markdown
stringlengths
0
1.02M
code
stringlengths
0
832k
output
stringlengths
0
1.02M
license
stringlengths
3
36
path
stringlengths
6
265
repo_name
stringlengths
6
127
Lab Task 1. In the cell below we define the pipeline for training and deploying our taxifare model. Fill in the code to accomplish four things:1. define the approrpriate `worker_pool_spec` for the training job1. use `ModelUploadOp` to upload the model artifacts after training to create the model in Vertex AI1. create ...
@kfp.dsl.pipeline(name="taxifare--train-upload-endpoint-deploy") def pipeline( project: str = PROJECT, model_display_name: str = MODEL_DISPLAY_NAME, ): train_task = training_op("taxifare training pipeline") experimental.run_as_aiplatform_custom_job( train_task, display_name=f"pipelines-t...
_____no_output_____
Apache-2.0
notebooks/building_production_ml_systems/labs/3_kubeflow_pipelines_vertex.ipynb
paras301/asl-ml-immersion
Compile and run the pipelineNow, you're ready to compile the pipeline:
if not os.path.isdir("vertex_pipelines"): os.mkdir("vertex_pipelines") compiler.Compiler().compile( pipeline_func=pipeline, package_path="./vertex_pipelines/train_upload_endpoint_deploy.json", )
_____no_output_____
Apache-2.0
notebooks/building_production_ml_systems/labs/3_kubeflow_pipelines_vertex.ipynb
paras301/asl-ml-immersion
The pipeline compilation generates the `train_upload_endpoint_deploy.json` job spec file.Next, instantiate the pipeline job object: Lab Task 2.Complete the code in the cell below to fill in the missing arguments.
pipeline_job = aiplatform.pipeline_jobs.PipelineJob( display_name= # TODO: Your code goes here. template_path= # TODO: Your code goes here. pipeline_root= # TODO: Your code goes here. project=PROJECT, location=REGION, )
_____no_output_____
Apache-2.0
notebooks/building_production_ml_systems/labs/3_kubeflow_pipelines_vertex.ipynb
paras301/asl-ml-immersion
Then, you run the defined pipeline like this:
pipeline_job.run()
_____no_output_____
Apache-2.0
notebooks/building_production_ml_systems/labs/3_kubeflow_pipelines_vertex.ipynb
paras301/asl-ml-immersion
Project 3: Smart Beta Portfolio and Portfolio Optimization OverviewSmart beta has a broad meaning, but we can say in practice that when we use the universe of stocks from an index, and then apply some weighting scheme other than market cap weighting, it can be considered a type of smart beta fund. A Smart Beta portfo...
import sys !{sys.executable} -m pip install -r requirements.txt
Requirement already satisfied: colour==0.1.5 in /opt/conda/lib/python3.6/site-packages (from -r requirements.txt (line 1)) (0.1.5) Requirement already satisfied: cvxpy==1.0.3 in /opt/conda/lib/python3.6/site-packages (from -r requirements.txt (line 2)) (1.0.3) Requirement already satisfied: cycler==0.10.0 in /opt/conda...
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
Load Packages
import pandas as pd import numpy as np import helper import project_helper import project_tests
_____no_output_____
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
Market Data Load DataFor this universe of stocks, we'll be selecting large dollar volume stocks. We're using this universe, since it is highly liquid.
df = pd.read_csv('../../data/project_3/eod-quotemedia.csv') percent_top_dollar = 0.2 high_volume_symbols = project_helper.large_dollar_volume_stocks(df, 'adj_close', 'adj_volume', percent_top_dollar) df = df[df['ticker'].isin(high_volume_symbols)] close = df.reset_index().pivot(index='date', columns='ticker', values=...
_____no_output_____
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
View DataTo see what one of these 2-d matrices looks like, let's take a look at the closing prices matrix.
project_helper.print_dataframe(close)
_____no_output_____
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
Part 1: Smart Beta PortfolioIn Part 1 of this project, you'll build a portfolio using dividend yield to choose the portfolio weights. A portfolio such as this could be incorporated into a smart beta ETF. You'll compare this portfolio to a market cap weighted index to see how well it performs. Note that in practice, y...
def generate_dollar_volume_weights(close, volume): """ Generate dollar volume weights. Parameters ---------- close : DataFrame Close price for each ticker and date volume : str Volume for each ticker and date Returns ------- dollar_volume_weights : DataFrame ...
Tests Passed
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
View DataLet's generate the index weights using `generate_dollar_volume_weights` and view them using a heatmap.
index_weights = generate_dollar_volume_weights(close, volume) project_helper.plot_weights(index_weights, 'Index Weights')
_____no_output_____
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
Portfolio WeightsNow that we have the index weights, let's choose the portfolio weights based on dividend. You would normally calculate the weights based on trailing dividend yield, but we'll simplify this by just calculating the total dividend yield over time.Implement `calculate_dividend_weights` to return the weigh...
def calculate_dividend_weights(dividends): """ Calculate dividend weights. Parameters ---------- dividends : DataFrame Dividend for each stock and date Returns ------- dividend_weights : DataFrame Weights for each stock and date """ #TODO: Implement function ...
Tests Passed
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
View DataJust like the index weights, let's generate the ETF weights and view them using a heatmap.
etf_weights = calculate_dividend_weights(dividends) project_helper.plot_weights(etf_weights, 'ETF Weights')
_____no_output_____
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
ReturnsImplement `generate_returns` to generate returns data for all the stocks and dates from price data. You might notice we're implementing returns and not log returns. Since we're not dealing with volatility, we don't have to use log returns.
def generate_returns(prices): """ Generate returns for ticker and date. Parameters ---------- prices : DataFrame Price for each ticker and date Returns ------- returns : Dataframe The returns for each ticker and date """ #TODO: Implement function return pri...
Tests Passed
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
View DataLet's generate the closing returns using `generate_returns` and view them using a heatmap.
returns = generate_returns(close) project_helper.plot_returns(returns, 'Close Returns')
_____no_output_____
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
Weighted ReturnsWith the returns of each stock computed, we can use it to compute the returns for an index or ETF. Implement `generate_weighted_returns` to create weighted returns using the returns and weights.
def generate_weighted_returns(returns, weights): """ Generate weighted returns. Parameters ---------- returns : DataFrame Returns for each ticker and date weights : DataFrame Weights for each ticker and date Returns ------- weighted_returns : DataFrame Weigh...
