'),
submit_button_props=dict(label=i18n.get('send')),
upload_button_props=dict(
label=i18n.get('upload_btn'),
file_types=['file', 'image', 'audio', 'video', 'text'],
file_count='multiple')),
create_chat_input:
mgr.MultimodalInput(
label=i18n.get('message'),
placeholder=i18n.get('message_placeholder'),
submit_button_props=dict(label=i18n.get('send'))),
user_chat_bot_suggest:
gr.Dataset(
components=[preview_chat_input],
label=i18n.get('prompt_suggestion')),
publish_accordion:
gr.Accordion(label=i18n.get('publish')),
header:
gr.Markdown(i18n.get('header')),
publish_alert_md:
gr.Markdown(f'{i18n.get("publish_alert")}'),
build_hint_md:
gr.Markdown(f'#### 1.{i18n.get("build_hint")}'),
publish_hint_md:
gr.Markdown(f'#### 2.{i18n.get("publish_hint")}'),
} | null |
179,764 | import os
import random
import shutil
import traceback
import gradio as gr
import json
import modelscope_gradio_components as mgr
import yaml
from builder_core import (beauty_output, gen_response_and_process,
init_builder_chatbot_agent)
from config_utils import (DEFAULT_AGENT_DIR, Config, get_avatar_image,
get_ci_dir, get_user_dir,
is_valid_plugin_configuration, parse_configuration,
save_avatar_image, save_builder_configuration,
save_plugin_configuration)
from gradio_utils import format_cover_html, format_goto_publish_html
from i18n import I18n
from modelscope_agent.schemas import Message
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_gradio_components.components.Chatbot.llm_thinking_presets import \
qwen
from publish_util import (pop_user_info_from_config, prepare_agent_zip,
reload_agent_zip)
from user_core import init_user_chatbot_agent
def init_user(uuid_str, state):
try:
seed = state.get('session_seed', random.randint(0, 1000000000))
user_agent, user_memory = init_user_chatbot_agent(uuid_str)
user_agent.seed = seed
state['user_agent'] = user_agent
state['user_memory'] = user_memory
except Exception as e:
logger.query_error(
uuid=uuid_str,
error=str(e),
details={'error_traceback': traceback.format_exc()})
return state
def init_builder(uuid_str, state):
try:
builder_agent, builder_memory = init_builder_chatbot_agent(uuid_str)
state['builder_agent'] = builder_agent
state['builder_memory'] = builder_memory
except Exception as e:
logger.query_error(
uuid=uuid_str,
error=str(e),
details={'error_traceback': traceback.format_exc()})
return state
def check_uuid(uuid_str):
if not uuid_str or uuid_str == '':
if os.getenv('MODELSCOPE_ENVIRONMENT') == 'studio':
raise gr.Error('请登陆后使用! (Please login first)')
else:
uuid_str = 'local_user'
return uuid_str
with demo:
uuid_str = gr.Textbox(label='modelscope_uuid', visible=False)
draw_seed = random.randint(0, 1000000000)
state = gr.State({'session_seed': draw_seed})
i18n = I18n('zh-cn')
with gr.Row():
with gr.Column(scale=5):
header = gr.Markdown(i18n.get('header'))
with gr.Column(scale=1):
language = gr.Dropdown(
choices=[('中文', 'zh-cn'), ('English', 'en')],
show_label=False,
container=False,
value='zh-cn',
interactive=True)
with gr.Row():
with gr.Column():
with gr.Tabs() as tabs:
with gr.Tab(i18n.get_whole('create'), id=0) as create_tab:
with gr.Column():
# "Create" 标签页的 Chatbot 组件
start_text = '欢迎使用agent创建助手。我可以帮助您创建一个定制agent。'\
'您希望您的agent主要用于什么领域或任务？比如，您可以说，我想做一个RPG游戏agent'
create_chatbot = mgr.Chatbot(
show_label=False,
value=[[None, start_text]],
flushing=False,
show_copy_button=True,
llm_thinking_presets=[
qwen(
action_input_title='调用 <Action>',
action_output_title='完成调用')
])
create_chat_input = mgr.MultimodalInput(
label=i18n.get('message'),
placeholder=i18n.get('message_placeholder'),
interactive=False,
upload_button_props=dict(visible=False),
submit_button_props=dict(
label=i18n.get('sendOnLoading')))
configure_tab = gr.Tab(i18n.get_whole('configure'), id=1)
with configure_tab:
with gr.Column():
# "Configure" 标签页的配置输入字段
with gr.Row():
bot_avatar_comp = gr.Image(
label=i18n.get('form_avatar'),
sources=['upload'],
interactive=True,
type='filepath',
scale=1,
width=182,
height=182,
)
with gr.Column(scale=4):
name_input = gr.Textbox(
label=i18n.get('form_name'),
placeholder=i18n.get(
'form_name_placeholder'))
description_input = gr.Textbox(
label=i18n.get('form_description'),
placeholder=i18n.get(
'form_description_placeholder'))
instructions_input = gr.Textbox(
label=i18n.get('form_instructions'),
placeholder=i18n.get(
'form_instructions_placeholder'),
lines=3)
model_selector = gr.Dropdown(
label=i18n.get('form_model'))
agent_language_selector = gr.Dropdown(
label=i18n.get('form_agent_language'),
choices=['zh', 'en'],
value='zh')
suggestion_input = gr.Dataframe(
show_label=False,
value=[['']],
datatype=['str'],
headers=[i18n.get_whole('form_prompt_suggestion')],
type='array',
col_count=(1, 'fixed'),
interactive=True)
gr.Markdown(
'*注意：知识库上传的文本文档默认按照\\n\\n切分，pdf默认按照页切分。如果片段'
'对应的字符大于[配置文件](https://github.com/modelscope/modelscope-agent/'
'blob/master/apps/agentfabric/config/model_config.json)中指定模型的'
'knowledge限制，则在被召回时有可能会被截断。*')
knowledge_input = gr.File(
label=i18n.get('form_knowledge'),
file_count='multiple',
file_types=[
'text', '.json', '.csv', '.pdf', '.md'
])
capabilities_checkboxes = gr.CheckboxGroup(
label=i18n.get('form_capabilities'))
with gr.Accordion(
i18n.get('open_api_accordion'),
open=False) as open_api_accordion:
openapi_schema = gr.Textbox(
label='Schema',
placeholder=
'Enter your OpenAPI schema here, JSON or YAML format only'
)
with gr.Group():
openapi_auth_type = gr.Radio(
label='Authentication Type',
choices=['None', 'API Key'],
value='None')
openapi_auth_apikey = gr.Textbox(
label='API Key',
placeholder='Enter your API Key here')
openapi_auth_apikey_type = gr.Radio(
label='API Key type', choices=['Bearer'])
openapi_privacy_policy = gr.Textbox(
label='Privacy Policy',
placeholder='Enter privacy policy URL')
configure_button = gr.Button(
i18n.get('form_update_button'))
with gr.Accordion(
label=i18n.get('import_config'),
open=False) as update_accordion:
with gr.Column():
update_space = gr.Textbox(
label=i18n.get('space_addr'),
placeholder=i18n.get('input_space_addr'))
import_button = gr.Button(
i18n.get_whole('import_space'))
gr.Markdown(
f'#### {i18n.get_whole("import_hint")}')
with gr.Column():
# Preview
preview_header = gr.HTML(
f"""<div class="preview_header">{i18n.get('preview')}<div>""")
user_chat_bot_cover = gr.HTML(format_cover_html({}, None))
user_chatbot = mgr.Chatbot(
value=[[None, None]],
elem_id='user_chatbot',
elem_classes=['markdown-body'],
avatar_images=get_avatar_image('', uuid_str),
height=650,
latex_delimiters=[],
show_label=False,
visible=False,
show_copy_button=True,
llm_thinking_presets=[
qwen(
action_input_title='调用 <Action>',
action_output_title='完成调用')
])
preview_chat_input = mgr.MultimodalInput(
interactive=False,
label=i18n.get('message'),
placeholder=i18n.get('message_placeholder'),
upload_button_props=dict(
label=i18n.get('upload_btn'),
file_types=['file', 'image', 'audio', 'video', 'text'],
file_count='multiple'),
submit_button_props=dict(label=i18n.get('sendOnLoading')))
user_chat_bot_suggest = gr.Dataset(
label=i18n.get('prompt_suggestion'),
components=[preview_chat_input],
samples=[])
user_chat_bot_suggest.select(
lambda evt: evt[0],
inputs=[user_chat_bot_suggest],
outputs=[preview_chat_input])
with gr.Accordion(
label=i18n.get('publish'),
open=False) as publish_accordion:
publish_alert_md = gr.Markdown(f'{i18n.get("publish_alert")}')
with gr.Row():
with gr.Column():
publish_button = gr.Button(i18n.get_whole('build'))
build_hint_md = gr.Markdown(
f'#### 1.{i18n.get("build_hint")}')
with gr.Column():
publish_link = gr.HTML(
value=format_goto_publish_html(
i18n.get_whole('publish'), '', {}, True))
publish_hint_md = gr.Markdown(
f'#### 2.{i18n.get("publish_hint")}')
configure_updated_outputs = [
state,
# config form
bot_avatar_comp,
name_input,
description_input,
instructions_input,
model_selector,
agent_language_selector,
suggestion_input,
knowledge_input,
capabilities_checkboxes,
# bot
user_chat_bot_cover,
user_chat_bot_suggest,
preview_chat_input,
create_chat_input,
]
# 初始化表单
def init_ui_config(uuid_str, _state, builder_cfg, model_cfg, tool_cfg):
logger.query_info(
uuid=uuid_str,
message='builder_cfg',
details={'builder_cfg': str(builder_cfg)})
# available models
models = list(model_cfg.keys())
capabilities = [(tool_cfg[tool_key]['name'], tool_key)
for tool_key in tool_cfg.keys()
if tool_cfg[tool_key].get('is_active', False)]
_state['model_cfg'] = model_cfg
_state['tool_cfg'] = tool_cfg
_state['capabilities'] = capabilities
bot_avatar = get_avatar_image(builder_cfg.get('avatar', ''),
uuid_str)[1]
suggests = builder_cfg.get('prompt_recommend', [''])
return {
state:
_state,
bot_avatar_comp:
gr.Image(value=bot_avatar),
name_input:
builder_cfg.get('name', ''),
description_input:
builder_cfg.get('description'),
instructions_input:
builder_cfg.get('instruction'),
model_selector:
gr.Dropdown(
value=builder_cfg.get('model', models[0]), choices=models),
agent_language_selector:
builder_cfg.get('language') or 'zh',
suggestion_input:
[[str] for str in suggests] if len(suggests) > 0 else [['']],
knowledge_input:
builder_cfg.get('knowledge', [])
if len(builder_cfg['knowledge']) > 0 else None,
capabilities_checkboxes:
gr.CheckboxGroup(
value=[
tool for tool in builder_cfg.get('tools', {}).keys()
if builder_cfg.get('tools').get(tool).get('use', False)
],
choices=capabilities),
# bot
user_chat_bot_cover:
format_cover_html(builder_cfg, bot_avatar),
user_chat_bot_suggest:
gr.Dataset(
components=[preview_chat_input],
samples=[[item] for item in suggests]),
}
# tab 切换的事件处理
configure_tab.select(
on_congifure_tab_select,
inputs=[state, uuid_str],
outputs=configure_updated_outputs)
# 配置 "Create" 标签页的消息发送功能
create_chat_input.submit(
create_send_message,
inputs=[create_chatbot, create_chat_input, state, uuid_str],
outputs=[
create_chatbot, user_chat_bot_cover, user_chatbot,
user_chat_bot_suggest, create_chat_input
])
# 配置 "Configure" 标签页的提交按钮功能
configure_button.click(
process_configuration,
inputs=[
uuid_str, bot_avatar_comp, name_input, description_input,
instructions_input, model_selector, agent_language_selector,
suggestion_input, knowledge_input, capabilities_checkboxes,
openapi_schema, openapi_auth_type, openapi_auth_apikey,
openapi_auth_apikey_type, openapi_privacy_policy, state
],
outputs=[
user_chat_bot_cover, user_chatbot, user_chat_bot_suggest,
suggestion_input
])
# 配置 "Preview" 的消息发送功能
preview_chat_input.submit(
preview_send_message,
inputs=[user_chatbot, preview_chat_input, state, uuid_str],
outputs=[user_chatbot, user_chat_bot_cover, preview_chat_input])
# configuration for publish
publish_button.click(
publish_agent,
inputs=[name_input, uuid_str, state],
outputs=[publish_link],
)
import_button.click(
import_space,
inputs=[update_space, uuid_str, state],
outputs=configure_updated_outputs,
)
language.select(
change_lang,
inputs=[language],
outputs=configure_updated_outputs + [
configure_button, create_chat_input, open_api_accordion,
preview_header, preview_chat_input, publish_accordion, header,
publish_alert_md, build_hint_md, publish_hint_md
])
demo.load(
init_all, inputs=[uuid_str, state], outputs=configure_updated_outputs)
def parse_configuration(uuid_str=''):
"""parse configuration
Args:
Returns:
dict: parsed configuration
"""
model_cfg_file = os.getenv('MODEL_CONFIG_FILE', DEFAULT_MODEL_CONFIG_FILE)
builder_cfg_file = get_user_cfg_file(uuid_str)
# use default if not exists
if not os.path.exists(builder_cfg_file):
# create parents directory
os.makedirs(os.path.dirname(builder_cfg_file), exist_ok=True)
# copy the template to the address
builder_cfg_file_temp = './config/builder_config.json'
if builder_cfg_file_temp != builder_cfg_file:
shutil.copy(builder_cfg_file_temp, builder_cfg_file)
tool_cfg_file = os.getenv('TOOL_CONFIG_FILE', DEFAULT_TOOL_CONFIG_FILE)
builder_cfg = Config.from_file(builder_cfg_file)
model_cfg = Config.from_file(model_cfg_file)
tool_cfg = Config.from_file(tool_cfg_file)
tools_info = builder_cfg.tools
available_tool_list = []
for key, value in tools_info.items():
if value['use']:
available_tool_list.append(key)
tool_cfg[key]['use'] = value['use']
openapi_plugin_file = get_user_openapi_plugin_cfg_file(uuid_str)
plugin_cfg = {}
available_plugin_list = []
if os.path.exists(openapi_plugin_file):
openapi_plugin_cfg = Config.from_file(openapi_plugin_file)
try:
config_dict = openapi_schema_convert(
schema=openapi_plugin_cfg.schema,
auth=openapi_plugin_cfg.auth.to_dict())
plugin_cfg = Config(config_dict)
for name, config in config_dict.items():
available_plugin_list.append(name)
except Exception as e:
logger.query_error(
uuid=uuid_str,
error=str(e),
content={
'error_traceback':
traceback.format_exc(),
'error_details':
'The format of the plugin config file is incorrect.'
})
return builder_cfg, model_cfg, tool_cfg, available_tool_list, plugin_cfg, available_plugin_list
def init_all(uuid_str, _state):
uuid_str = check_uuid(uuid_str)
builder_cfg, model_cfg, tool_cfg, available_tool_list, _, _ = parse_configuration(
uuid_str)
ret = init_ui_config(uuid_str, _state, builder_cfg, model_cfg,
tool_cfg)
yield ret
init_user(uuid_str, _state)
init_builder(uuid_str, _state)
yield {
state:
_state,
preview_chat_input:
mgr.MultimodalInput(
submit_button_props=dict(label=i18n.get('send')),
interactive=True),
create_chat_input:
mgr.MultimodalInput(
submit_button_props=dict(label=i18n.get('send')),
interactive=True),
} | null |
179,765 |
The provided code snippet includes necessary dependencies for implementing the `parse_action_response` function. Write a Python function `def parse_action_response(response: str) -> Tuple[str, Dict]` to solve the following problem:
parse response of llm to get tool name and parameters Args: response (str): llm response, it should conform to some predefined format Returns: tuple[str, dict]: tuple of tool name and parameters
Here is the function:
def parse_action_response(response: str) -> Tuple[str, Dict]:
"""parse response of llm to get tool name and parameters
Args:
response (str): llm response, it should conform to some predefined format
Returns:
tuple[str, dict]: tuple of tool name and parameters
"""
if 'Action' not in response or 'Action Input:' not in response:
return None, None
action, action_para = '', ''
# use regular expression to get result from MRKL format
re_pattern1 = re.compile(pattern=r'Action:([\s\S]+)Action Input:([\s\S]+)')
res = re_pattern1.search(response)
action = res.group(1).strip()
action_para = res.group(2)
parameters = json.loads(action_para.replace('\n', ''))
return action, parameters | parse response of llm to get tool name and parameters Args: response (str): llm response, it should conform to some predefined format Returns: tuple[str, dict]: tuple of tool name and parameters |
179,766 |
def convert_url(text, new_filename):
# Define the pattern to search for
# This pattern captures the text inside the square brackets, the path, and the filename
pattern = r'!\[([^\]]+)\]\(([^)]+)\)'
# Define the replacement pattern
# \1 is a backreference to the text captured by the first group ([^\]]+)
replacement = rf''
# Replace the pattern in the text with the replacement
return re.sub(pattern, replacement, text) | null |
179,767 |
def postprocess_messages(
message_pairs: list[list[str | tuple[str] | tuple[str, str] | None]
| tuple]
) -> list[list[str | dict | None]]:
if message_pairs is None:
return []
processed_messages = []
for message_pair in message_pairs:
assert isinstance(
message_pair, (tuple, list)
), f'Expected a list of lists or list of tuples. Received: {message_pair}'
assert (
len(message_pair) == 2
), f'Expected a list of lists of length 2 or list of tuples of length 2. Received: {message_pair}'
user_message, bot_message = message_pair
# TODO: 是否需要进行任何其他的后处理
processed_messages.append([
user_message,
bot_message,
])
return processed_messages | null |
179,768 | import os
from typing import Dict, Iterator, List, Optional
from zhipuai import ZhipuAI
from .base import BaseChatModel, register_llm
def stream_output(response, **kwargs):
func_call = {
'name': None,
'arguments': '',
}
for chunk in response:
delta = chunk.choices[0].delta
if delta.tool_calls:
# TODO : multi tool_calls
tool_call = delta.tool_calls[0]
print(f'tool_call: {tool_call}')
func_call['name'] = tool_call.function.name
func_call['arguments'] += tool_call.function.arguments
if chunk.choices[0].finish_reason == 'tool_calls':
yield {'function_call': func_call}
else:
yield delta.content | null |
179,769 | from abc import ABC, abstractmethod
from typing import Dict, Iterator, List, Optional, Union
from modelscope_agent.utils.retry import retry
from modelscope_agent.utils.tokenization_utils import count_tokens
from modelscope_agent.utils.utils import print_traceback
LLM_REGISTRY = {}
def register_llm(name):
def decorator(cls):
LLM_REGISTRY[name] = cls
return cls
return decorator | null |
179,770 | import os
from http import HTTPStatus
from typing import Dict, Iterator, List, Optional, Union
import dashscope
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_agent.utils.tokenization_utils import count_tokens
from .base import BaseChatModel, register_llm
def stream_output(response, **kwargs):
last_len = 0
delay_len = 5
in_delay = False
text = ''
for trunk in response:
if trunk.status_code == HTTPStatus.OK:
# logger at the first for the request_id, and the last time for whole output
if not text or trunk.output.choices[0].finish_reason != 'null':
logger.info(
f'call dashscope generation api success, '
f'request_id: { trunk.request_id}, output: { trunk.output}'
)
text = trunk.output.choices[0].message.content
if (len(text) - last_len) <= delay_len:
in_delay = True
continue
else:
in_delay = False
real_text = text[:-delay_len]
now_rsp = real_text[last_len:]
yield now_rsp
last_len = len(real_text)
else:
logger.query_error(
uuid=kwargs.get('uuid_str', ''),
details={
'dashscope.request_id': trunk.request_id,
'dashscope.status_code': trunk.status_code,
'dashscope.code': trunk.code,
'dashscope.message': trunk.message
},
message='call dashscope generation api error')
err = '\nError code: %s. Error message: %s with request id %s' % (
trunk.code, trunk.message, trunk.request_id)
if trunk.code == 'DataInspectionFailed':
err += '\n错误码: 数据检查失败。错误信息: 输入数据可能包含不适当的内容。由于该不适当内容会一直存在历史对话中，后续的对话大概率仍会触发此错误。建议刷新重置页面。'
text = ''
yield f'{err}'
# with open('debug.json', 'w', encoding='utf-8') as writer:
# writer.write(json.dumps(trunk, ensure_ascii=False))
if text and (in_delay or (last_len != len(text))):
yield text[last_len:] | null |
179,771 | import hashlib
import os
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_agent.utils.utils import (print_traceback, read_text_from_file,
save_text_to_file)
from .base import BaseStorage
def hash_sha256(key):
hash_object = hashlib.sha256(key.encode())
key = hash_object.hexdigest()
return key | null |
179,772 | import copy
import datetime
import os
import re
from typing import Dict, List
from urllib.parse import unquote, urlparse
import json
import json5
from modelscope_agent.schemas import Document
from modelscope_agent.storage import BaseStorage, DocumentStorage
from modelscope_agent.tools.base import BaseTool, register_tool
from modelscope_agent.tools.similarity_search import (RefMaterialInput,
RefMaterialInputItem)
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_agent.utils.parse_doc import (count_tokens, parse_doc,
parse_html_bs)
from modelscope_agent.utils.utils import print_traceback, save_text_to_file
def sanitize_chrome_file_path(file_path: str) -> str:
class Document(BaseModel):
def parse_doc(path):
def parse_html_bs(path):
def print_traceback():
def save_text_to_file(path, text):
def process_file(url: str,
content: str = None,
source: str = None,
db: BaseStorage = None):
logger.info('Starting cache pages...')
url = url
if url.split('.')[-1].lower() in ['pdf', 'docx', 'pptx', 'txt']:
date1 = datetime.datetime.now()
# generate one processing record
db.put(url, '')
if url.startswith('https://') or url.startswith('http://'):
pdf_path = url
else:
parsed_url = urlparse(url)
pdf_path = unquote(parsed_url.path)
pdf_path = sanitize_chrome_file_path(pdf_path)
try:
pdf_content = parse_doc(pdf_path)
date2 = datetime.datetime.now()
logger.info('Parsing pdf time: ' + str(date2 - date1))
content = pdf_content
source = 'pdf'
title = pdf_path.split('/')[-1].split('\\')[-1].split('.')[0]
except Exception:
print_traceback()
# del the processing record
db.delete(url)
return 'failed'
elif content and source == 'html':
# generate one processing record
db.put(url, '')
try:
tmp_html_file = os.path.join(db.root, 'tmp.html')
save_text_to_file(tmp_html_file, content)
content = parse_html_bs(tmp_html_file)
title = content[0]['metadata']['title']
except Exception:
print_traceback()
# del the processing record
db.delete(url)
return 'failed'
else:
logger.error(
'Only Support the Following File Types: [\'.html\', \'.pdf\', \'.docx\', \'.pptx\']'
)
raise NotImplementedError
# save real data
now_time = str(datetime.date.today())
new_record = Document(
url=url,
time=now_time,
source=source,
raw=content,
title=title,
topic='',
checked=True,
session=[]).model_dump()
new_record_str = json.dumps(new_record, ensure_ascii=False)
db.add(url, new_record_str)
meta_info = db.search('meta_info')
if meta_info == 'Not Exist':
meta_info = {}
else:
meta_info = json5.loads(meta_info)
if isinstance(meta_info, list):
logger.info('update meta_info to new format')
new_meta_info = {}
for x in meta_info:
new_meta_info[x['url']] = x
meta_info = new_meta_info
meta_info[url] = {
'url': url,
'time': now_time,
'title': title,
'checked': True,
}
db.add('meta_info', json.dumps(meta_info, ensure_ascii=False))
return new_record_str | null |
179,773 | import copy
import datetime
import os
import re
from typing import Dict, List
from urllib.parse import unquote, urlparse
import json
import json5
from modelscope_agent.schemas import Document
from modelscope_agent.storage import BaseStorage, DocumentStorage
from modelscope_agent.tools.base import BaseTool, register_tool
from modelscope_agent.tools.similarity_search import (RefMaterialInput,
RefMaterialInputItem)
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_agent.utils.parse_doc import (count_tokens, parse_doc,
parse_html_bs)
from modelscope_agent.utils.utils import print_traceback, save_text_to_file
def token_counter_backup(records):
new_records = []
for record in records:
if not record['raw']:
continue
if 'token' not in record['raw'][0]['page_content']:
tmp = []
for page in record['raw']:
new_page = copy.deepcopy(page)
new_page['token'] = count_tokens(page['page_content'])
tmp.append(new_page)
record['raw'] = tmp
new_records.append(record)
return new_records | null |
179,774 | import copy
import datetime
import os
import re
from typing import Dict, List
from urllib.parse import unquote, urlparse
import json
import json5
from modelscope_agent.schemas import Document
from modelscope_agent.storage import BaseStorage, DocumentStorage
from modelscope_agent.tools.base import BaseTool, register_tool
from modelscope_agent.tools.similarity_search import (RefMaterialInput,
RefMaterialInputItem)
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_agent.utils.parse_doc import (count_tokens, parse_doc,
parse_html_bs)
from modelscope_agent.utils.utils import print_traceback, save_text_to_file
The provided code snippet includes necessary dependencies for implementing the `read_data_by_condition` function. Write a Python function `def read_data_by_condition(db: BaseStorage = None, **kwargs)` to solve the following problem:
filter records from meta-data
Here is the function:
def read_data_by_condition(db: BaseStorage = None, **kwargs):
"""
filter records from meta-data
"""
meta_info = db.get('meta_info')
if meta_info == 'Not Exist':
records = []
else:
records = json5.loads(meta_info)
records = records.values()
if 'time_limit' in kwargs:
filter_records = []
for x in records:
if kwargs['time_limit'][0] <= x['time'] <= kwargs['time_limit'][1]:
filter_records.append(x)
records = filter_records
if 'checked' in kwargs:
filter_records = []
for x in records:
if x['checked']:
filter_records.append(x)
records = filter_records
return records | filter records from meta-data |
179,775 | import copy
import datetime
import os
import re
from typing import Dict, List
from urllib.parse import unquote, urlparse
import json
import json5
from modelscope_agent.schemas import Document
from modelscope_agent.storage import BaseStorage, DocumentStorage
from modelscope_agent.tools.base import BaseTool, register_tool
from modelscope_agent.tools.similarity_search import (RefMaterialInput,
RefMaterialInputItem)
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_agent.utils.parse_doc import (count_tokens, parse_doc,
parse_html_bs)
from modelscope_agent.utils.utils import print_traceback, save_text_to_file
class RefMaterialInputItem(BaseModel):
content: str
token: int
def to_dict(self) -> dict:
return {'content': self.content, 'token': self.token}
class RefMaterialInput(BaseModel):
"""
The knowledge data format input to the retrieval
"""
url: str
text: List[RefMaterialInputItem]
def to_dict(self) -> dict:
return {'url': self.url, 'text': [x.to_dict() for x in self.text]}
def format_records(records: List[Dict]):
formatted_records = []
for record in records:
formatted_records.append(
RefMaterialInput(
url=record['url'],
text=[
RefMaterialInputItem(
content=x['page_content'], token=x['token'])
for x in record['raw']
]).to_dict())
return formatted_records | null |
179,776 | import os
from enum import Enum
from pydantic import BaseModel, model_validator
class AuthenticationKey(Enum):
def to_dict(cls):
class BingWebSearcher(WebSearcher):
def __init__(
self,
timeout=3000,
mkt='en-US',
endpoint='https://api.bing.microsoft.com/v7.0/search',
**kwargs,
):
def __call__(self, query, **kwargs):
class KuakeWebSearcher(WebSearcher):
def __call__(self, query, **kwargs):
def get_websearcher_cls():
def get_env(authentication_key: str):
env = os.environ
return env.get(authentication_key, None)
cls_dict = {}
if get_env(AuthenticationKey.bing.value):
from .searcher.bing import BingWebSearcher
cls_dict[AuthenticationKey.bing.name] = BingWebSearcher
if get_env(AuthenticationKey.kuake.value):
from .searcher.kuake import KuakeWebSearcher
cls_dict[AuthenticationKey.kuake.name] = KuakeWebSearcher
return cls_dict | null |
179,777 | import os
import re
from typing import List, Optional
import json
import requests
from jsonschema import RefResolver
from modelscope_agent.tools.base import BaseTool, register_tool
from pydantic import BaseModel, ValidationError
from requests.exceptions import RequestException, Timeout
def parse_responses_parameters(param_name, param_info, parameters_list):
param_type = param_info['type']
param_description = param_info.get('description',
f'调用api返回的{param_name}') # 按需更改描述
try:
if param_type == 'object':
properties = param_info.get('properties')
if properties:
# If the argument type is an object and has a non-empty "properties"
# field, its internal properties are parsed recursively
for inner_param_name, inner_param_info in properties.items():
param_type = inner_param_info['type']
param_description = inner_param_info.get(
'description',
f'调用api返回的{param_name}.{inner_param_name}')
parameters_list.append({
'name': f'{param_name}.{inner_param_name}',
'description': param_description,
'type': param_type,
})
else:
# Non-nested parameters are added directly to the parameter list
parameters_list.append({
'name': param_name,
'description': param_description,
'type': param_type,
})
except Exception as e:
raise ValueError(f'{e}:schema结构出错') | null |
179,778 | import os
import subprocess
from http import HTTPStatus
from typing import Any, Dict, List, Optional
from modelscope_agent.tools.base import BaseTool, register_tool
from pydantic import ValidationError
def _preprocess(input_file, output_file):
ret = subprocess.call([
'ffmpeg', '-y', '-i', input_file, '-f', 's16le', '-acodec',
'pcm_s16le', '-ac', '1', '-ar', '16000', '-loglevel', 'quiet',
output_file
])
if ret != 0:
raise ValueError(f'Failed to preprocess audio file {input_file}') | null |
179,779 | from abc import ABC, abstractmethod
from typing import Dict, List, Optional, Union
import json
import json5
from modelscope_agent.tools.base import BaseTool
from modelscope_agent.utils.utils import has_chinese_chars
TOOL_REGISTRY = {}
def register_tool(name):
def decorator(cls):
TOOL_REGISTRY[name] = cls
return cls
return decorator | null |
179,780 | import math import os
import re
import signal
import json
import matplotlib
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import seaborn as sns
from matplotlib.font_manager import FontProperties
from sympy import Eq, solve, symbols
def input(*args, **kwargs): # noqa
raise NotImplementedError('Python input() function is disabled.') | null |
179,781 | import math import os
import re
import signal
import json
import matplotlib
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import seaborn as sns
from matplotlib.font_manager import FontProperties
from sympy import Eq, solve, symbols
def _m6_timout_handler(_signum=None, _frame=None):
raise TimeoutError('M6_CODE_INTERPRETER_TIMEOUT') | null |
179,782 | from abc import ABC, abstractmethod
from typing import Dict, List, Optional, Union
import json
import json5
from modelscope_agent.utils.utils import has_chinese_chars
TOOL_REGISTRY = {}
def register_tool(name):
def decorator(cls):
TOOL_REGISTRY[name] = cls
return cls
return decorator | null |
179,783 | import hashlib
from typing import Dict, Optional
from modelscope_agent.storage import DocumentStorage
from modelscope_agent.tools.base import BaseTool, register_tool
def hash_sha256(key):
hash_object = hashlib.sha256(key.encode())
key = hash_object.hexdigest()
return key | null |
179,784 | import os
import re
import tempfile
import uuid
from typing import Dict
import numpy as np
import requests
from PIL import Image
from requests.exceptions import RequestException
class OutputWrapper:
"""
Wrapper for output of tool execution when output is image, video, audio, etc.