Tests Passed
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
View DataLet's generate the ETF and index returns using `generate_weighted_returns` and view them using a heatmap.
index_weighted_returns = generate_weighted_returns(returns, index_weights) etf_weighted_returns = generate_weighted_returns(returns, etf_weights) project_helper.plot_returns(index_weighted_returns, 'Index Returns') project_helper.plot_returns(etf_weighted_returns, 'ETF Returns')
_____no_output_____
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
Cumulative ReturnsTo compare performance between the ETF and Index, we're going to calculate the tracking error. Before we do that, we first need to calculate the index and ETF comulative returns. Implement `calculate_cumulative_returns` to calculate the cumulative returns over time given the returns.
def calculate_cumulative_returns(returns): """ Calculate cumulative returns. Parameters ---------- returns : DataFrame Returns for each ticker and date Returns ------- cumulative_returns : Pandas Series Cumulative returns for each date """ #TODO: Implement funct...
Tests Passed
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
View DataLet's generate the ETF and index cumulative returns using `calculate_cumulative_returns` and compare the two.
index_weighted_cumulative_returns = calculate_cumulative_returns(index_weighted_returns) etf_weighted_cumulative_returns = calculate_cumulative_returns(etf_weighted_returns) project_helper.plot_benchmark_returns(index_weighted_cumulative_returns, etf_weighted_cumulative_returns, 'Smart Beta ETF vs Index')
_____no_output_____
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
Tracking ErrorIn order to check the performance of the smart beta portfolio, we can calculate the annualized tracking error against the index. Implement `tracking_error` to return the tracking error between the ETF and benchmark.For reference, we'll be using the following annualized tracking error function:$$ TE = \sq...
def tracking_error(benchmark_returns_by_date, etf_returns_by_date): """ Calculate the tracking error. Parameters ---------- benchmark_returns_by_date : Pandas Series The benchmark returns for each date etf_returns_by_date : Pandas Series The ETF returns for each date Return...
Tests Passed
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
View DataLet's generate the tracking error using `tracking_error`.
smart_beta_tracking_error = tracking_error(np.sum(index_weighted_returns, 1), np.sum(etf_weighted_returns, 1)) print('Smart Beta Tracking Error: {}'.format(smart_beta_tracking_error))
Smart Beta Tracking Error: 0.10207614832007529
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
Part 2: Portfolio OptimizationNow, let's create a second portfolio. We'll still reuse the market cap weighted index, but this will be independent of the dividend-weighted portfolio that we created in part 1.We want to both minimize the portfolio variance and also want to closely track a market cap weighted index. In...
def get_covariance_returns(returns): """ Calculate covariance matrices. Parameters ---------- returns : DataFrame Returns for each ticker and date Returns ------- returns_covariance : 2 dimensional Ndarray The covariance of the returns """ #TODO: Implement func...
Tests Passed
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
View DataLet's look at the covariance generated from `get_covariance_returns`.
covariance_returns = get_covariance_returns(returns) covariance_returns = pd.DataFrame(covariance_returns, returns.columns, returns.columns) covariance_returns_correlation = np.linalg.inv(np.diag(np.sqrt(np.diag(covariance_returns)))) covariance_returns_correlation = pd.DataFrame( covariance_returns_correlation.do...
_____no_output_____
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
portfolio varianceWe can write the portfolio variance $\sigma^2_p = \mathbf{x^T} \mathbf{P} \mathbf{x}$Recall that the $\mathbf{x^T} \mathbf{P} \mathbf{x}$ is called the quadratic form.We can use the cvxpy function `quad_form(x,P)` to get the quadratic form. Distance from index weightsWe want portfolio weights that tr...
import cvxpy as cvx def get_optimal_weights(covariance_returns, index_weights, scale=2.0): """ Find the optimal weights. Parameters ---------- covariance_returns : 2 dimensional Ndarray The covariance of the returns index_weights : Pandas Series Index weights for all tickers at...
_____no_output_____
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
Optimized PortfolioUsing the `get_optimal_weights` function, let's generate the optimal ETF weights without rebalanceing. We can do this by feeding in the covariance of the entire history of data. We also need to feed in a set of index weights. We'll go with the average weights of the index over time.
raw_optimal_single_rebalance_etf_weights = get_optimal_weights(covariance_returns.values, index_weights.iloc[-1]) optimal_single_rebalance_etf_weights = pd.DataFrame( np.tile(raw_optimal_single_rebalance_etf_weights, (len(returns.index), 1)), returns.index, returns.columns)
_____no_output_____
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
With our ETF weights built, let's compare it to the index. Run the next cell to calculate the ETF returns and compare it to the index returns.
optim_etf_returns = generate_weighted_returns(returns, optimal_single_rebalance_etf_weights) optim_etf_cumulative_returns = calculate_cumulative_returns(optim_etf_returns) project_helper.plot_benchmark_returns(index_weighted_cumulative_returns, optim_etf_cumulative_returns, 'Optimized ETF vs Index') optim_etf_tracking...
_____no_output_____
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
Rebalance Portfolio Over TimeThe single optimized ETF portfolio used the same weights for the entire history. This might not be the optimal weights for the entire period. Let's rebalance the portfolio over the same period instead of using the same weights. Implement `rebalance_portfolio` to rebalance a portfolio.Rebla...
def rebalance_portfolio(returns, index_weights, shift_size, chunk_size): """ Get weights for each rebalancing of the portfolio. Parameters ---------- returns : DataFrame Returns for each ticker and date index_weights : DataFrame Index weight for each ticker and date shift_si...
Tests Passed
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
Run the following cell to get the portfolio turnover from `get_portfolio turnover`.
chunk_size = 250 shift_size = 5 all_rebalance_weights = rebalance_portfolio(returns, index_weights, shift_size, chunk_size) print(get_portfolio_turnover(all_rebalance_weights, shift_size, len(all_rebalance_weights) - 1))
16.72683266050277
MIT
Smart_Beta_and_Portfolio_Optimization.ipynb
parinp/Ai-for-Trading
**In this notebook, we embed the abstract of the papers into a low dimensional space (using either sentencetransformers library or doc2vec from Gensim) and associate to each author his abstracts embedding**
!pip install -U sentence-transformers from tqdm import tqdm_notebook as tqdm from sentence_transformers import SentenceTransformer import pandas as pd import gzip import pickle import numpy as np import torch from gensim.models.doc2vec import Doc2Vec, TaggedDocument from string import digits, ascii_letters, punctuation...
_____no_output_____
MIT
notebook_utils/abstracts_2_vec_per_author.ipynb
omarsou/altegrad_challenge_hindex
Load Abstracts
tmp = load_dataset_file('/content/drive/MyDrive/altegrad_datachallenge/files_generated/preprocess_abstracts.txt') ## Cleaning V2 (before conditioned on word with word.isalpha() as a condition) valid = ascii_letters + digits + punctuation + printable paper_id = [] text = [] for key in tqdm(tmp.keys()): txt = ''.joi...