In this wrapper, __repr__() is implemented to return the str representation of the output for llm.
Each wrapper have below attributes:
path: the path where the output is stored
raw_data: the raw data, e.g. image, video, audio, etc. In remote mode, it should be None
"""
def __init__(self) -> None:
self._repr = None
self._path = None
self._raw_data = None
self.root_path = os.environ.get('OUTPUT_FILE_DIRECTORY', None)
if self.root_path and not os.path.exists(self.root_path):
try:
os.makedirs(self.root_path)
except Exception:
self.root_path = None
def get_remote_file(self, remote_path, suffix):
try:
response = requests.get(remote_path)
obj = response.content
directory = tempfile.mkdtemp(dir=self.root_path)
path = os.path.join(directory, str(uuid.uuid4()) + f'.{suffix}')
with open(path, 'wb') as f:
f.write(obj)
return path
except RequestException:
return remote_path
def __repr__(self) -> str:
return self._repr
def path(self):
return self._path
def raw_data(self):
return self._raw_data
def get_raw_output(exec_result: Dict):
# get rwa data of exec_result
res = {}
for k, v in exec_result.items():
if isinstance(v, OutputWrapper):
# In remote mode, raw data maybe None
res[k] = v.raw_data or str(v)
else:
res[k] = v
return res | null |
179,785 | import datetime
import os
import re
import socket
import sys
import traceback
from typing import Literal, Optional, Union
from urllib.parse import unquote_plus, urlparse
import jieba
import json5
from dashscope.common.error import InvalidInput, UploadFileException
from dashscope.utils.oss_utils import OssUtils
from jieba import analyse
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_agent.utils.tokenization_utils import count_tokens
def get_local_ip():
s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
try:
# doesn't even have to be reachable
s.connect(('10.255.255.255', 1))
IP = s.getsockname()[0]
except Exception:
IP = '127.0.0.1'
finally:
s.close()
return IP | null |
179,786 | import datetime
import os
import re
import socket
import sys
import traceback
from typing import Literal, Optional, Union
from urllib.parse import unquote_plus, urlparse
import jieba
import json5
from dashscope.common.error import InvalidInput, UploadFileException
from dashscope.utils.oss_utils import OssUtils
from jieba import analyse
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_agent.utils.tokenization_utils import count_tokens
def has_chinese_chars(data) -> bool:
text = f'{data}'
return len(re.findall(r'[\u4e00-\u9fff]+', text)) > 0 | null |
179,787 | import datetime
import os
import re
import socket
import sys
import traceback
from typing import Literal, Optional, Union
from urllib.parse import unquote_plus, urlparse
import jieba
import json5
from dashscope.common.error import InvalidInput, UploadFileException
from dashscope.utils.oss_utils import OssUtils
from jieba import analyse
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_agent.utils.tokenization_utils import count_tokens
def get_current_date_str(
lang: Literal['en', 'zh'] = 'en',
hours_from_utc: Optional[int] = None,
) -> str:
if hours_from_utc is None:
cur_time = datetime.datetime.now()
else:
cur_time = datetime.datetime.utcnow() + datetime.timedelta(
hours=hours_from_utc)
if lang == 'en':
date_str = 'Current date: ' + cur_time.strftime('%A, %B %d, %Y')
elif lang == 'zh':
cur_time = cur_time.timetuple()
date_str = f'当前时间：{cur_time.tm_year}年{cur_time.tm_mon}月{cur_time.tm_mday}日，星期'
date_str += ['一', '二', '三', '四', '五', '六', '日'][cur_time.tm_wday]
date_str += '。'
else:
raise NotImplementedError
return date_str | null |
179,788 | import datetime
import os
import re
import socket
import sys
import traceback
from typing import Literal, Optional, Union
from urllib.parse import unquote_plus, urlparse
import jieba
import json5
from dashscope.common.error import InvalidInput, UploadFileException
from dashscope.utils.oss_utils import OssUtils
from jieba import analyse
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_agent.utils.tokenization_utils import count_tokens
def read_text_from_file(path):
with open(path, 'r', encoding='utf-8') as file:
file_content = file.read()
return file_content | null |
179,789 | import datetime
import os
import re
import socket
import sys
import traceback
from typing import Literal, Optional, Union
from urllib.parse import unquote_plus, urlparse
import jieba
import json5
from dashscope.common.error import InvalidInput, UploadFileException
from dashscope.utils.oss_utils import OssUtils
from jieba import analyse
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_agent.utils.tokenization_utils import count_tokens
ignore_words = [
'', ' ', '\t', '\n', '\\', 'is', 'are', 'am', 'what', 'how', '的', '吗', '是',
'了', '啊', '呢', '怎么', '如何', '什么', '？', '?', '！', '!', '“', '”', '‘', '’',
"'", "'", '"', '"', ':', '：', '讲了', '描述', '讲', '说说', '讲讲', '介绍', '总结下',
'总结一下', '文档', '文章', '文稿', '稿子', '论文', 'PDF', 'pdf', '这个', '这篇', '这', '我',
'帮我', '那个', '下', '翻译'
]
def get_split_word(text):
def get_keyword_by_llm(text):
try:
res = json5.loads(text)
except Exception:
return get_split_word(text)
# json format
_wordlist = []
try:
if 'keywords_zh' in res and isinstance(res['keywords_zh'], list):
_wordlist.extend([kw.lower() for kw in res['keywords_zh']])
if 'keywords_en' in res and isinstance(res['keywords_en'], list):
_wordlist.extend([kw.lower() for kw in res['keywords_en']])
wordlist = []
for x in _wordlist:
if x in ignore_words:
continue
wordlist.append(x)
wordlist.extend(get_split_word(res['text']))
return wordlist
except Exception:
return get_split_word(text) | null |
179,790 | import datetime
import os
import re
import socket
import sys
import traceback
from typing import Literal, Optional, Union
from urllib.parse import unquote_plus, urlparse
import jieba
import json5
from dashscope.common.error import InvalidInput, UploadFileException
from dashscope.utils.oss_utils import OssUtils
from jieba import analyse
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_agent.utils.tokenization_utils import count_tokens
ignore_words = [
'', ' ', '\t', '\n', '\\', 'is', 'are', 'am', 'what', 'how', '的', '吗', '是',
'了', '啊', '呢', '怎么', '如何', '什么', '？', '?', '！', '!', '“', '”', '‘', '’',
"'", "'", '"', '"', ':', '：', '讲了', '描述', '讲', '说说', '讲讲', '介绍', '总结下',
'总结一下', '文档', '文章', '文稿', '稿子', '论文', 'PDF', 'pdf', '这个', '这篇', '这', '我',
'帮我', '那个', '下', '翻译'
]
def get_key_word(text):
text = text.lower()
_wordlist = analyse.extract_tags(text)
wordlist = []
for x in _wordlist:
if x in ignore_words:
continue
wordlist.append(x)
print('wordlist: ', wordlist)
return wordlist | null |
179,791 | import datetime
import os
import re
import socket
import sys
import traceback
from typing import Literal, Optional, Union
from urllib.parse import unquote_plus, urlparse
import jieba
import json5
from dashscope.common.error import InvalidInput, UploadFileException
from dashscope.utils.oss_utils import OssUtils
from jieba import analyse
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_agent.utils.tokenization_utils import count_tokens
def get_last_one_line_context(text):
lines = text.split('\n')
n = len(lines)
res = ''
for i in range(n - 1, -1, -1):
if lines[i].strip():
res = lines[i]
break
return res | null |
179,792 | import datetime
import os
import re
import socket
import sys
import traceback
from typing import Literal, Optional, Union
from urllib.parse import unquote_plus, urlparse
import jieba
import json5
from dashscope.common.error import InvalidInput, UploadFileException
from dashscope.utils.oss_utils import OssUtils
from jieba import analyse
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_agent.utils.tokenization_utils import count_tokens
def print_traceback():
def extract_urls(text):
def extract_obs(text):
def extract_code(text):
def parse_latest_plugin_call(text):
def format_answer(text):
action, action_input, output = parse_latest_plugin_call(text)
if 'code_interpreter' in text:
rsp = ''
code = extract_code(action_input)
rsp += ('\n```py\n' + code + '\n```\n')
obs = extract_obs(text)
if '![fig' in obs:
rsp += obs
return rsp
elif 'image_gen' in text:
# get url of FA
# img_urls = URLExtract().find_urls(text.split("Final Answer:")[-1].strip())
obs = text.split('Observation:')[-1].split('\nThought:')[0].strip()
img_urls = []
if obs:
logger.info(repr(obs))
try:
obs = json5.loads(obs)
img_urls.append(obs['image_url'])
except Exception:
print_traceback()
img_urls = []
if not img_urls:
img_urls = extract_urls(text.split('Final Answer:')[-1].strip())
logger.info(img_urls)
rsp = ''
for x in img_urls:
rsp += '\n + ')'
return rsp
else:
return text.split('Final Answer:')[-1].strip() | null |
179,793 | import datetime
import os
import re
import socket
import sys
import traceback
from typing import Literal, Optional, Union
from urllib.parse import unquote_plus, urlparse
import jieba
import json5
from dashscope.common.error import InvalidInput, UploadFileException
from dashscope.utils.oss_utils import OssUtils
from jieba import analyse
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_agent.utils.tokenization_utils import count_tokens
FILE_PATH_SCHEMA = 'file://'
The provided code snippet includes necessary dependencies for implementing the `get_upload_url` function. Write a Python function `def get_upload_url(model: str, file_to_upload: str, api_key: str)` to solve the following problem:
This function is used to convert local file to get its oss url. Args: model(str): Theoretically, you can set this parameter freely. It will only affect the information of the oss url and will not affect the function function. file_to_upload(str): the local file path which you need to convert to oss url.And it should start with 'file://'. api_key(str): dashscope_api_key which you have set in enviroment. Returns: An oss type url. Raises: InvalidInput: the file path you upload is not exists.
Here is the function:
def get_upload_url(model: str, file_to_upload: str, api_key: str):
"""This function is used to convert local file to get its oss url.
Args:
model(str): Theoretically, you can set this parameter freely. It will only affect
the information of the oss url and will not affect the function function.
file_to_upload(str): the local file path which you need to convert to oss url.And it should
start with 'file://'.
api_key(str): dashscope_api_key which you have set in enviroment.
Returns:
An oss type url.
Raises:
InvalidInput: the file path you upload is not exists.
"""
if file_to_upload.startswith(FILE_PATH_SCHEMA):
parse_result = urlparse(file_to_upload)
if parse_result.netloc:
file_path = parse_result.netloc + unquote_plus(parse_result.path)
else:
file_path = unquote_plus(parse_result.path)
if os.path.exists(file_path):
file_url = OssUtils.upload(
model=model, file_path=file_path, api_key=api_key)
if file_url is None:
raise UploadFileException('Uploading file: %s failed'
% file_to_upload)
return file_url
else:
raise InvalidInput('The file: %s is not exists!' % file_path)
return None | This function is used to convert local file to get its oss url. Args: model(str): Theoretically, you can set this parameter freely. It will only affect the information of the oss url and will not affect the function function. file_to_upload(str): the local file path which you need to convert to oss url.And it should start with 'file://'. api_key(str): dashscope_api_key which you have set in enviroment. Returns: An oss type url. Raises: InvalidInput: the file path you upload is not exists. |
179,794 | import datetime
import os
import re
import socket
import sys
import traceback
from typing import Literal, Optional, Union
from urllib.parse import unquote_plus, urlparse
import jieba
import json5
from dashscope.common.error import InvalidInput, UploadFileException
from dashscope.utils.oss_utils import OssUtils
from jieba import analyse
from modelscope_agent.utils.logger import agent_logger as logger
from modelscope_agent.utils.tokenization_utils import count_tokens
def count_tokens(text):
tokens = tokenizer.tokenize(text)
return len(tokens)
The provided code snippet includes necessary dependencies for implementing the `check_and_limit_input_length` function. Write a Python function `def check_and_limit_input_length(check_body: Union[list, str], max_length: int) -> str` to solve the following problem:
Check the input length and limit the length, Args: check_body: the input to be checked, should be a list of message or single prompt string max_length: the maximum length of the check_body Returns: the output with the length limited
Here is the function:
def check_and_limit_input_length(check_body: Union[list, str],
max_length: int) -> str:
"""
Check the input length and limit the length,
Args:
check_body: the input to be checked, should be a list of message or single prompt string
max_length: the maximum length of the check_body
Returns:
the output with the length limited
"""
if isinstance(check_body, str):
if len(check_body) <= max_length:
return check_body
check_result = check_body[:max_length]
return check_result
else:
# limit the length by limit the history of message from far to near
used_length = 0
output_messages = []
start_index = 0
if check_body[0]['role'] == 'system':
output_messages.append(check_body[0])
used_length += count_tokens(check_body[0]['content'])
start_index = 1
for message in reversed(check_body[start_index:]):
used_length += count_tokens(message['content'])
if used_length <= max_length:
# add to the output messages first index
output_messages.insert(start_index, message)
else:
break
return output_messages | Check the input length and limit the length, Args: check_body: the input to be checked, should be a list of message or single prompt string max_length: the maximum length of the check_body Returns: the output with the length limited |
179,795 | import time
from functools import wraps
from modelscope_agent.utils.logger import agent_logger as logger
The provided code snippet includes necessary dependencies for implementing the `retry` function. Write a Python function `def retry(max_retries=3, delay_seconds=1, return_str=False)` to solve the following problem:
Retry decorator with exponential backoff. Args: max_retries: max retry times delay_seconds: delay seconds between retries return_str: want to return in str format, set it to True Returns:func
Here is the function:
def retry(max_retries=3, delay_seconds=1, return_str=False):
"""
Retry decorator with exponential backoff.
Args:
max_retries: max retry times
delay_seconds: delay seconds between retries
return_str: want to return in str format, set it to True
Returns:func
"""
def decorator(func):
@wraps(func)
def wrapper(*args, **kwargs):
attempts = 0
while attempts < max_retries:
try:
return func(*args, **kwargs)
except AssertionError as e:
raise AssertionError(e)
except Exception as e:
logger.warning(
f'Attempt to run {func.__name__} {attempts + 1} failed: {e}'
)
attempts += 1
time.sleep(delay_seconds)
if return_str:
return f'Max retries reached. Attempt to run {func.__name__} failed after {max_retries} times'
else:
raise Exception('Max retries reached. Failed to get result')
return wrapper
return decorator | Retry decorator with exponential backoff. Args: max_retries: max retry times delay_seconds: delay seconds between retries return_str: want to return in str format, set it to True Returns:func |
179,796 | import base64
import logging
import os
import unicodedata
from dataclasses import dataclass, field
from pathlib import Path
from typing import Collection, Dict, List, Set, Tuple, Union
import tiktoken
def _load_tiktoken_bpe(tiktoken_bpe_file: str) -> Dict[bytes, int]:
with open(tiktoken_bpe_file, 'rb') as f:
contents = f.read()
return {
base64.b64decode(token): int(rank)
for token, rank in (line.split() for line in contents.splitlines()
if line)
} | null |
179,797 | import logging
import os
from datetime import datetime
from logging.handlers import RotatingFileHandler
from typing import Dict
import json
class JsonFormatter(logging.Formatter):
"""
Custom formatter to output logs in JSON format.
"""
def format(self, record):
log_record = {
'timestamp': datetime.now().strftime('%Y-%m-%d %H:%M:%S'),
'level': record.levelname,
'message': record.getMessage(),
# Extract additional fields if they are in the 'extra' dict
'uuid': getattr(record, 'uuid', None),
# 'request_id': getattr(record, 'request_id', None),
'details': getattr(record, 'details', None),
'error': getattr(record, 'error', None),
'step': getattr(record, 'step', None)
}
# Clean up any extra fields that are None (not provided)
log_record = {k: v for k, v in log_record.items() if v is not None}
if record.exc_info:
log_record['exc_info'] = self.formatException(record.exc_info)
return json.dumps(log_record, ensure_ascii=False)
class TextFormatter(logging.Formatter):
"""
Custom formatter to output logs in text format.
"""
def format(self, record):
timestamp = datetime.now().strftime('%Y-%m-%d %H:%M:%S.%f')[:-3]
level = record.levelname
message = record.getMessage()
# Collect additional fields if they are in the 'extra' dict
uuid = getattr(record, 'uuid', '-')
# request_id = getattr(record, 'request_id', '-')
details = getattr(record, 'details', '-')
step = getattr(record, 'step', '-')
error = getattr(record, 'error', '-')
# Format the log record as text
log_message = f'{timestamp} - {LOG_NAME} - {level} - '
log_message = log_message + f' | message: {message}'
if uuid != '-':
log_message += f' | uuid: {uuid}'
# if request_id != '-':
# log_message += f' | request_id: {request_id}'
if details != '-':
log_message += f' | details: {details}'
if step != '-':
log_message += f' | step: {step}'
if error != '-':
log_message += f' | error: {error}'
if record.exc_info:
log_message += f' | Exception: {self.formatException(record.exc_info)}'
return log_message
def get_formatter(log_format_env):
if log_format_env == 'json':
formatter = JsonFormatter()
else:
formatter = TextFormatter()
return formatter | null |
179,798 | import os
from dataclasses import dataclass, field
from functools import partial
from typing import Optional
import json
import torch
from swift import Swift, get_logger
from swift.utils import parse_args, print_model_info, seed_everything
from tqdm import tqdm
from transformers import BitsAndBytesConfig, GenerationConfig, TextStreamer
from utils import (DEFAULT_PROMPT, MODEL_MAPPING, evaluate,
get_model_tokenizer, get_ms_tool_dataset_test, inference,
process_dataset, select_bnb, select_dtype, show_layers)
logger = get_logger()
class InferArguments:
model_type: str = field(
default='qwen-7b', metadata={'choices': list(MODEL_MAPPING.keys())})
sft_type: str = field(
default='lora', metadata={'choices': ['lora', 'full']})
ckpt_dir: str = '/path/to/your/vx_xxx/checkpoint-xxx'
eval_human: bool = False # False: eval test_dataset
seed: int = 42
dtype: str = field(
default='bf16', metadata={'choices': {'bf16', 'fp16', 'fp32'}})
ignore_args_error: bool = False # True: notebook compatibility
dataset: str = field(
default='alpaca-en,alpaca-zh', metadata={'help': 'dataset'})
dataset_seed: int = 42
dataset_sample: int = 20000 # -1: all dataset
dataset_test_size: float = 0.01
prompt: str = DEFAULT_PROMPT
max_length: Optional[int] = 1024
quantization_bit: Optional[int] = field(
default=None, metadata={'choices': {4, 8}})
bnb_4bit_comp_dtype: str = field(
default='fp16', metadata={'choices': {'fp16', 'bf16', 'fp32'}})
bnb_4bit_quant_type: str = field(
default='nf4', metadata={'choices': {'fp4', 'nf4'}})
bnb_4bit_use_double_quant: bool = True
max_new_tokens: int = 512
do_sample: bool = True
temperature: float = 0.9
top_k: int = 10
top_p: float = 0.8
def __post_init__(self):
if not os.path.isdir(self.ckpt_dir):
raise ValueError(f'Please enter a valid ckpt_dir: {self.ckpt_dir}')
self.torch_dtype, _, _ = select_dtype(self.dtype)
self.bnb_4bit_compute_dtype, self.load_in_4bit, self.load_in_8bit = select_bnb(
self.quantization_bit, self.bnb_4bit_comp_dtype)
def llm_infer(args: InferArguments) -> None:
logger.info(f'device_count: {torch.cuda.device_count()}')
seed_everything(args.seed)
# ### Loading Model and Tokenizer
kwargs = {'low_cpu_mem_usage': True, 'device_map': 'auto'}
if args.load_in_8bit or args.load_in_4bit:
quantization_config = BitsAndBytesConfig(
args.load_in_8bit,
args.load_in_4bit,
bnb_4bit_compute_dtype=args.bnb_4bit_compute_dtype,
bnb_4bit_quant_type=args.bnb_4bit_quant_type,
bnb_4bit_use_double_quant=args.bnb_4bit_use_double_quant)
logger.info(f'quantization_config: {quantization_config.__dict__}')
kwargs['quantization_config'] = quantization_config
if args.sft_type == 'full':
kwargs['model_dir'] = args.ckpt_dir
model, tokenizer = get_model_tokenizer(
args.model_type, torch_dtype=args.torch_dtype, **kwargs)
# ### Preparing lora
if args.sft_type == 'lora':
model = Swift.from_pretrained(model, args.ckpt_dir)
show_layers(model)
print_model_info(model)
# ### Inference
generation_config = GenerationConfig(
max_new_tokens=args.max_new_tokens,
temperature=args.temperature,
top_k=args.top_k,
top_p=args.top_p,
do_sample=args.do_sample,
pad_token_id=tokenizer.pad_token_id,
eos_token_id=tokenizer.eos_token_id)
logger.info(f'generation_config: {generation_config}')
dataset = get_ms_tool_dataset_test(args.dataset)
test_dataset, _ = process_dataset(dataset, args.dataset_test_size,
args.dataset_sample, args.dataset_seed)
del dataset
preds = []
labels = []
for data in tqdm(test_dataset):
label = data['labels']
inputs = data['inputs']
preds.append(inference(inputs, model, tokenizer, generation_config))
labels.append(label)
return labels, preds | null |
179,799 | import os
from dataclasses import dataclass, field
from functools import partial
from typing import List, Optional
import torch
import torch.distributed as dist
from swift import (HubStrategy, LoraConfig, Seq2SeqTrainer,
Seq2SeqTrainingArguments, Swift, get_logger)
from swift.llm.utils import data_collate_fn, print_example, stat_dataset
from swift.llm.utils.model import fix_gradient_checkpointing_warning
from swift.utils import (add_version_to_work_dir, get_model_info, is_master,
parse_args, seed_everything)
from transformers import BitsAndBytesConfig
from utils import (DEFAULT_PROMPT, MODEL_MAPPING, broadcast_string,
find_all_linear_for_lora, get_dist_setting,
get_model_tokenizer, get_ms_tool_dataset, is_dist,
plot_images, process_dataset, select_bnb, select_dtype,
show_layers, tokenize_function)
logger = get_logger()
class SftArguments:
model_type: str = field(
default='qwen-7b', metadata={'choices': list(MODEL_MAPPING.keys())})
# qwen-7b: lora+4bitQ: 10G, lora+8bitQ: 14G, lora: 22G; full: 95G
sft_type: str = field(
default='lora', metadata={'choices': ['lora', 'full']})
output_dir: str = 'runs'
# currently, DDP+MP is not supported
ddp_backend: Optional[str] = field(
default=None, metadata={'choices': ['nccl', 'gloo', 'mpi', 'ccl']})
seed: int = 42
resume_from_ckpt: Optional[str] = None
dtype: str = field(
default='fp16', metadata={'choices': {'bf16', 'fp16', 'fp32'}})
ignore_args_error: bool = False # True: notebook compatibility
dataset: str = field(
default='alpaca-en,alpaca-zh', metadata={'help': 'dataset'})
dataset_seed: int = 42
dataset_sample: int = 20000 # -1: all dataset
dataset_test_size: float = 0.01
prompt: str = DEFAULT_PROMPT
max_length: Optional[int] = 1024
# If you want to use qlora, set the quantization_bit to 8 or 4.
# And you need to install bitsandbytes: `pip install bitsandbytes -U`
# note: bf16 and quantization have requirements for gpu architecture
quantization_bit: Optional[int] = field(
default=None, metadata={'choices': {4, 8}})
bnb_4bit_comp_dtype: str = field(
default='fp16', metadata={'choices': {'fp16', 'bf16', 'fp32'}})
bnb_4bit_quant_type: str = field(
default='nf4', metadata={'choices': {'fp4', 'nf4'}})
bnb_4bit_use_double_quant: bool = True
lora_target_modules: Optional[List[str]] = None
lora_rank: int = 8
lora_alpha: int = 32
lora_dropout_p: float = 0.1
gradient_checkpoint: bool = True
batch_size: int = 1
num_train_epochs: int = 1
optim: str = 'adamw_torch'
learning_rate: Optional[float] = None
weight_decay: float = 0.01
gradient_accumulation_steps: int = 16
max_grad_norm: float = 1.
lr_scheduler_type: str = 'cosine'
warmup_ratio: float = 0.1
eval_steps: int = 50
save_steps: Optional[int] = None
save_total_limit: int = 2
logging_steps: int = 5
skip_memory_metrics: bool = True
push_to_hub: bool = False
# 'user_name/repo_name' or 'repo_name'
hub_model_id: Optional[str] = None
hub_private_repo: bool = True
hub_strategy: HubStrategy = HubStrategy.EVERY_SAVE
# None: use env var `MODELSCOPE_API_TOKEN`
hub_token: Optional[str] = None
# fsdp
fsdp: Optional[str] = None
fsdp_config: Optional[str] = None
# deepspeed
deepspeed: Optional[str] = None
# other
use_flash_attn: Optional[bool] = field(
default=None,
metadata={
'help': "This parameter is used only when model_type='qwen-7b'"
})
def __post_init__(self):
if is_dist():
rank, _, _ = get_dist_setting()
self.seed += rank # Avoid the same dropout
if self.ddp_backend is None:
self.ddp_backend = 'nccl'
# Initialize in advance
dist.init_process_group(backend=self.ddp_backend)
if self.sft_type == 'lora':
if self.learning_rate is None:
self.learning_rate = 1e-4
if self.save_steps is None:
self.save_steps = self.eval_steps
elif self.sft_type == 'full':
assert self.quantization_bit is None, 'not supported'
assert self.dtype != 'fp16', 'please use bf16 or fp32'
if self.learning_rate is None:
self.learning_rate = 1e-5
if self.save_steps is None:
# Saving the model takes a long time
self.save_steps = self.eval_steps * 4
else:
raise ValueError(f'sft_type: {self.sft_type}')
self.output_dir = os.path.join(self.output_dir, self.model_type)
if self.lora_target_modules is None:
self.lora_target_modules = MODEL_MAPPING[
self.model_type]['lora_TM']
self.torch_dtype, self.fp16, self.bf16 = select_dtype(self.dtype)
self.bnb_4bit_compute_dtype, self.load_in_4bit, self.load_in_8bit = select_bnb(
self.quantization_bit, self.bnb_4bit_comp_dtype)
if self.hub_model_id is None:
self.hub_model_id = f'{self.model_type}-sft'
if self.use_flash_attn is None:
self.use_flash_attn = 'auto'
def llm_sft(args: SftArguments) -> None:
logger.info(f'device_count: {torch.cuda.device_count()}')
rank, local_rank, world_size = get_dist_setting()
logger.info(
f'rank: {rank}, local_rank: {local_rank}, world_size: {world_size}')
seed_everything(args.seed)
# ### Loading Model and Tokenizer
kwargs = {'low_cpu_mem_usage': True}
if is_dist():
kwargs['device_map'] = {'': local_rank}
else:
kwargs['device_map'] = 'auto'
if args.load_in_8bit or args.load_in_4bit:
quantization_config = BitsAndBytesConfig(
args.load_in_8bit,
args.load_in_4bit,
bnb_4bit_compute_dtype=args.bnb_4bit_compute_dtype,
bnb_4bit_quant_type=args.bnb_4bit_quant_type,
bnb_4bit_use_double_quant=args.bnb_4bit_use_double_quant)
logger.info(f'quantization_config: {quantization_config.__dict__}')
kwargs['quantization_config'] = quantization_config
if args.model_type == 'qwen-7b':
kwargs['use_flash_attn'] = args.use_flash_attn
model, tokenizer = get_model_tokenizer(
args.model_type, torch_dtype=args.torch_dtype, **kwargs)
# ### Preparing lora
if args.sft_type == 'lora':
if 'ALL' in args.lora_target_modules:
assert len(args.lora_target_modules) == 1
args.lora_target_modules = find_all_linear_for_lora(
model, args.quantization_bit, args.model_type)
logger.info(
f'Setting lora_target_modules: {args.lora_target_modules}')
if args.resume_from_ckpt is None:
lora_config = LoraConfig(
r=args.lora_rank,
target_modules=args.lora_target_modules,
lora_alpha=args.lora_alpha,
lora_dropout=args.lora_dropout_p,
task_type='CAUSAL_LM')
logger.info(f'lora_config: {lora_config}')
model = Swift.prepare_model(model, lora_config)
else:
model = Swift.from_pretrained(
model, args.resume_from_ckpt, is_trainable=True)
is_logging = False
for p in model.parameters():
if p.requires_grad and p.dtype == torch.float16:
if not is_logging:
logger.info(
'Convert trainable parameters from fp16 to fp32.')