_____no_output_____
MIT
notebook_utils/abstracts_2_vec_per_author.ipynb
omarsou/altegrad_challenge_hindex
Abstract Embedding STSB Roberta Base
model = SentenceTransformer('stsb-roberta-base') model.cuda() embeddings = model.encode(text) emb_per_paper = {} for idx, id in enumerate(paper_id): emb_per_paper[id] = embeddings[idx] save(emb_per_paper, '/content/drive/MyDrive/altegrad_datachallenge/embedding_per_paper_clean.txt')
_____no_output_____
MIT
notebook_utils/abstracts_2_vec_per_author.ipynb
omarsou/altegrad_challenge_hindex
Doc2Vec
stop_words = set(stopwords.words('english')) doc = [] for txt in tqdm(text): p = txt.split() p_clean = [l for l in p if l not in stop_words] doc.append(p_clean) del text tagged_data = [TaggedDocument(d, [i]) for i, d in enumerate(doc)] model = Doc2Vec(tagged_data, vector_size = 256, window = 5, min_count ...
_____no_output_____
MIT
notebook_utils/abstracts_2_vec_per_author.ipynb
omarsou/altegrad_challenge_hindex
Abstract Per Author EmbeddingAssociate each author with his articles
# read the file to create a dictionary with author key and paper list as value f = open("/content/drive/MyDrive/altegrad_datachallenge/author_papers.txt","r") papers_set = set() d = {} for l in f: auth_paps = [paper_id.strip() for paper_id in l.split(":")[1].replace("[","").replace("]","").replace("\n","").replace(...
_____no_output_____
MIT
notebook_utils/abstracts_2_vec_per_author.ipynb
omarsou/altegrad_challenge_hindex
Using Roberta Embedding
emb_per_paper = load_dataset_file('/content/drive/MyDrive/altegrad_datachallenge/embedding_per_paper_clean.txt') df = open("/content/drive/MyDrive/altegrad_datachallenge/author_embedding_clean.csv","w") for id_author in tqdm(d.keys()): tot_embedding = np.zeros(768) c = 0 for id_paper in d[id_author]: ...
_____no_output_____
MIT
notebook_utils/abstracts_2_vec_per_author.ipynb
omarsou/altegrad_challenge_hindex
Using Doc2Vec
emb_per_paper = load_dataset_file('/content/drive/MyDrive/altegrad_datachallenge/doc2vec_paper_embedding.txt') df = open("/content/drive/MyDrive/altegrad_datachallenge/doc2vec_author_embedding.csv","w") for id_author in tqdm(d.keys()): tot_embedding = np.zeros(256) c = 0 for id_paper in d[id_author]: ...
_____no_output_____
MIT
notebook_utils/abstracts_2_vec_per_author.ipynb
omarsou/altegrad_challenge_hindex
Visualization using the Graphviz Library**Goal:** Visualize all (or some) of the DAG defining the LSHTC3 data.Source: https://graphviz.readthedocs.io/en/stable/examples.html Load the Data
with open("./data/hierarchyWikipediaMedium.txt", 'r') as edges: lines = [] for line in edges.readlines(): line = line.rstrip('\r\n') line = line.split(' ') lines.append(line) print(lines[0:100]) from graphviz import Graph g = Graph('G', filename='process.gv', engine='sfdp') for edge ...
_____no_output_____
MIT
DAG_viz.ipynb
djliden/LSHTC3_DAG
The aim of this is to be able to classify names as either being African in origin or not
import pandas as pd import ast from ethnicolr import pred_wiki_ln, pred_wiki_name
Using TensorFlow backend.
Apache-2.0
working_ipynbs/name_classification.ipynb
TamatiB/restitution_africa2021
Load puplications, get authors and how many publications of theirs we have
from collections import Counter data = pd.read_csv("bb_pulications.csv") data['author'] = data['bib'].apply(lambda x: ast.literal_eval(x)['author']) data['year'] = data['bib'].apply(lambda x: ast.literal_eval(x)['pub_year']) data['title'] = data['bib'].apply(lambda x: ast.literal_eval(x)['title'])
_____no_output_____
Apache-2.0
working_ipynbs/name_classification.ipynb
TamatiB/restitution_africa2021
Clean author names a little bit
def clean_author(x): """ x list of authors for a publication """ clean_list = [] for item in x: clean_list.append(item.lower()) return clean_list data['author_cleaned'] = data['author'].apply(lambda x: clean_author(x)) authors = data['author_cleaned'].sum() Counter(authors)
_____no_output_____
Apache-2.0
working_ipynbs/name_classification.ipynb
TamatiB/restitution_africa2021
Get surnames
# first have to searate from initials and then put back togetehr again surnames = [] for name in authors: surname = name.split(' ')[1:] surname_str = ' '.join(surname) surnames.append(surname_str) Counter(surnames)
_____no_output_____
Apache-2.0
working_ipynbs/name_classification.ipynb
TamatiB/restitution_africa2021
Hokay, lets try this Classifier
#drop duplicates surnames = list(set(surnames)) df = pd.DataFrame(surnames, columns=["surnames"]) preds = pred_wiki_ln(df, "surnames") preds preds['race'].value_counts() preds[preds['race'] == 'GreaterAfrican,Africans']
_____no_output_____
Apache-2.0
working_ipynbs/name_classification.ipynb
TamatiB/restitution_africa2021
Text drift detection on IMDB movie reviews MethodWe detect drift on text data using both the [Maximum Mean Discrepancy](https://docs.seldon.io/projects/alibi-detect/en/latest/methods/mmddrift.html) and [Kolmogorov-Smirnov (K-S)](https://docs.seldon.io/projects/alibi-detect/en/latest/methods/ksdrift.html) detectors. In...
import nlp import numpy as np import os import tensorflow as tf from transformers import AutoTokenizer from alibi_detect.cd import KSDrift, MMDDrift from alibi_detect.utils.saving import save_detector, load_detector
_____no_output_____
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Load tokenizer
model_name = 'bert-base-cased' tokenizer = AutoTokenizer.from_pretrained(model_name)
_____no_output_____
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Load data
def load_dataset(dataset: str, split: str = 'test'): data = nlp.load_dataset(dataset) X, y = [], [] for x in data[split]: X.append(x['text']) y.append(x['label']) X = np.array(X) y = np.array(y) return X, y X, y = load_dataset('imdb', split='train') print(X.shape, y.shape)
INFO:nlp.load:Checking /home/avl/.cache/huggingface/datasets/d3b7716978cb901261e59327d43b04c52d6d29e50eeac39bea0816865a584081.7c39fd6270c5ee55bcf2e4de23af77ef299e0df65be3f3e84454dcef7175844a.py for additional imports. INFO:filelock:Lock 140070637965264 acquired on /home/avl/.cache/huggingface/datasets/d3b7716978cb90126...