is_logging = True
p.data = p.data.to(dtype=torch.float32)
# # for fsdp
# if args.fp16:
# model = model.half()
# if args.bf16:
# model = model.bfloat16()
show_layers(model)
get_model_info(model)
# ### Loading Dataset
dataset = get_ms_tool_dataset(args.dataset)
train_dataset, val_dataset = process_dataset(dataset,
args.dataset_test_size,
args.dataset_sample,
args.dataset_seed)
tokenize_func = partial(
tokenize_function, tokenizer=tokenizer, max_length=args.max_length)
train_dataset = train_dataset.map(tokenize_func)
val_dataset = val_dataset.map(tokenize_func)
del dataset
# Data analysis
stat_dataset(train_dataset)
stat_dataset(val_dataset)
data_collator = partial(data_collate_fn, tokenizer=tokenizer)
# print_example(train_dataset[0], tokenizer)
# ### Setting trainer_args
output_dir = None
if is_master():
output_dir = add_version_to_work_dir(args.output_dir)
if is_dist():
output_dir = broadcast_string(output_dir)
trainer_args = Seq2SeqTrainingArguments(
output_dir=output_dir,
do_train=True,
do_eval=True,
evaluation_strategy='steps',
per_device_train_batch_size=args.batch_size,
per_device_eval_batch_size=1,
gradient_accumulation_steps=args.gradient_accumulation_steps,
learning_rate=args.learning_rate,
weight_decay=args.weight_decay,
max_grad_norm=args.max_grad_norm,
num_train_epochs=args.num_train_epochs,
lr_scheduler_type=args.lr_scheduler_type,
warmup_ratio=args.warmup_ratio,
logging_steps=args.logging_steps,
save_strategy='steps',
save_steps=args.save_steps,
save_total_limit=args.save_total_limit,
bf16=args.bf16,
fp16=args.fp16,
eval_steps=args.eval_steps,
dataloader_num_workers=1,
load_best_model_at_end=True,
metric_for_best_model='loss',
greater_is_better=False,
sortish_sampler=True,
optim=args.optim,
hub_model_id=args.hub_model_id,
hub_private_repo=args.hub_private_repo,
hub_strategy=args.hub_strategy,
hub_token=args.hub_token,
push_to_hub=args.push_to_hub,
resume_from_checkpoint=args.resume_from_ckpt,
ddp_backend=args.ddp_backend,
gradient_checkpointing=args.gradient_checkpoint,
local_rank=local_rank,
skip_memory_metrics=args.skip_memory_metrics,
fsdp=args.fsdp or False,
fsdp_config=args.fsdp_config,
deepspeed=args.deepspeed)
if args.gradient_checkpoint:
# fix: gradients will be None
model.enable_input_require_grads()
if is_dist():
trainer_args.ddp_find_unused_parameters = False
trainer_args.ddp_broadcast_buffers = False
logger.info(f'trainer_args: {trainer_args}')
trainer = Seq2SeqTrainer(
model=model,
args=trainer_args,
data_collator=data_collator,
train_dataset=train_dataset,
eval_dataset=val_dataset,
tokenizer=tokenizer,
)
train_result = trainer.train(trainer_args.resume_from_checkpoint)
# save metrics
metrics = train_result.metrics
trainer.log_metrics("train", metrics)
trainer.save_metrics("train", metrics)
# ### Visualization
# if is_master():
# images_dir = os.path.join(output_dir, 'images')
# tb_dir = os.path.join(output_dir, 'runs')
# folder_name = os.listdir(tb_dir)[0]
# tb_dir = os.path.join(tb_dir, folder_name)
# plot_images(images_dir, tb_dir, ['train/loss'], 0.9)
# if args.push_to_hub:
# trainer._add_patterns_to_gitignores(['images/'])
# trainer.push_to_hub() | null |
179,800 | import os
import re
from typing import List, Optional, Tuple
import json
import matplotlib.pyplot as plt
import torch
import torch.distributed as dist
from rouge import Rouge
from swift import get_logger
from swift.utils.tb_utils import (TB_COLOR, TB_COLOR_SMOOTH,
read_tensorboard_file, tensorboard_smoothing)
from torch import dtype as Dtype
from torch.nn import Module
from transformers import GenerationConfig, TextStreamer
os.environ['TOKENIZERS_PARALLELISM'] = 'true'
def is_dist():
rank = int(os.getenv('RANK', -1))
return rank >= 0 | null |
179,801 | import os
import re
from typing import List, Optional, Tuple
import json
import matplotlib.pyplot as plt
import torch
import torch.distributed as dist
from rouge import Rouge
from swift import get_logger
from swift.utils.tb_utils import (TB_COLOR, TB_COLOR_SMOOTH,
read_tensorboard_file, tensorboard_smoothing)
from torch import dtype as Dtype
from torch.nn import Module
from transformers import GenerationConfig, TextStreamer
logger = get_logger()
def show_layers(model: Module, max_lines: Optional[int] = 20) -> None:
named_p = list(model.named_parameters())
for i, (n, p) in enumerate(named_p):
if max_lines is not None and i >= max_lines:
logger.info('...')
break
logger.info(
f'[{n}]: requires_grad={p.requires_grad}, dtype={p.dtype}, device={p.device}'
) | null |
179,802 | import os
import re
from typing import List, Optional, Tuple
import json
import matplotlib.pyplot as plt
import torch
import torch.distributed as dist
from rouge import Rouge
from swift import get_logger
from swift.utils.tb_utils import (TB_COLOR, TB_COLOR_SMOOTH,
read_tensorboard_file, tensorboard_smoothing)
from torch import dtype as Dtype
from torch.nn import Module
from transformers import GenerationConfig, TextStreamer
os.environ['TOKENIZERS_PARALLELISM'] = 'true'
def plot_images(images_dir: str,
tb_dir: str,
smooth_key: List[str],
smooth_val: float = 0.9,
figsize: Tuple[int, int] = (8, 5),
dpi: int = 100) -> None:
os.makedirs(images_dir, exist_ok=True)
fname = [
fname for fname in os.listdir(tb_dir)
if os.path.isfile(os.path.join(tb_dir, fname))
][0]
tb_path = os.path.join(tb_dir, fname)
data = read_tensorboard_file(tb_path)
for k in data.keys():
_data = data[k]
steps = [d['step'] for d in _data]
values = [d['value'] for d in _data]
if len(values) == 0:
continue
_, ax = plt.subplots(1, 1, squeeze=True, figsize=figsize, dpi=dpi)
ax.set_title(k)
if len(values) == 1:
ax.scatter(steps, values, color=TB_COLOR_SMOOTH)
elif k in smooth_key:
ax.plot(steps, values, color=TB_COLOR)
values_s = tensorboard_smoothing(values, smooth_val)
ax.plot(steps, values_s, color=TB_COLOR_SMOOTH)
else:
ax.plot(steps, values, color=TB_COLOR_SMOOTH)
fpath = os.path.join(images_dir, k.replace('/', '_'))
plt.savefig(fpath, dpi=dpi, bbox_inches='tight') | null |
179,803 | import os
import re
from typing import List, Optional, Tuple
import json
import matplotlib.pyplot as plt
import torch
import torch.distributed as dist
from rouge import Rouge
from swift import get_logger
from swift.utils.tb_utils import (TB_COLOR, TB_COLOR_SMOOTH,
read_tensorboard_file, tensorboard_smoothing)
from torch import dtype as Dtype
from torch.nn import Module
from transformers import GenerationConfig, TextStreamer
def inference(
data,
model,
tokenizer,
generation_config: Optional[GenerationConfig] = None,
) -> str:
input_ids = tokenizer(data, return_tensors='pt').input_ids.cuda()
input_len = input_ids.shape[1]
attention_mask = torch.ones_like(input_ids)
result_id = model.generate(
input_ids=input_ids,
attention_mask=attention_mask,
generation_config=generation_config)
result_id[result_id >= tokenizer.vocab_size] = tokenizer.pad_token_id
result_id = result_id[0].tolist()[input_len:]
result = tokenizer.decode(result_id, skip_special_tokens=True)
return result | null |
179,804 | import os
import re
from typing import List, Optional, Tuple
import json
import matplotlib.pyplot as plt
import torch
import torch.distributed as dist
from rouge import Rouge
from swift import get_logger
from swift.utils.tb_utils import (TB_COLOR, TB_COLOR_SMOOTH,
read_tensorboard_file, tensorboard_smoothing)
from torch import dtype as Dtype
from torch.nn import Module
from transformers import GenerationConfig, TextStreamer
logger = get_logger()
DTYPE_MAPPING = {
'fp16': torch.float16,
'bf16': torch.bfloat16,
'fp32': torch.float32
}
The provided code snippet includes necessary dependencies for implementing the `select_dtype` function. Write a Python function `def select_dtype(dtype: str) -> Tuple[Dtype, bool, bool]` to solve the following problem:
dtype: Literal['fp16', 'bf16', 'fp32']
Here is the function:
def select_dtype(dtype: str) -> Tuple[Dtype, bool, bool]:
"""
dtype: Literal['fp16', 'bf16', 'fp32']
"""
torch_dtype = DTYPE_MAPPING[dtype]
assert torch_dtype in {torch.float16, torch.bfloat16, torch.float32}
if torch_dtype == torch.float16:
fp16, bf16 = True, False
elif torch_dtype == torch.bfloat16:
support_bf16 = torch.cuda.is_bf16_supported()
if not support_bf16:
logger.warning(f'support_bf16: {support_bf16}')
fp16, bf16 = False, True
else:
fp16, bf16 = False, False
return torch_dtype, fp16, bf16 | dtype: Literal['fp16', 'bf16', 'fp32'] |
179,805 | import os
import re
from typing import List, Optional, Tuple
import json
import matplotlib.pyplot as plt
import torch
import torch.distributed as dist
from rouge import Rouge
from swift import get_logger
from swift.utils.tb_utils import (TB_COLOR, TB_COLOR_SMOOTH,
read_tensorboard_file, tensorboard_smoothing)
from torch import dtype as Dtype
from torch.nn import Module
from transformers import GenerationConfig, TextStreamer
The provided code snippet includes necessary dependencies for implementing the `find_all_linear_for_lora` function. Write a Python function `def find_all_linear_for_lora(model: Module, quantization_bit: Optional[int], model_type: Optional[str] = None) -> List[str]` to solve the following problem:
ref: https://github.com/artidoro/qlora
Here is the function:
def find_all_linear_for_lora(model: Module,
quantization_bit: Optional[int],
model_type: Optional[str] = None) -> List[str]:
"""ref: https://github.com/artidoro/qlora"""
head_module_name = 'lm_head'
if model_type.startswith('chatglm2-6b'):
head_module_name = 'output_layer'
if model_type.startswith('qwen-vl'):
return ['c_attn', 'attn.c_proj', 'w1', 'w2']
if quantization_bit == 4:
from bitsandbytes.nn import Linear4bit
linear_cls = Linear4bit
elif quantization_bit == 8:
from bitsandbytes.nn import Linear8bitLt
linear_cls = Linear8bitLt
else:
linear_cls = Linear
lora_module_names = set()
for name, module in model.named_modules():
if isinstance(module, linear_cls):
module_name = name.split('.')[-1]
if head_module_name not in module_name:
lora_module_names.add(module_name)
return list(lora_module_names) | ref: https://github.com/artidoro/qlora |
179,806 | import os
import re
from typing import List, Optional, Tuple
import json
import matplotlib.pyplot as plt
import torch
import torch.distributed as dist
from rouge import Rouge
from swift import get_logger
from swift.utils.tb_utils import (TB_COLOR, TB_COLOR_SMOOTH,
read_tensorboard_file, tensorboard_smoothing)
from torch import dtype as Dtype
from torch.nn import Module
from transformers import GenerationConfig, TextStreamer
DTYPE_MAPPING = {
'fp16': torch.float16,
'bf16': torch.bfloat16,
'fp32': torch.float32
}
def select_bnb(quantization_bit: Optional[int],
bnb_4bit_compute_dtype: str) -> Tuple[Dtype, bool, bool]:
bnb_4bit_compute_dtype = DTYPE_MAPPING[bnb_4bit_compute_dtype]
assert bnb_4bit_compute_dtype in {
torch.float16, torch.bfloat16, torch.float32
}
if quantization_bit == 4:
load_in_4bit, load_in_8bit = True, False
elif quantization_bit == 8:
load_in_4bit, load_in_8bit = False, True
else:
load_in_4bit, load_in_8bit = False, False
return bnb_4bit_compute_dtype, load_in_4bit, load_in_8bit | null |
179,807 | import os
import re
from typing import List, Optional, Tuple
import json
import matplotlib.pyplot as plt
import torch
import torch.distributed as dist
from rouge import Rouge
from swift import get_logger
from swift.utils.tb_utils import (TB_COLOR, TB_COLOR_SMOOTH,
read_tensorboard_file, tensorboard_smoothing)
from torch import dtype as Dtype
from torch.nn import Module
from transformers import GenerationConfig, TextStreamer
def get_dist_setting() -> Tuple[int, int, int]:
"""return rank, local_rank, world_size"""
rank = int(os.getenv('RANK', -1))
local_rank = int(os.getenv('LOCAL_RANK', -1))
world_size = int(os.getenv('WORLD_SIZE', 1))
return rank, local_rank, world_size
The provided code snippet includes necessary dependencies for implementing the `broadcast_string` function. Write a Python function `def broadcast_string(string: Optional[str], buffer_size: int = 100) -> str` to solve the following problem:
string: main rank: str other rank: None return: all rank: str
Here is the function:
def broadcast_string(string: Optional[str], buffer_size: int = 100) -> str:
"""
string: main rank: str
other rank: None
return: all rank: str
"""
assert dist.is_initialized()
rank, local_rank, _ = get_dist_setting()
assert rank >= 0
if rank == 0:
assert string is not None
tensor = torch.tensor(
[ord(c) for c in string] + [0] * (buffer_size - len(string)),
dtype=torch.int64,
device=local_rank)
else:
tensor = torch.zeros(buffer_size, dtype=torch.int64, device=local_rank)
dist.broadcast(tensor, 0)
first_zero = (tensor == 0).nonzero()[0].item()
res = tensor.tolist()[:first_zero]
return ''.join([chr(x) for x in res]) | string: main rank: str other rank: None return: all rank: str |
179,808 | import os
import re
from typing import List, Optional, Tuple
import json
import matplotlib.pyplot as plt
import torch
import torch.distributed as dist
from rouge import Rouge
from swift import get_logger
from swift.utils.tb_utils import (TB_COLOR, TB_COLOR_SMOOTH,
read_tensorboard_file, tensorboard_smoothing)
from torch import dtype as Dtype
from torch.nn import Module
from transformers import GenerationConfig, TextStreamer
def evaluate(refs, preds):
# action: em
# action input: em
# answer: rouge
re_pattern1 = re.compile(
pattern=r'<\|startofthink\|>([\s\S]+)<\|endofthink\|>')
re_pattern2 = re.compile(r'{[\s\S]+}')
action_em = []
input_em = []
ref_seqs = []
pred_seqs = []
for i, (ref, pred) in enumerate(zip(refs, preds)):
try:
think_content = re_pattern1.search(ref).group(1)
think_content = re_pattern2.search(think_content).group()
r = json.loads(think_content.replace('\n', ''))
try:
think_content = re_pattern1.search(pred).group(1)
think_content = re_pattern2.search(think_content).group()
p = json.loads(think_content.replace('\n', ''))
if p['api_name'] == r['api_name']:
action_em.append(1)
else:
action_em.append(0)
r_input = r['parameters']
p_input = p['parameters']
match = True
for k, v in r_input.items():
if k in p_input.keys() and p_input[k] == v:
continue
else:
match = False
break
for k in p_input.keys():
if k not in r_input.keys():
match = False
break
if match:
input_em.append(1)
else:
input_em.append(0)
except Exception:
action_em.append(0)
input_em.append(0)
except Exception:
ref_seqs.append(ref)
pred_seqs.append(pred if pred else ' ')
rouge = Rouge()
rouge_score = rouge.get_scores(hyps=pred_seqs, refs=ref_seqs, avg=True)
rougel = rouge_score["rouge-l"]["f"]
return {
'action_em': sum(action_em) / len(action_em),
'input_em': sum(input_em) / len(input_em),
'rouge': rougel
} | null |
179,809 | import os
from types import MethodType
from typing import Any, Dict, NamedTuple, Optional
import torch
from swift import get_logger
from torch import dtype as Dtype
from modelscope import (AutoConfig, AutoModelForCausalLM, AutoTokenizer, Model,
read_config, snapshot_download)
from modelscope.models.nlp.chatglm2 import ChatGLM2Config, ChatGLM2Tokenizer
from modelscope.models.nlp.qwen import QWenConfig, QWenTokenizer
def get_model_tokenizer_from_repo(model_dir: str,
torch_dtype: Dtype,
load_model: bool = True,
model_config=None,
**model_kwargs):
"""load from an independent repository"""
if model_config is None:
model_config = AutoConfig.from_pretrained(
model_dir, trust_remote_code=True)
model_config.torch_dtype = torch_dtype
logger.info(f'model_config: {model_config}')
tokenizer = AutoTokenizer.from_pretrained(
model_dir, trust_remote_code=True)
model = None
if load_model:
model = AutoModelForCausalLM.from_pretrained(
model_dir,
config=model_config,
torch_dtype=torch_dtype,
trust_remote_code=True,
**model_kwargs)
return model, tokenizer
def get_model_tokenizer_baichuan13b(model_dir: str,
torch_dtype: Dtype,
load_model: bool = True,
**model_kwargs):
# baichuan-13b does not implement the `get_input_embeddings` function
model, tokenizer = get_model_tokenizer_from_repo(model_dir, torch_dtype,
load_model,
**model_kwargs)
model.get_input_embeddings = MethodType(
lambda self: self.model.embed_tokens, model)
return model, tokenizer | null |
179,810 | import os
from types import MethodType
from typing import Any, Dict, NamedTuple, Optional
import torch
from swift import get_logger
from torch import dtype as Dtype
from modelscope import (AutoConfig, AutoModelForCausalLM, AutoTokenizer, Model,
read_config, snapshot_download)
from modelscope.models.nlp.chatglm2 import ChatGLM2Config, ChatGLM2Tokenizer
from modelscope.models.nlp.qwen import QWenConfig, QWenTokenizer
def get_model_tokenizer_from_sdk(config_class,
tokenizer_class,
model_dir: str,
torch_dtype: Dtype,
load_model: bool = True,
**model_kwargs):
"""load from ms library"""
config = read_config(model_dir)
logger.info(config)
model_config = config_class.from_pretrained(model_dir)
model_config.torch_dtype = torch_dtype
logger.info(model_config)
tokenizer = tokenizer_class.from_pretrained(model_dir)
model = None
if load_model:
model = Model.from_pretrained(
model_dir,
cfg_dict=config,
config=model_config,
torch_dtype=torch_dtype,
**model_kwargs)
return model, tokenizer
def get_model_tokenizer_chatglm2(model_dir: str,
torch_dtype: Dtype,
load_model: bool = True,
**model_kwargs):
if 'quantization_config' in model_kwargs:
model_kwargs['quantization_config'].llm_int8_skip_modules = [
'output_layer'
]
return get_model_tokenizer_from_sdk(ChatGLM2Config, ChatGLM2Tokenizer,
model_dir, torch_dtype, load_model,
**model_kwargs) | null |
179,811 | import os
from types import MethodType
from typing import Any, Dict, NamedTuple, Optional
import torch
from swift import get_logger
from torch import dtype as Dtype
from modelscope import (AutoConfig, AutoModelForCausalLM, AutoTokenizer, Model,
read_config, snapshot_download)
from modelscope.models.nlp.chatglm2 import ChatGLM2Config, ChatGLM2Tokenizer
from modelscope.models.nlp.qwen import QWenConfig, QWenTokenizer
def get_model_tokenizer_from_repo(model_dir: str,
torch_dtype: Dtype,
load_model: bool = True,
model_config=None,
**model_kwargs):
"""load from an independent repository"""
if model_config is None:
model_config = AutoConfig.from_pretrained(
model_dir, trust_remote_code=True)
model_config.torch_dtype = torch_dtype
logger.info(f'model_config: {model_config}')
tokenizer = AutoTokenizer.from_pretrained(
model_dir, trust_remote_code=True)
model = None
if load_model:
model = AutoModelForCausalLM.from_pretrained(
model_dir,
config=model_config,
torch_dtype=torch_dtype,
trust_remote_code=True,
**model_kwargs)
return model, tokenizer
def get_model_tokenizer_qwen(model_dir: str,
torch_dtype: Dtype,
load_model: bool = True,
**kwargs):
model_config = AutoConfig.from_pretrained(
model_dir, trust_remote_code=True)
mapper = {
torch.float16: 'fp16',
torch.bfloat16: 'bf16',
torch.float32: 'fp32'
}
k_true = mapper[torch_dtype]
for k in mapper.values():
v = False
if k == k_true:
v = True
setattr(model_config, k, v)
use_flash_attn = kwargs.pop('use_flash_attn', 'auto')
model_config.use_flash_attn = use_flash_attn
return get_model_tokenizer_from_repo(model_dir, torch_dtype, load_model,
model_config, **kwargs) | null |
179,812 | import os
from types import MethodType
from typing import Any, Dict, NamedTuple, Optional
import torch
from swift import get_logger
from torch import dtype as Dtype
from modelscope import (AutoConfig, AutoModelForCausalLM, AutoTokenizer, Model,
read_config, snapshot_download)
from modelscope.models.nlp.chatglm2 import ChatGLM2Config, ChatGLM2Tokenizer
from modelscope.models.nlp.qwen import QWenConfig, QWenTokenizer
def _add_special_token(tokenizer, special_token_mapper: Dict[str,
Any]) -> None:
for k, v in special_token_mapper.items():
setattr(tokenizer, k, v)
assert tokenizer.eos_token is not None
if tokenizer.pad_token is None:
tokenizer.pad_token = tokenizer.eos_token
def get_model_tokenizer_from_repo(model_dir: str,
torch_dtype: Dtype,
load_model: bool = True,
model_config=None,
**model_kwargs):
"""load from an independent repository"""
if model_config is None:
model_config = AutoConfig.from_pretrained(
model_dir, trust_remote_code=True)
model_config.torch_dtype = torch_dtype
logger.info(f'model_config: {model_config}')
tokenizer = AutoTokenizer.from_pretrained(
model_dir, trust_remote_code=True)
model = None
if load_model:
model = AutoModelForCausalLM.from_pretrained(
model_dir,
config=model_config,
torch_dtype=torch_dtype,
trust_remote_code=True,
**model_kwargs)
return model, tokenizer
MODEL_MAPPING = {
'baichuan-7b': {
'model_id': 'baichuan-inc/baichuan-7B', # model id or model dir
'revision': 'v1.0.7',
'lora_TM': LoRATM.baichuan
},
'baichuan-13b': {
'model_id': 'baichuan-inc/Baichuan-13B-Base',
'revision': 'v1.0.3',
'lora_TM': LoRATM.baichuan
},
'chatglm2-6b': {
'model_id': 'ZhipuAI/chatglm2-6b',
'revision': 'v1.0.7',
'get_function': get_model_tokenizer_chatglm2,
'lora_TM': LoRATM.chatglm2
},
'llama2-7b': {
'model_id': 'modelscope/Llama-2-7b-ms',
'revision': 'v1.0.2',
'ignore_file_pattern': [r'.+\.bin$'], # use safetensors
'lora_TM': LoRATM.llama2
},
'llama2-13b': {
'model_id': 'modelscope/Llama-2-13b-ms',
'revision': 'v1.0.2',
'ignore_file_pattern': [r'.+\.bin$'],
'lora_TM': LoRATM.llama2
},
'llama2-70b': {
'model_id': 'modelscope/Llama-2-70b-ms',
'revision': 'v1.0.0',
'ignore_file_pattern': [r'.+\.bin$'],
'lora_TM': LoRATM.llama2
},
'openbuddy-llama2-13b': {
'model_id': 'OpenBuddy/openbuddy-llama2-13b-v8.1-fp16',
'revision': 'v1.0.0',
'lora_TM': LoRATM.llama2,
},
'qwen-7b': {
'model_id': 'qwen/Qwen-7B',
'revision': 'v1.1.8',
'get_function': get_model_tokenizer_qwen,
'lora_TM': LoRATM.qwen,
'special_token_mapper': {
'eos_token': '<|endoftext|>'
}
},
'modelscope-agent-7b': {
'model_id': 'damo/ModelScope-Agent-7B',
'revision': 'v1.0.0',
'get_function': get_model_tokenizer_qwen,
'lora_TM': LoRATM.qwen,
'special_token_mapper': {
'eos_token': '<|endoftext|>'
}
}
}
def get_model_tokenizer(model_type: str,
torch_dtype: Optional[Dtype] = None,
load_model: bool = True,
**kwargs):
data = MODEL_MAPPING.get(model_type)
if data is None:
raise ValueError(f'model_type: {model_type}')
model_id = data['model_id']
get_function = data.get('get_function', get_model_tokenizer_from_repo)
ignore_file_pattern = data.get('ignore_file_pattern', [])
special_token_mapper = data.get('special_token_mapper', {})
if torch_dtype is None:
torch_dtype = data.get('torch_dtype', torch.float16)
model_dir = kwargs.pop('model_dir', None)
if model_dir is None:
model_dir = model_id
if not os.path.exists(model_id):
revision = data.get('revision', 'master')
model_dir = snapshot_download(
model_id, revision, ignore_file_pattern=ignore_file_pattern)
model, tokenizer = get_function(model_dir, torch_dtype, load_model,
**kwargs)
_add_special_token(tokenizer, special_token_mapper)
return model, tokenizer | null |
179,813 | import ast
import os
import re
from typing import List, Optional, Tuple
import json
import numpy as np
import torch
from datasets import Dataset as HfDataset
from datasets import IterableDataset, concatenate_datasets
from swift.utils import get_seed
from modelscope import MsDataset
def get_ms_tool_dataset(dataset_name_or_file) -> HfDataset:
# ms_tool_dataset: for train
# each data may contain multiple segments, they are organized as a list
# and split by flag. 0 for user input/ tool execute result..., 1 for label
if os.path.isfile(dataset_name_or_file):
with open(dataset_name_or_file, 'r') as f:
if dataset_name_or_file.endswith('.json'):
origin_data = json.load(f)
elif dataset_name_or_file.endswith('.jsonl'):
origin_data = []
for line in f:
origin_data.append(json.loads(line))
else:
origin_data = MsDataset.load('damo/MSAgent-Bench', split='train')
all_inputs_str = []
all_inputs_flag = []
for d in origin_data:
content = d['conversations']
if isinstance(content, str):
content = ast.literal_eval(content)
# ilegal data
if len(content) == 0 or content[0]['from'] != 'system':
continue
system_str = '<|system|>:' + content[0]['value']
inputs_str = [system_str] # segment of conservations
inputs_flag = [
0
] # a flag to indicate whether the segment is assistant response
for i in range(len(content) // 2):
if len(content[2 * i + 2]['value']) == 0:
continue
assert content[2 * i + 1]['from'] == 'user'
assert content[2 * i + 2]['from'] == 'assistant'
# user input
inputs_str.append('\n\n<|user|>:' + content[2 * i + 1]['value'])
inputs_flag.append(0)
# assistant response
origin_response_str = '\n\n<|assistant|>:' + content[
2 * i + 2]['value'] + '\n\n</s>'
idx1, idx2 = -1, 0
iter1 = re.finditer(r'<\|startofexec\|>', origin_response_str)
iter2 = re.finditer(r'<\|endofexec\|>', origin_response_str)
for i1, i2 in zip(iter1, iter2):
idx1 = i1.start()
# llm response
inputs_str.append(origin_response_str[idx2:idx1])
inputs_flag.append(1)
idx2 = i2.end()
# exec result
inputs_str.append(origin_response_str[idx1:idx2])
inputs_flag.append(0)
if idx2 != len(origin_response_str):
inputs_str.append(origin_response_str[idx2:])
inputs_flag.append(1)
if len(inputs_flag) == 1:
continue
all_inputs_str.append(inputs_str)
all_inputs_flag.append(inputs_flag)
dataset = HfDataset.from_dict({
'inputs': all_inputs_str,
'flags': all_inputs_flag
})
return dataset | null |
179,814 | import ast
import os
import re
from typing import List, Optional, Tuple
import json
import numpy as np
import torch
from datasets import Dataset as HfDataset
from datasets import IterableDataset, concatenate_datasets
from swift.utils import get_seed
from modelscope import MsDataset
def get_ms_tool_dataset_test(dataset_name_or_file):
# ms_tool_dataset: for train
# each data may contain multiple segments, they are organized as different samples
all_inputs_str = []
all_labels_str = []
if os.path.isfile(dataset_name_or_file):
with open(dataset_json_file, 'r') as f:
if dataset_json_file.endswith('.json'):
origin_data = json.load(f)
elif dataset_json_file.endswith('.jsonl'):
origin_data = []
for line in f:
origin_data.append(json.loads(line))
else:
origin_data = MsDataset.load('damo/MSAgent-Bench', split='validation')
for d in origin_data:
content = d['conversations']
if isinstance(content, str):
content = ast.literal_eval(content)
# ilegal data
if len(content) == 0 or content[0]['from'] != 'system':
continue
system_str = '<|system|>:' + content[0]['value']
input_str = system_str
for i in range(len(content) // 2):
if len(content[2 * i + 2]['value']) == 0:
continue