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Let's take a look at respectively a negative and positive review:
labels = ['Negative', 'Positive'] print(labels[y[-1]]) print(X[-1]) print(labels[y[2]]) print(X[2])
Positive Brilliant over-acting by Lesley Ann Warren. Best dramatic hobo lady I have ever seen, and love scenes in clothes warehouse are second to none. The corn on face is a classic, as good as anything in Blazing Saddles. The take on lawyers is also superb. After being accused of being a turncoat, selling out his boss...
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
We split the original test set in a reference dataset and a dataset which should not be rejected under the *H0* of the statistical test. We also create imbalanced datasets and inject selected words in the reference set.
def random_sample(X: np.ndarray, y: np.ndarray, proba_zero: float, n: int): if len(y.shape) == 1: idx_0 = np.where(y == 0)[0] idx_1 = np.where(y == 1)[0] else: idx_0 = np.where(y[:, 0] == 1)[0] idx_1 = np.where(y[:, 1] == 1)[0] n_0, n_1 = int(n * proba_zero), int(n * (1 - pro...
_____no_output_____
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Reference, *H0* and imbalanced data:
# proba_zero = fraction with label 0 (=negative sentiment) n_sample = 1000 X_ref = random_sample(X, y, proba_zero=.5, n=n_sample)[0] X_h0 = random_sample(X, y, proba_zero=.5, n=n_sample)[0] n_imb = [.1, .9] X_imb = {_: random_sample(X, y, proba_zero=_, n=n_sample)[0] for _ in n_imb}
_____no_output_____
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Inject words in reference data:
words = ['fantastic', 'good', 'bad', 'horrible'] perc_chg = [1., 5.] # % of tokens to change in an instance words_tf = tokenizer(words)['input_ids'] words_tf = [token[1:-1][0] for token in words_tf] max_len = 100 tokens = tokenizer(list(X_ref), pad_to_max_length=True, max_length=max_len, return_te...
Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precise...
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
PreprocessingFirst we need to specify the type of embedding we want to extract from the BERT model. We can extract embeddings from the ...- **pooler_output**: Last layer hidden-state of the first token of the sequence (classification token; CLS) further processed by a Linear layer and a Tanh activation function. The L...
from alibi_detect.models.tensorflow import TransformerEmbedding emb_type = 'hidden_state' n_layers = 8 layers = [-_ for _ in range(1, n_layers + 1)] embedding = TransformerEmbedding(model_name, emb_type, layers)
Some layers from the model checkpoint at bert-base-cased were not used when initializing TFBertModel: ['nsp___cls', 'mlm___cls'] - This IS expected if you are initializing TFBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification m...
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Let's check what an embedding looks like:
tokens = tokenizer(list(X[:5]), pad_to_max_length=True, max_length=max_len, return_tensors='tf') x_emb = embedding(tokens) print(x_emb.shape)
(5, 768)
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
So the BERT model's embedding space used by the drift detector consists of a $768$-dimensional vector for each instance. We will therefore first apply a dimensionality reduction step with an Untrained AutoEncoder (*UAE*) before conducting the statistical hypothesis test. We use the embedding model as the input for the ...
tf.random.set_seed(0) from alibi_detect.cd.tensorflow import UAE enc_dim = 32 shape = (x_emb.shape[1],) uae = UAE(input_layer=embedding, shape=shape, enc_dim=enc_dim)
_____no_output_____
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Let's test this again:
emb_uae = uae(tokens) print(emb_uae.shape)
(5, 32)
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
K-S detector InitializeWe proceed to initialize the drift detector. From here on the detector works the same as for other modalities such as images. Please check the [images](https://docs.seldon.io/projects/alibi-detect/en/latest/examples/cd_ks_cifar10.html) example or the [K-S detector documentation](https://docs.sel...
from functools import partial from alibi_detect.cd.tensorflow import preprocess_drift # define preprocessing function preprocess_fn = partial(preprocess_drift, model=uae, tokenizer=tokenizer, max_len=max_len, batch_size=32) # initialize detector cd = KSDrift(X_ref, p_val=.05, preprocess_fn=pr...
WARNING:alibi_detect.utils.saving:Directory my_path does not exist and is now created.
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Detect driftLet’s first check if drift occurs on a similar sample from the training set as the reference data.
preds_h0 = cd.predict(X_h0) labels = ['No!', 'Yes!'] print('Drift? {}'.format(labels[preds_h0['data']['is_drift']])) print('p-value: {}'.format(preds_h0['data']['p_val']))
Truncation was not explicitly activated but `max_length` is provided a specific value, please use `truncation=True` to explicitly truncate examples to max length. Defaulting to 'longest_first' truncation strategy. If you encode pairs of sequences (GLUE-style) with the tokenizer you can select this strategy more precise...
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Detect drift on imbalanced and perturbed datasets:
for k, v in X_imb.items(): preds = cd.predict(v) print('% negative sentiment {}'.format(k * 100)) print('Drift? {}'.format(labels[preds['data']['is_drift']])) print('p-value: {}'.format(preds['data']['p_val'])) print('') for w, probas in X_word.items(): for p, v in probas.items(): preds ...
Word: fantastic -- % perturbed: 1.0 Drift? No! p-value: [0.9540582 0.01293455 0.26338065 0.722555 0.34099194 0.04281518 0.04841881 0.31356168 0.14833806 0.96887016 0.85929435 0.50035924 0.00532228 0.8879386 0.9998709 0.99870795 0.85929435 0.9882611 0.06155144 0.7590978 0.79439443 0.2406036 0.10828251 0.722555...