assert content[2 * i + 1]['from'] == 'user'
assert content[2 * i + 2]['from'] == 'assistant'
# user input
input_str += ('\n\n<|user|>:' + content[2 * i + 1]['value'])
# assistant response
origin_response_str = '\n\n<|assistant|>:' + content[2 * i
+ 2]['value']
idx2 = 0
iter1 = re.finditer(r'<\|startofexec\|>', origin_response_str)
iter2 = re.finditer(r'<\|endofexec\|>', origin_response_str)
for i1, i2 in zip(iter1, iter2):
idx1 = i1.start()
# llm response
llm_response = origin_response_str[idx2:idx1]
all_inputs_str.append(input_str)
all_labels_str.append(llm_response)
input_str += llm_response
idx2 = i2.end()
# exec result
exec_result = origin_response_str[idx1:idx2]
input_str += exec_result
# summarize
if idx2 != len(origin_response_str):
final_summarize = origin_response_str[idx2:]
all_inputs_str.append(input_str)
all_labels_str.append(final_summarize)
dataset = HfDataset.from_dict({
'inputs': all_inputs_str,
'labels': all_labels_str
})
return dataset | null |
179,815 | import ast
import os
import re
from typing import List, Optional, Tuple
import json
import numpy as np
import torch
from datasets import Dataset as HfDataset
from datasets import IterableDataset, concatenate_datasets
from swift.utils import get_seed
from modelscope import MsDataset
def process_dataset(dataset: HfDataset, dataset_test_size: float,
dataset_sample: int,
dataset_seed: int) -> Tuple[HfDataset, HfDataset]:
random_state = np.random.RandomState(dataset_seed)
if dataset_sample >= 0:
index = random_state.permutation(len(dataset))[:dataset_sample]
dataset = dataset.select(index)
if dataset_test_size == 1.0:
return dataset, None
dataset = dataset.train_test_split(
dataset_test_size, seed=get_seed(random_state))
return dataset['train'], dataset['test'] | null |
179,816 | import ast
import os
import re
from typing import List, Optional, Tuple
import json
import numpy as np
import torch
from datasets import Dataset as HfDataset
from datasets import IterableDataset, concatenate_datasets
from swift.utils import get_seed
from modelscope import MsDataset
IGNORE_INDEX = -100
def tokenize_function(example, tokenizer, max_length: Optional[int] = None):
input_str, input_flag = example['inputs'], example['flags']
input_tokenized = tokenizer(
input_str,
# return_tensors='pt',
add_special_tokens=True,
return_attention_mask=False)['input_ids']
# transform list to tensor and cat them in dim=0
input_id = torch.cat([torch.tensor(inp) for inp in input_tokenized], dim=0)
# if flag = 1, the token should be origin result, if 0, it should be ignored
label = []
for inp, flag in zip(input_tokenized, input_flag):
inp = torch.tensor(inp)
label.append(inp.clone() if flag == 1 else torch.ones_like(inp)
* IGNORE_INDEX)
label = torch.cat(label, dim=0)
if max_length is not None and input_id.shape[0] > max_length:
input_id = input_id[-max_length:]
label = label[-max_length:]
if torch.max(label) == IGNORE_INDEX:
print(input_str)
print('Warning: no label in this data')
return dict(input_ids=input_id, labels=label) | null |
179,817 | import argparse
import json
from tqdm import tqdm
import datasets
import transformers
def preprocess(tokenizer, config, example, max_seq_length, version):
if version == 'v1':
prompt = example["context"]
target = example["target"]
prompt_ids = tokenizer.encode(prompt, max_length=max_seq_length, truncation=True)
target_ids = tokenizer.encode(
target,
max_length=max_seq_length,
truncation=True,
add_special_tokens=False)
input_ids = prompt_ids + target_ids + [config.eos_token_id]
return {"input_ids": input_ids, "seq_len": len(prompt_ids)}
if version == 'v2':
query = example["context"]
target = example["target"]
history = None
prompt = tokenizer.build_prompt(query, history)
a_ids = tokenizer.encode(text=prompt, add_special_tokens=True, truncation=True,
max_length=max_seq_length)
b_ids = tokenizer.encode(text=target, add_special_tokens=False, truncation=True,
max_length=max_seq_length)
input_ids = a_ids + b_ids + [tokenizer.eos_token_id]
return {"input_ids": input_ids, "seq_len": len(a_ids)}
def read_jsonl(path, max_seq_length, chatglm_path, version='v1', skip_overlength=False):
tokenizer = transformers.AutoTokenizer.from_pretrained(
chatglm_path, trust_remote_code=True)
config = transformers.AutoConfig.from_pretrained(
chatglm_path, trust_remote_code=True, device_map='auto')
with open(path, "r") as f:
for line in tqdm(f.readlines()):
example = json.loads(line)
# feature = preprocess(tokenizer, config, example, max_seq_length)
feature = preprocess(tokenizer, config, example, max_seq_length, version)
if skip_overlength and len(feature["input_ids"]) > max_seq_length:
continue
# feature["input_ids"] = feature["input_ids"][:max_seq_length]
yield feature | null |
179,818 | import argparse
import json
from tqdm import tqdm
def format_example(example: dict) -> dict:
context = f"Instruction: {example['instruction']}\n"
if example.get("input"):
context += f"Input: {example['input']}\n"
context += "Answer: "
target = example["output"]
return {"context": context, "target": target} | null |
179,819 | import os
from collections import defaultdict
from datetime import datetime, timedelta
from typing import Optional
from github import Github
from github.Issue import Issue
from github.Repository import Repository
from pytz import timezone
import typer
from typer import Typer
class IssueData:
def __init__(self, issue: Issue) -> None:
def get_label_to_issues(
github: Github, repository: Repository
) -> defaultdict[str, list[Issue]]:
def get_label_to_issue_data(
label_to_issues: defaultdict[str, list[Issue]]
) -> dict[str, list[IssueData]]:
def get_error_message_to_erroneous_issues(
github: Github, repository: Repository
) -> defaultdict[str, list[Issue]]:
def get_error_message_to_erroneous_issue_data(
error_message_to_erroneous_issues: defaultdict[str, list[Issue]],
) -> dict[str, list[IssueData]]:
def get_issue_maps(
github: Github, repository: Repository
) -> tuple[dict[str, list[IssueData]], dict[str, list[IssueData]]]:
label_to_issues: defaultdict[str, list[Issue]] = get_label_to_issues(
github, repository
)
label_to_issue_data: dict[str, list[IssueData]] = get_label_to_issue_data(
label_to_issues
)
error_message_to_erroneous_issues: defaultdict[
str, list[Issue]
] = get_error_message_to_erroneous_issues(github, repository)
error_message_to_erroneous_issue_data: dict[
str, list[IssueData]
] = get_error_message_to_erroneous_issue_data(error_message_to_erroneous_issues)
# Create a new dictionary with labels ordered by the summation the of likes on the associated issues
labels = list(label_to_issue_data.keys())
labels.sort(
key=lambda label: sum(
issue_data.like_count for issue_data in label_to_issue_data[label]
),
reverse=True,
)
label_to_issue_data = {label: label_to_issue_data[label] for label in labels}
return (
label_to_issue_data,
error_message_to_erroneous_issue_data,
) | null |
179,820 | import os
from collections import defaultdict
from datetime import datetime, timedelta
from typing import Optional
from github import Github
from github.Issue import Issue
from github.Repository import Repository
from pytz import timezone
import typer
from typer import Typer
DATETIME_FORMAT: str = "%m/%d/%Y %I:%M %p"
CORE_LABELS: set[str] = set(
[
"defect",
"design",
"documentation",
"enhancement",
"panic / crash",
]
)
IGNORED_LABELS: set[str] = set(
[
"meta",
"linux",
"web",
"windows",
]
)
class IssueData:
def __init__(self, issue: Issue) -> None:
self.url: str = issue.html_url
self.like_count: int = issue._rawData["reactions"]["+1"] # type: ignore [attr-defined]
self.creation_datetime: str = issue.created_at.strftime(DATETIME_FORMAT)
# TODO: Change script to support storing labels here, rather than directly in the script
self.labels: set[str] = set(label["name"] for label in issue._rawData["labels"]) # type: ignore [attr-defined]
def get_highest_ranking_issues_lines(
label_to_issue_data: dict[str, list[IssueData]],
) -> list[str]:
highest_ranking_issues_lines: list[str] = []
if label_to_issue_data:
for label, issue_data in label_to_issue_data.items():
highest_ranking_issues_lines.append(f"\n## {label}\n")
for i, issue_data in enumerate(issue_data):
markdown_bullet_point: str = (
f"{issue_data.url} ({issue_data.like_count} :thumbsup:)"
)
markdown_bullet_point = f"{i + 1}. {markdown_bullet_point}"
highest_ranking_issues_lines.append(markdown_bullet_point)
return highest_ranking_issues_lines
def get_erroneous_issues_lines(
error_message_to_erroneous_issue_data,
) -> list[str]:
erroneous_issues_lines: list[str] = []
if error_message_to_erroneous_issue_data:
for (
error_message,
erroneous_issue_data,
) in error_message_to_erroneous_issue_data.items():
erroneous_issues_lines.append(f"\n#### {error_message}\n")
for errorneous_issue_data in erroneous_issue_data:
erroneous_issues_lines.append(f"- {errorneous_issue_data.url}")
return erroneous_issues_lines
def get_issue_text(
label_to_issue_data: dict[str, list[IssueData]],
error_message_to_erroneous_issue_data: dict[str, list[IssueData]],
) -> str:
tz = timezone("america/new_york")
current_datetime: str = datetime.now(tz).strftime(f"{DATETIME_FORMAT} (%Z)")
highest_ranking_issues_lines: list[str] = get_highest_ranking_issues_lines(
label_to_issue_data
)
issue_text_lines: list[str] = [
f"*Updated on {current_datetime}*",
*highest_ranking_issues_lines,
"",
"---\n",
]
erroneous_issues_lines: list[str] = get_erroneous_issues_lines(
error_message_to_erroneous_issue_data
)
if erroneous_issues_lines:
core_labels_text: str = ", ".join(
f'"{core_label}"' for core_label in CORE_LABELS
)
ignored_labels_text: str = ", ".join(
f'"{ignored_label}"' for ignored_label in IGNORED_LABELS
)
issue_text_lines.extend(
[
"## errors with issues (this section only shows when there are errors with issues)\n",
f"This script expects every issue to have at least one of the following core labels: {core_labels_text}",
f"This script currently ignores issues that have one of the following labels: {ignored_labels_text}\n",
"### what to do?\n",
"- Adjust the core labels on an issue to put it into a correct state or add a currently-ignored label to the issue",
"- Adjust the core and ignored labels registered in this script",
*erroneous_issues_lines,
"",
"---\n",
]
)
issue_text_lines.extend(
[
"*For details on how this issue is generated, [see the script](https://github.com/zed-industries/community/blob/main/scripts/update_top_ranking_issues/main.py)*",
]
)
return "\n".join(issue_text_lines) | null |
179,821 | import os
import re
from dataclasses import dataclass
from typing import Callable, Generic, Iterable, TypeVar
The provided code snippet includes necessary dependencies for implementing the `iter_folder` function. Write a Python function `def iter_folder(folder, is_dir=False, ext=None)` to solve the following problem:
Args: folder (str): is_dir (bool): True to iter directories only ext (str): File extension, such as `.yaml` Yields: str: Absolute path of files
Here is the function:
def iter_folder(folder, is_dir=False, ext=None):
"""
Args:
folder (str):
is_dir (bool): True to iter directories only
ext (str): File extension, such as `.yaml`
Yields:
str: Absolute path of files
"""
for file in os.listdir(folder):
sub = os.path.join(folder, file)
if is_dir:
if os.path.isdir(sub):
yield sub.replace('\\\\', '/').replace('\\', '/')
elif ext is not None:
if not os.path.isdir(sub):
_, extension = os.path.splitext(file)
if extension == ext:
yield os.path.join(folder, file).replace('\\\\', '/').replace('\\', '/')
else:
yield os.path.join(folder, file).replace('\\\\', '/').replace('\\', '/') | Args: folder (str): is_dir (bool): True to iter directories only ext (str): File extension, such as `.yaml` Yields: str: Absolute path of files |
179,822 | import os
import re
from dataclasses import dataclass
from typing import Callable, Generic, Iterable, TypeVar
class DataProcessInfo:
proc: object # psutil.Process or psutil._pswindows.Process
pid: int
def name(self):
name = self.proc.name()
return name
def cmdline(self):
try:
cmdline = self.proc.cmdline()
except:
# psutil.AccessDenied
cmdline = []
cmdline = ' '.join(cmdline).replace(r'\\', '/').replace('\\', '/')
return cmdline
def __str__(self):
# Don't print `proc`, it will take some time to get process properties
return f'DataProcessInfo(name="{self.name}", pid={self.pid}, cmdline="{self.cmdline}")'
__repr__ = __str__
def iter_process() -> Iterable[DataProcessInfo]:
try:
import psutil
except ModuleNotFoundError:
return
if psutil.WINDOWS:
# Since this is a one-time-usage, we access psutil._psplatform.Process directly
# to bypass the call of psutil.Process.is_running().
# This only costs about 0.017s.
for pid in psutil.pids():
proc = psutil._psplatform.Process(pid)
yield DataProcessInfo(
proc=proc,
pid=proc.pid,
)
else:
# This will cost about 0.45s, even `attr` is given.
for proc in psutil.process_iter():
yield DataProcessInfo(
proc=proc,
pid=proc.pid,
) | null |
179,823 | import os
import re
from deploy.Windows.logger import logger
def patch_trust_env(file):
"""
People use proxies, but they never realize that proxy software leaves a
global proxy pointing to itself even when the software is not running.