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
MMD TensorFlow detector InitializeAgain check the [images](https://docs.seldon.io/projects/alibi-detect/en/latest/examples/cd_mmd_cifar10.html) example or the [MMD detector documentation](https://docs.seldon.io/projects/alibi-detect/en/latest/methods/mmddrift.html) for more information about each of the possible param...
cd = MMDDrift(X_ref, p_val=.05, preprocess_fn=preprocess_fn, n_permutations=100, input_shape=(max_len,))
_____no_output_____
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Detect drift*H0*:
preds_h0 = cd.predict(X_h0) labels = ['No!', 'Yes!'] print('Drift? {}'.format(labels[preds_h0['data']['is_drift']])) print('p-value: {}'.format(preds_h0['data']['p_val']))
Drift? No! p-value: 0.9
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Imbalanced data:
for k, v in X_imb.items(): preds = cd.predict(v) print('% negative sentiment {}'.format(k * 100)) print('Drift? {}'.format(labels[preds['data']['is_drift']])) print('p-value: {}'.format(preds['data']['p_val'])) print('')
% negative sentiment 10.0 Drift? Yes! p-value: 0.0 % negative sentiment 90.0 Drift? Yes! p-value: 0.0
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Perturbed data:
for w, probas in X_word.items(): for p, v in probas.items(): preds = cd.predict(v) print('Word: {} -- % perturbed: {}'.format(w, p)) print('Drift? {}'.format(labels[preds['data']['is_drift']])) print('p-value: {}'.format(preds['data']['p_val'])) print('')
Word: fantastic -- % perturbed: 1.0 Drift? Yes! p-value: 0.01 Word: fantastic -- % perturbed: 5.0 Drift? Yes! p-value: 0.0 Word: good -- % perturbed: 1.0 Drift? No! p-value: 0.57 Word: good -- % perturbed: 5.0 Drift? Yes! p-value: 0.0 Word: bad -- % perturbed: 1.0 Drift? No! p-value: 0.4 Word: bad -- % perturbed: ...
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
MMD PyTorch detector InitializeWe can run the same detector with *PyTorch* backend for both the preprocessing step and MMD implementation:
import torch import torch.nn as nn # set random seed and device seed = 0 torch.manual_seed(seed) torch.cuda.manual_seed(seed) device = torch.device('cuda' if torch.cuda.is_available() else 'cpu') print(device) from alibi_detect.cd.pytorch import preprocess_drift from alibi_detect.models.pytorch import TransformerEmbe...
INFO:filelock:Lock 140068554309968 acquired on /home/avl/.cache/huggingface/transformers/092cc582560fc3833e556b3f833695c26343cb54b7e88cd02d40821462a74999.1f48cab6c959fc6c360d22bea39d06959e90f5b002e77e836d2da45464875cda.lock
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Detect drift*H0*:
preds_h0 = cd.predict(X_h0) labels = ['No!', 'Yes!'] print('Drift? {}'.format(labels[preds_h0['data']['is_drift']])) print('p-value: {}'.format(preds_h0['data']['p_val']))
Drift? No! p-value: 0.3400000035762787
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Imbalanced data:
for k, v in X_imb.items(): preds = cd.predict(v) print('% negative sentiment {}'.format(k * 100)) print('Drift? {}'.format(labels[preds['data']['is_drift']])) print('p-value: {}'.format(preds['data']['p_val'])) print('')
% negative sentiment 10.0 Drift? Yes! p-value: 0.0 % negative sentiment 90.0 Drift? Yes! p-value: 0.0
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Perturbed data:
for w, probas in X_word.items(): for p, v in probas.items(): preds = cd.predict(v) print('Word: {} -- % perturbed: {}'.format(w, p)) print('Drift? {}'.format(labels[preds['data']['is_drift']])) print('p-value: {}'.format(preds['data']['p_val'])) print('')
Word: fantastic -- % perturbed: 1.0 Drift? No! p-value: 0.07999999821186066 Word: fantastic -- % perturbed: 5.0 Drift? Yes! p-value: 0.0 Word: good -- % perturbed: 1.0 Drift? No! p-value: 0.7099999785423279 Word: good -- % perturbed: 5.0 Drift? Yes! p-value: 0.0 Word: bad -- % perturbed: 1.0 Drift? No! p-value: 0.1...
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Train embeddings from scratchSo far we used pre-trained embeddings from a BERT model. We can however also use embeddings from a model trained from scratch. First we define and train a simple classification model consisting of an embedding and LSTM layer in *TensorFlow*. Load data and train model
from tensorflow.keras.datasets import imdb, reuters from tensorflow.keras.layers import Dense, Embedding, Input, LSTM from tensorflow.keras.preprocessing import sequence from tensorflow.keras.utils import to_categorical INDEX_FROM = 3 NUM_WORDS = 10000 def print_sentence(tokenized_sentence: str, id2w: dict): pri...
_____no_output_____
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Load and tokenize data:
(X_train, y_train), (X_test, y_test), (word2token, token2word) = \ get_dataset(dataset='imdb', max_len=max_len)
<string>:6: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray /home/avl/anaconda3/envs/detect/lib/pytho...
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Let's check out an instance:
print_sentence(X_train[0], token2word)
cry at a film it must have been good and this definitely was also <UNK> to the two little boy's that played the <UNK> of norman and paul they were just brilliant children are often left out of the <UNK> list i think because the stars that play them all grown up are such a big profile for the whole film but these childr...
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Define and train a simple model:
model = imdb_model(X=X_train, num_words=NUM_WORDS, emb_dim=256, lstm_dim=128, output_dim=2) model.fit(X_train, y_train, batch_size=32, epochs=2, shuffle=True, validation_data=(X_test, y_test))
Epoch 1/2 782/782 [==============================] - 96s 121ms/step - loss: 0.5019 - accuracy: 0.7397 - val_loss: 0.3452 - val_accuracy: 0.8514 Epoch 2/2 782/782 [==============================] - 93s 118ms/step - loss: 0.2649 - accuracy: 0.8943 - val_loss: 0.3628 - val_accuracy: 0.8454
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Extract the embedding layer from the trained model and combine with UAE preprocessing step:
embedding = tf.keras.Model(inputs=model.inputs, outputs=model.layers[1].output) x_emb = embedding(X_train[:5]) print(x_emb.shape) tf.random.set_seed(0) shape = tuple(x_emb.shape[1:]) uae = UAE(input_layer=embedding, shape=shape, enc_dim=enc_dim)
_____no_output_____
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Again, create reference, *H0* and perturbed datasets. Also test against the *Reuters* news topic classification dataset.
X_ref, y_ref = random_sample(X_test, y_test, proba_zero=.5, n=n_sample) X_h0, y_h0 = random_sample(X_test, y_test, proba_zero=.5, n=n_sample) tokens = [word2token[w] for w in words] X_word = {} for i, t in enumerate(tokens): X_word[words[i]] = {} for p in perc_chg: X_word[words[i]][p] = inject_word(t, X...
/home/avl/anaconda3/envs/detect/lib/python3.7/site-packages/tensorflow/python/keras/datasets/reuters.py:148: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you...
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Initialize detector and detect drift
from alibi_detect.cd.tensorflow import preprocess_drift # define preprocessing function preprocess_fn = partial(preprocess_drift, model=uae, batch_size=128) # initialize detector cd = KSDrift(X_ref, p_val=.05, preprocess_fn=preprocess_fn)
_____no_output_____
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
*H0*:
preds_h0 = cd.predict(X_h0) labels = ['No!', 'Yes!'] print('Drift? {}'.format(labels[preds_h0['data']['is_drift']])) print('p-value: {}'.format(preds_h0['data']['p_val']))
Drift? No! p-value: [0.93558097 0.64755726 0.50035924 0.85929435 0.04281518 0.93558097 0.9801618 0.50035924 0.8879386 0.43243074 0.5726548 0.6852314 0.60991895 0.9134755 0.18111965 0.722555 0.5726548 0.21933001 0.5360543 0.6852314 0.85929435 0.31356168 0.9801618 0.18111965 0.34099194 0.722555 0.04841881...
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
Perturbed data:
for w, probas in X_word.items(): for p, v in probas.items(): preds = cd.predict(v) print('Word: {} -- % perturbed: {}'.format(w, p)) print('Drift? {}'.format(labels[preds['data']['is_drift']])) print('p-value: {}'.format(preds['data']['p_val'])) print('')
Word: fantastic -- % perturbed: 1.0 Drift? No! p-value: [0.9882611 0.79439443 0.9999727 0.9882611 0.7590978 0.8879386 0.996931 0.82795686 0.64755726 0.7590978 0.85929435 0.99870795 0.93558097 0.82795686 0.99365413 0.996931 0.85929435 0.8879386 0.85929435 0.9540582 0.96887016 0.9801618 0.50035924 0.9998709...
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
The detector is not as sensitive as the Transformer-based K-S drift detector. The embeddings trained from scratch only trained on a small dataset and a simple model with cross-entropy loss function for 2 epochs. The pre-trained BERT model on the other hand captures semantics of the data better.Sample from the Reuters d...
preds_ood = cd.predict(X_ood) labels = ['No!', 'Yes!'] print('Drift? {}'.format(labels[preds_ood['data']['is_drift']])) print('p-value: {}'.format(preds_ood['data']['p_val']))
Drift? Yes! p-value: [5.72654784e-01 7.26078229e-04 2.73716728e-15 3.49877549e-09 1.29345525e-02 2.24637091e-02 4.95470906e-14 1.34916729e-04 8.27956855e-01 4.00471032e-01 6.20218972e-03 1.97469308e-09 6.15514442e-02 5.06567594e-04 5.46463318e-02 7.59097815e-01 1.97830971e-07 4.56308130e-10 4.15714254e-08 4.3243074...
ECL-2.0
examples/cd_text_imdb.ipynb
cliveseldon/alibi-detect
# 1.2.1 신경망 추론 전체 그림 import numpy as np def sigmoid(x): return 1/(1+np.exp(-x)) x=np.random.randn(10,2) W1=np.random.randn(2,4) b1=np.random.randn(4) W2=np.random.randn(4,3) b2=np.random.randn(3) h=np.matmul(x,W1)+b1 a=sigmoid(h) s=np.matmul(a,W2)+b2 # 1.2.2 계층으로 클래스화 및 순전파 구현 class Sigmoid: def __init__(self)...
_____no_output_____
MIT
ch01/1-2.ipynb
jjhsnail0822/deep-learning-from-scratch-2
https://github.com/cyberFund/ethdrain Python script allowing to copy the Ethereum blockchain towards ElasticSearch, PostgreSQL and csv in an efficient way by connecting to a local RPC node
import requests import json def print_json(json_for_print): print(json.dumps(json_for_print, indent=4, sort_keys=True)) return
_____no_output_____
Apache-2.0
Ethereum API.ipynb
Snedashkovsky/Course_blockchain_data
Request function
def http_post_request(url, request): print('url: {}'.format(str(url))) print('request: {} \n'.format(str(request))) return requests.post(url, data=request, headers={"content-type": "application/json"}).json()
_____no_output_____
Apache-2.0
Ethereum API.ipynb
Snedashkovsky/Course_blockchain_data
APIEthereum JSON RPC API: https://github.com/ethereum/wiki/wiki/JSON-RPC getBlockByNumber
def make_request_getBlockByNumber(block_nb, use_hex=True): return json.dumps({ "jsonrpc": "2.0", "method": "eth_getBlockByNumber", "params": [hex(block_nb) if use_hex else block_nb, True], "id": 1 ...
url: https://mainnet.infura.io/TzMi1NSXsXK2SzUuEY9Q request: {"params": ["latest", true], "method": "eth_getBlockByNumber", "id": 1, "jsonrpc": "2.0"} { "id": 1, "jsonrpc": "2.0", "result": { "difficulty": "0xb91590c441438", "extraData": "0x6e616e6f706f6f6c2e6f7267", "gasLimit": "0...
Apache-2.0
Ethereum API.ipynb
Snedashkovsky/Course_blockchain_data
eth_getBalance
def make_request_eth_getBalance(address, block_nb = 'latest', use_hex=False): return json.dumps({ "jsonrpc": "2.0", "method": "eth_getBalance", "params": [address, hex(block_nb) if use_hex else block_nb], ...