In most situations we set `session.trust_env = False` in requests, but this
does not effect the `pip` command.
To handle untrusted user environment for good. We patch the code file in
requests directly. Of course, the patch only effect the python env inside
Alas.
Returns:
bool: If patched.
"""
try:
with open(file, 'r', encoding='utf-8') as f:
content = f.read()
except FileNotFoundError:
logger.info(f'{file} trust_env not exist')
return
if re.search('self.trust_env = True', content):
content = re.sub('self.trust_env = True', 'self.trust_env = False', content)
with open(file, 'w', encoding='utf-8') as f:
f.write(content)
logger.info(f'{file} trust_env patched')
elif re.search('self.trust_env = False', content):
logger.info(f'{file} trust_env already patched')
else:
logger.info(f'{file} trust_env is not in the file')
def check_running_directory():
"""
An fool-proof mechanism.
Show error if user is running Easy Install in compressing software,
since Alas can't install in temp directories.
"""
file = __file__.replace(r"\\", "/").replace("\\", "/")
# C:/Users/<user>/AppData/Local/Temp/360zip$temp/360$3/AzurLaneAutoScript
if 'Temp/360zip' in file:
logger.critical('请先解压Alas的压缩包，再安装Alas')
exit(1)
# C:/Users/<user>/AppData/Local/Temp/Rar$EXa9248.23428/AzurLaneAutoScript
if 'Temp/Rar' in file or 'Local/Temp' in file:
logger.critical('Please unzip ALAS installer first')
exit(1)
def patch_uiautomator2():
"""
uiautomator2 download assets from https://tool.appetizer.io first then fallback to https://github.com/openatx.
https://tool.appetizer.io is added to bypass the wall in China but https://tool.appetizer.io is slow outside of CN
plus some CN users cannot access it for unknown reason.
So we patch `uiautomator2/init.py` to a local assets cache `uiautomator2cache/cache`.
appdir = os.path.join(os.path.expanduser('~'), '.uiautomator2')
to:
appdir = os.path.join(__file__, '../../uiautomator2cache')
And we also remove minicap installations since emulators doesn't need it.
for url in self.minicap_urls:
self.push_url(url)
to:
for url in []:
self.push_url(url)
"""
init_file = './toolkit/Lib/site-packages/uiautomator2/init.py'
cache_dir = './toolkit/Lib/site-packages/uiautomator2cache/cache'
appdir = "os.path.join(__file__, '../../uiautomator2cache')"
modified = False
try:
with open(init_file, 'r', encoding='utf-8') as f:
content = f.read()
except FileNotFoundError:
logger.info(f'{init_file} not exist')
return
# Patch minicap_urls
res = re.search(r'self.minicap_urls', content)
if res:
content = re.sub(r'self.minicap_urls', '[]', content)
modified = True
logger.info(f'{init_file} minicap_urls patched')
else:
logger.info(f'{init_file} minicap_urls no need to patch')
# Patch appdir
if os.path.exists(cache_dir):
res = re.search(r'appdir ?=(.*)\n', content)
if res:
prev = res.group(1).strip()
if prev == appdir:
logger.info(f'{init_file} appdir already patched')
else:
content = re.sub(r'appdir ?=.*\n', f'appdir = {appdir}\n', content)
modified = True
logger.info(f'{init_file} appdir patched')
else:
logger.info(f'{init_file} appdir not found')
else:
logger.info('uiautomator2cache is not installed skip patching')
# Save file
if modified:
with open(init_file, 'w', encoding='utf-8') as f:
f.write(content)
logger.info(f'{init_file} content saved')
def patch_apkutils2():
"""
`adbutils/mixin.py` `ShellMixin.install` imports `apkutils2`, but `apkutils2` does not provide wheel files,
it may failed to install for unknown reasons. Since we never used that method, we just remove the import.
"""
mixin = './toolkit/Lib/site-packages/adbutils/mixin.py'
try:
with open(mixin, 'r', encoding='utf-8') as f:
content = f.read()
except FileNotFoundError:
logger.info(f'{mixin} not exist')
return
res = re.search(r'import apkutils2', content)
if res:
content = re.sub(r'import apkutils2', '', content)
with open(mixin, 'w', encoding='utf-8') as f:
f.write(content)
logger.info(f'{mixin} apkutils2 patched')
else:
logger.info(f'{mixin} apkutils2 no need to patch')
def pre_checks():
check_running_directory()
# patch_trust_env
patch_trust_env('./toolkit/Lib/site-packages/requests/sessions.py')
patch_trust_env('./toolkit/Lib/site-packages/pip/_vendor/requests/sessions.py')
patch_uiautomator2()
patch_apkutils2() | null |
179,824 | import logging
import os
from deploy.Windows.emulator import EmulatorManager
from deploy.Windows.logger import Progress, logger
logger = logging.getLogger("deploy")
logger.addHandler(hdlr)
logger.setLevel(logging.INFO)
logger.hr = hr
logger.attr = attr
def show_fix_tip(module):
logger.info(f"""
To fix this:
1. Open console.bat
2. Execute the following commands:
pip uninstall -y {module}
pip install --no-cache-dir {module}
3. Re-open Alas.exe
""") | null |
179,826 | import sys
import typing as t
from deploy.Windows.utils import poor_yaml_read, poor_yaml_write, DEPLOY_TEMPLATE
def get_args() -> t.Dict[str, str]:
args = {}
for arg in sys.argv[1:]:
if '=' not in arg:
continue
k, v = arg.split('=')
k, v = k.strip(), v.strip()
args[k] = v
return args | null |
179,827 | import itertools
from pponnxcr.predict_system import BoxedResult
from module.base.utils import area_in_area, area_offset
def area_cross_area(area1, area2, thres_x=20, thres_y=20):
"""
Args:
area1: (upper_left_x, upper_left_y, bottom_right_x, bottom_right_y).
area2: (upper_left_x, upper_left_y, bottom_right_x, bottom_right_y).
thres_x:
thres_y:
Returns:
bool:
"""
# https://www.yiiven.cn/rect-is-intersection.html
xa1, ya1, xa2, ya2 = area1
xb1, yb1, xb2, yb2 = area2
return abs(xb2 + xb1 - xa2 - xa1) <= xa2 - xa1 + xb2 - xb1 + thres_x * 2 \
and abs(yb2 + yb1 - ya2 - ya1) <= ya2 - ya1 + yb2 - yb1 + thres_y * 2
def _merge_boxed_result(left: BoxedResult, right: BoxedResult) -> BoxedResult:
left.box = _merge_area(left.box, right.box)
left.ocr_text = left.ocr_text + right.ocr_text
return left
The provided code snippet includes necessary dependencies for implementing the `merge_buttons` function. Write a Python function `def merge_buttons(buttons: list[BoxedResult], thres_x=20, thres_y=20) -> list[BoxedResult]` to solve the following problem:
Args: buttons: thres_x: Merge results with horizontal box distance <= `thres_x` thres_y: Merge results with vertical box distance <= `thres_y` Returns:
Here is the function:
def merge_buttons(buttons: list[BoxedResult], thres_x=20, thres_y=20) -> list[BoxedResult]:
"""
Args:
buttons:
thres_x: Merge results with horizontal box distance <= `thres_x`
thres_y: Merge results with vertical box distance <= `thres_y`
Returns:
"""
if thres_x <= 0 and thres_y <= 0:
return buttons
dic_button = {button.box: button for button in buttons}
set_merged = set()
for left, right in itertools.combinations(dic_button.items(), 2):
left_box, left = left
right_box, right = right
if area_cross_area(left.box, right.box, thres_x=thres_x, thres_y=thres_y):
left = _merge_boxed_result(left, right)
dic_button[left_box] = left
dic_button[right_box] = left
set_merged.add(right_box)
return [button for box, button in dic_button.items() if box not in set_merged] | Args: buttons: thres_x: Merge results with horizontal box distance <= `thres_x` thres_y: Merge results with vertical box distance <= `thres_y` Returns: |
179,828 | import itertools
from pponnxcr.predict_system import BoxedResult
from module.base.utils import area_in_area, area_offset
def split_and_pair_buttons(buttons, split_func, relative_area):
"""
Pair buttons in group1 with those in group2 in the relative_area.
Args:
buttons (list[OcrResultButton]):
split_func (callable):
A function that accepts an OcrResultButton object returns a bool,
button that has a True return join group1, False join group2.
relative_area (tuple[int, int, int, int]):
Yields:
OcrResultButton, OcrResultButton:
"""
group1 = [button for button in buttons if split_func(button)]
group2 = [button for button in buttons if not split_func(button)]
for ret in pair_buttons(group1, group2, relative_area):
yield ret
The provided code snippet includes necessary dependencies for implementing the `split_and_pair_button_attr` function. Write a Python function `def split_and_pair_button_attr(buttons, split_func, relative_area)` to solve the following problem:
Pair buttons in group1 with those in group2 in the relative_area, and treat group2 as the BUTTON attribute of group1. Args: buttons (list[OcrResultButton]): split_func (callable): A function that accepts an OcrResultButton object returns a bool, button that has a True return join group1, False join group2. relative_area (tuple[int, int, int, int]): Yields: OcrResultButton:
Here is the function:
def split_and_pair_button_attr(buttons, split_func, relative_area):
"""
Pair buttons in group1 with those in group2 in the relative_area,
and treat group2 as the BUTTON attribute of group1.
Args:
buttons (list[OcrResultButton]):
split_func (callable):
A function that accepts an OcrResultButton object returns a bool,
button that has a True return join group1, False join group2.
relative_area (tuple[int, int, int, int]):
Yields:
OcrResultButton:
"""
for button1, button2 in split_and_pair_buttons(buttons, split_func, relative_area):
button1.button = button2.button
yield button1 | Pair buttons in group1 with those in group2 in the relative_area, and treat group2 as the BUTTON attribute of group1. Args: buttons (list[OcrResultButton]): split_func (callable): A function that accepts an OcrResultButton object returns a bool, button that has a True return join group1, False join group2. relative_area (tuple[int, int, int, int]): Yields: OcrResultButton: |
179,829 | from pponnxcr import TextSystem as TextSystem_
from module.base.decorator import cached_property, del_cached_property
from module.exception import ScriptError
DIC_LANG_TO_MODEL = {
'cn': 'zhs',
'en': 'en',
'jp': 'ja',
'tw': 'zht',
}
The provided code snippet includes necessary dependencies for implementing the `lang2model` function. Write a Python function `def lang2model(lang: str) -> str` to solve the following problem:
Args: lang: In-game language name, defined in VALID_LANG Returns: str: Model name, defined in pponnxcr.utility
Here is the function:
def lang2model(lang: str) -> str:
"""
Args:
lang: In-game language name, defined in VALID_LANG
Returns:
str: Model name, defined in pponnxcr.utility
"""
return DIC_LANG_TO_MODEL.get(lang, lang) | Args: lang: In-game language name, defined in VALID_LANG Returns: str: Model name, defined in pponnxcr.utility |
179,830 | from pponnxcr import TextSystem as TextSystem_
from module.base.decorator import cached_property, del_cached_property
from module.exception import ScriptError
DIC_LANG_TO_MODEL = {
'cn': 'zhs',
'en': 'en',
'jp': 'ja',
'tw': 'zht',
}
The provided code snippet includes necessary dependencies for implementing the `model2lang` function. Write a Python function `def model2lang(model: str) -> str` to solve the following problem:
Args: model: Model name, defined in pponnxcr.utility Returns: str: In-game language name, defined in VALID_LANG
Here is the function:
def model2lang(model: str) -> str:
"""
Args:
model: Model name, defined in pponnxcr.utility
Returns:
str: In-game language name, defined in VALID_LANG
"""
for k, v in DIC_LANG_TO_MODEL.items():
if model == v:
return k
return model | Args: model: Model name, defined in pponnxcr.utility Returns: str: In-game language name, defined in VALID_LANG |
179,831 | import re
from dataclasses import dataclass
from functools import cached_property
from typing import ClassVar
import module.config.server as server
from module.exception import ScriptError
def parse_name(n):
n = REGEX_PUNCTUATION.sub('', str(n)).lower()
return n.strip() | null |
179,832 | import time
import typing as t
import numpy as np
from rich.table import Table
from rich.text import Text
from module.base.utils import float2str as float2str_
from module.base.utils import random_rectangle_point
from module.daemon.daemon_base import DaemonBase
from module.exception import RequestHumanTakeover
from module.logger import logger
def float2str(n, decimal=3):
if not isinstance(n, (float, int)):
return str(n)
else:
return float2str_(n, decimal=decimal) + 's' | null |
179,833 | import os
from typing import Callable, Generic, TypeVar
The provided code snippet includes necessary dependencies for implementing the `iter_folder` function. Write a Python function `def iter_folder(folder, is_dir=False, ext=None)` to solve the following problem:
Args: folder (str): is_dir (bool): True to iter directories only ext (str): File extension, such as `.yaml` Yields: str: Absolute path of files
Here is the function:
def iter_folder(folder, is_dir=False, ext=None):
"""
Args:
folder (str):
is_dir (bool): True to iter directories only
ext (str): File extension, such as `.yaml`
Yields:
str: Absolute path of files
"""
try:
files = os.listdir(folder)
except FileNotFoundError:
return
for file in files:
sub = os.path.join(folder, file)
if is_dir:
if os.path.isdir(sub):
yield sub.replace('\\\\', '/').replace('\\', '/')
elif ext is not None:
if not os.path.isdir(sub):
_, extension = os.path.splitext(file)
if extension == ext:
yield os.path.join(folder, file).replace('\\\\', '/').replace('\\', '/')
else:
yield os.path.join(folder, file).replace('\\\\', '/').replace('\\', '/') | Args: folder (str): is_dir (bool): True to iter directories only ext (str): File extension, such as `.yaml` Yields: str: Absolute path of files |
179,834 | import codecs
import os
import re
import typing as t
import winreg
from dataclasses import dataclass
from module.device.platform.utils import cached_property, iter_folder
from module.device.platform.emulator_base import EmulatorBase, EmulatorInstanceBase, EmulatorManagerBase
class RegValue:
name: str
value: str
typ: int
The provided code snippet includes necessary dependencies for implementing the `list_reg` function. Write a Python function `def list_reg(reg) -> t.List[RegValue]` to solve the following problem:
List all values in a reg key
Here is the function:
def list_reg(reg) -> t.List[RegValue]:
"""
List all values in a reg key
"""
rows = []
index = 0
try:
while 1:
value = RegValue(*winreg.EnumValue(reg, index))
index += 1
rows.append(value)
except OSError:
pass
return rows | List all values in a reg key |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.