url: https://mainnet.infura.io/TzMi1NSXsXK2SzUuEY9Q request: {"params": ["0xc8f88d1c1259060a799af77120db270cdce07e37", "0x517025"], "method": "eth_getBalance", "id": 1, "jsonrpc": "2.0"} { "id": 1, "jsonrpc": "2.0", "result": "0x1b9c02a0a23a70" }
Apache-2.0
Ethereum API.ipynb
Snedashkovsky/Course_blockchain_data
**Name:** Omar Khaled Mahmoud Safwat Mohamed Safwat**Group:** Alex group 3
%matplotlib inline import numpy as np import matplotlib.pyplot as plt from sklearn.metrics import r2_score from sklearn.preprocessing import StandardScaler # Standardize data for faster convergence import seaborn as sns import Linear_Regression as lr sns.set()
_____no_output_____
MIT
Optimization in ML/Gradient_descent/Practical_4.ipynb
Omar-Safwat/Numerical_Methods_Projects
Data generation
X = np.array([list(range(1, 20))]).T y = -1 * X + 2 # Plot data plt.scatter(X, y) plt.xlabel("x") plt.ylabel("y")
_____no_output_____
MIT
Optimization in ML/Gradient_descent/Practical_4.ipynb
Omar-Safwat/Numerical_Methods_Projects
Adagrad 1st trial
lr_adagrad = lr.Linear_Regression(X, y) theta = lr_adagrad.fit(solver="Adagrad", alpha=0.01, max_epochs=1e3, standardize=False, stop_criteria=1e-3, eps=1e-8) lr_adagrad.show_summary() lr_adagrad.plot_LR_2D(show_trials=True, N_trials_to_show= 10) lr_adagrad.plot_MSE()
Solver summary: =============== Number of iterations: 1000 MSE: 6.797968808367863 Stop criteria was reached first: False Model Training accuracy: 0.5468020794421424
MIT
Optimization in ML/Gradient_descent/Practical_4.ipynb
Omar-Safwat/Numerical_Methods_Projects
2nd Trial
lr_adagrad = lr.Linear_Regression(X, y) theta = lr_adagrad.fit(solver="Adagrad", alpha=0.01, max_epochs=1e4, standardize=False, stop_criteria=1e-5, eps=1e-8) lr_adagrad.show_summary() lr_adagrad.plot_LR_2D(show_trials=True, N_trials_to_show= 10) lr_adagrad.plot_MSE()
Solver summary: =============== Number of iterations: 10000 MSE: 0.7111622267407923 Stop criteria was reached first: False Model Training accuracy: 0.9525891848839472
MIT
Optimization in ML/Gradient_descent/Practical_4.ipynb
Omar-Safwat/Numerical_Methods_Projects
RMSprop 1st trial
lr_rms = lr.Linear_Regression(X, y) theta = lr_rms.fit(solver="RMSprop", alpha=0.01, max_epochs=1e3, standardize=False, stop_criteria=1e-3, eps=1e-8, beta_grad=0.9) lr_rms.show_summary() lr_rms.plot_LR_2D(show_trials=True, N_trials_to_show= 10) lr_rms.plot_MSE()
Solver summary: =============== Number of iterations: 300 MSE: 0.32095489048071413 Stop criteria was reached first: True Model Training accuracy: 0.9786030073012857
MIT
Optimization in ML/Gradient_descent/Practical_4.ipynb
Omar-Safwat/Numerical_Methods_Projects
2nd Trial
lr_rms = lr.Linear_Regression(X, y) theta = lr_rms.fit(solver="RMSprop", alpha=0.001, max_epochs=1e4, standardize=False, stop_criteria=1e-4, eps=1e-8, beta_grad=0.9) lr_rms.show_summary() lr_rms.plot_LR_2D(show_trials=True, N_trials_to_show= 10) lr_rms.plot_MSE()
Solver summary: =============== Number of iterations: 3117 MSE: 0.022779696979792863 Stop criteria was reached first: True Model Training accuracy: 0.9984813535346805
MIT
Optimization in ML/Gradient_descent/Practical_4.ipynb
Omar-Safwat/Numerical_Methods_Projects
Adam 1st trial
lr_adam = lr.Linear_Regression(X, y) theta = lr_adam.fit( solver="Adam", alpha=0.001, max_epochs=1e3, standardize=False, stop_criteria=1e-3, eps=1e-7, beta_grad=0.9, beta_nu=0.8) lr_adam.show_summary() lr_adam.plot_LR_2D(show_trials=True, N_trials_to_show= 10) lr_adam.plot_MSE()
Solver summary: =============== Number of iterations: 909 MSE: 0.8091644884352912 Stop criteria was reached first: True Model Training accuracy: 0.9460557007709806
MIT
Optimization in ML/Gradient_descent/Practical_4.ipynb
Omar-Safwat/Numerical_Methods_Projects
2nd trial
lr_adam = lr.Linear_Regression(X, y) theta = lr_adam.fit( solver="Adam", alpha=0.01, max_epochs=1e4, standardize=False, stop_criteria=1e-3, eps=1e-7, beta_grad=0.9, beta_nu=0.8) lr_adam.show_summary() lr_adam.plot_LR_2D(show_trials=True, N_trials_to_show= 10) lr_adam.plot_MSE()
Solver summary: =============== Number of iterations: 424 MSE: 0.023983926232788545 Stop criteria was reached first: True Model Training accuracy: 0.9984010715844808
MIT
Optimization in ML/Gradient_descent/Practical_4.ipynb
Omar-Safwat/Numerical_Methods_Projects
Compare between three algorithmsAt * alpha = 0.01* max_epochs=1e4, * standardize=False, * stop_criteria=1e-4, * eps=1e-7, * beta_grad=0.9, * beta_nu=0.8
# Adagrad lr_adagrad = lr.Linear_Regression(X, y) theta = lr_adagrad.fit(solver="Adagrad", alpha=0.01, max_epochs=1e4, standardize=False, stop_criteria=1e-4, eps=1e-7) lr_adagrad.show_summary() lr_adagrad.plot_LR_2D(show_trials=True, N_trials_to_show= 10) lr_adagrad.plot_MSE() # RMS lr_rms = lr.Linear_Regression(X, y) ...
Solver summary: =============== Number of iterations: 473 MSE: 0.0004158716712860208 Stop criteria was reached first: True Model Training accuracy: 0.9999722752219142
MIT
Optimization in ML/Gradient_descent/Practical_4.ipynb
Omar-Safwat/Numerical_Methods_Projects
Character-Level LSTM in PyTorchIn this notebook, I'll construct a character-level LSTM with PyTorch. The network will train character by character on some text, then generate new text character by character. As an example, I will train on Anna Karenina. **This model will be able to generate new text based on the text ...
import numpy as np import torch from torch import nn import torch.nn.functional as F
_____no_output_____
Apache-2.0
Recurrent Neural Networks/Character_Level_RNN_Exercise.ipynb
sayakpaul/Favorite-Execises-from-Udacity-s-Deep-Learning-Course
Load in DataThen, we'll load the Anna Karenina text file and convert it into integers for our network to use.
# open text file and read in data as `text` with open('data/anna.txt', 'r') as f: text = f.read()
_____no_output_____
Apache-2.0
Recurrent Neural Networks/Character_Level_RNN_Exercise.ipynb
sayakpaul/Favorite-Execises-from-Udacity-s-Deep-Learning-Course
Let's check out the first 100 characters, make sure everything is peachy. According to the [American Book Review](http://americanbookreview.org/100bestlines.asp), this is the 6th best first line of a book ever.
text[:100]
_____no_output_____
Apache-2.0
Recurrent Neural Networks/Character_Level_RNN_Exercise.ipynb
sayakpaul/Favorite-Execises-from-Udacity-s-Deep-Learning-Course
TokenizationIn the cells, below, I'm creating a couple **dictionaries** to convert the characters to and from integers. Encoding the characters as integers makes it easier to use as input in the network.
# encode the text and map each character to an integer and vice versa # we create two dictionaries: # 1. int2char, which maps integers to characters # 2. char2int, which maps characters to unique integers chars = tuple(set(text)) int2char = dict(enumerate(chars)) char2int = {ch: ii for ii, ch in int2char.items()} # e...
_____no_output_____
Apache-2.0
Recurrent Neural Networks/Character_Level_RNN_Exercise.ipynb
sayakpaul/Favorite-Execises-from-Udacity-s-Deep-Learning-Course
And we can see those same characters from above, encoded as integers.
encoded[:100]
_____no_output_____
Apache-2.0
Recurrent Neural Networks/Character_Level_RNN_Exercise.ipynb
sayakpaul/Favorite-Execises-from-Udacity-s-Deep-Learning-Course
Pre-processing the dataAs you can see in our char-RNN image above, our LSTM expects an input that is **one-hot encoded** meaning that each character is converted into an integer (via our created dictionary) and *then* converted into a column vector where only it's corresponding integer index will have the value of 1 a...
def one_hot_encode(arr, n_labels): # Initialize the the encoded array one_hot = np.zeros((arr.size, n_labels), dtype=np.float32) # Fill the appropriate elements with ones one_hot[np.arange(one_hot.shape[0]), arr.flatten()] = 1. # Finally reshape it to get back to the original array ...
[[[0. 0. 0. 1. 0. 0. 0. 0.] [0. 0. 0. 0. 0. 1. 0. 0.] [0. 1. 0. 0. 0. 0. 0. 0.]]]
Apache-2.0
Recurrent Neural Networks/Character_Level_RNN_Exercise.ipynb
sayakpaul/Favorite-Execises-from-Udacity-s-Deep-Learning-Course
Making training mini-batchesTo train on this data, we also want to create mini-batches for training. Remember that we want our batches to be multiple sequences of some desired number of sequence steps. Considering a simple example, our batches would look like this:In this example, we'll take the encoded characters (pa...
def get_batches(arr, batch_size, seq_length): '''Create a generator that returns batches of size batch_size x seq_length from arr. Arguments --------- arr: Array you want to make batches from batch_size: Batch size, the number of sequences per batch seq_length: Numb...
_____no_output_____
Apache-2.0
Recurrent Neural Networks/Character_Level_RNN_Exercise.ipynb
sayakpaul/Favorite-Execises-from-Udacity-s-Deep-Learning-Course
Test Your ImplementationNow I'll make some data sets and we can check out what's going on as we batch data. Here, as an example, I'm going to use a batch size of 8 and 50 sequence steps.
batches = get_batches(encoded, 8, 50) x, y = next(batches) # printing out the first 10 items in a sequence print('x\n', x[:10, :10]) print('\ny\n', y[:10, :10])
x [[52 66 23 3 30 72 76 2 26 36] [71 50 82 2 30 66 23 30 2 23] [72 82 57 2 50 76 2 23 2 20] [71 2 30 66 72 2 48 66 25 72] [ 2 71 23 29 2 66 72 76 2 30] [48 33 71 71 25 50 82 2 23 82] [ 2 74 82 82 23 2 66 23 57 2] [34 14 56 50 82 71 0 44 65 2]] y [[66 23 3 30 72 76 2 26 36 36] [50 82 2 30 6...
Apache-2.0
Recurrent Neural Networks/Character_Level_RNN_Exercise.ipynb
sayakpaul/Favorite-Execises-from-Udacity-s-Deep-Learning-Course
If you implemented `get_batches` correctly, the above output should look something like ```x [[25 8 60 11 45 27 28 73 1 2] [17 7 20 73 45 8 60 45 73 60] [27 20 80 73 7 28 73 60 73 65] [17 73 45 8 27 73 66 8 46 27] [73 17 60 12 73 8 27 28 73 45] [66 64 17 17 46 7 20 73 60 20] [73 76 20 20 60 73 8 60 80 73] [4...
# check if GPU is available train_on_gpu = torch.cuda.is_available() if(train_on_gpu): print('Training on GPU!') else: print('No GPU available, training on CPU; consider making n_epochs very small.') class CharRNN(nn.Module): def __init__(self, tokens, n_hidden=256, n_layers=2, ...
_____no_output_____
Apache-2.0
Recurrent Neural Networks/Character_Level_RNN_Exercise.ipynb
sayakpaul/Favorite-Execises-from-Udacity-s-Deep-Learning-Course
Time to trainThe train function gives us the ability to set the number of epochs, the learning rate, and other parameters.Below we're using an Adam optimizer and cross entropy loss since we are looking at character class scores as output. We calculate the loss and perform backpropagation, as usual!A couple of details ...
from tqdm import tqdm def train(net, data, epochs=10, batch_size=10, seq_length=50, lr=0.001, clip=5, val_frac=0.1, print_every=10): ''' Training a network Arguments --------- net: CharRNN network data: text data to train the network epochs: Nu...
_____no_output_____
Apache-2.0
Recurrent Neural Networks/Character_Level_RNN_Exercise.ipynb
sayakpaul/Favorite-Execises-from-Udacity-s-Deep-Learning-Course
Instantiating the modelNow we can actually train the network. First we'll create the network itself, with some given hyperparameters. Then, define the mini-batches sizes, and start training!
# define and print the net n_hidden=256 n_layers=2 net = CharRNN(chars, n_hidden, n_layers) print(net)
CharRNN( (lstm): LSTM(83, 256, num_layers=2, batch_first=True, dropout=0.5) (dropout): Dropout(p=0.5) (fc): Linear(in_features=256, out_features=83, bias=True) )
Apache-2.0
Recurrent Neural Networks/Character_Level_RNN_Exercise.ipynb
sayakpaul/Favorite-Execises-from-Udacity-s-Deep-Learning-Course
Set your training hyperparameters!
batch_size = 128 seq_length = 75 n_epochs = 5 # start small if you are just testing initial behavior # train the model train(net, encoded, epochs=n_epochs, batch_size=batch_size, seq_length=seq_length, lr=0.001, print_every=10)
0%| | 0/5 [00:00<?, ?it/s]
Apache-2.0
Recurrent Neural Networks/Character_Level_RNN_Exercise.ipynb
sayakpaul/Favorite-Execises-from-Udacity-s-Deep-Learning-Course
Getting the best modelTo set your hyperparameters to get the best performance, you'll want to watch the training and validation losses. If your training loss is much lower than the validation loss, you're overfitting. Increase regularization (more dropout) or use a smaller network. If the training and validation losse...
# change the name, for saving multiple files model_name = 'rnn_sayak_5.net' checkpoint = {'n_hidden': net.n_hidden, 'n_layers': net.n_layers, 'state_dict': net.state_dict(), 'tokens': net.chars} with open(model_name, 'wb') as f: torch.save(checkpoint, f)
_____no_output_____
Apache-2.0
Recurrent Neural Networks/Character_Level_RNN_Exercise.ipynb
sayakpaul/Favorite-Execises-from-Udacity-s-Deep-Learning-Course
--- Making PredictionsNow that the model is trained, we'll want to sample from it and make predictions about next characters! To sample, we pass in a character and have the network predict the next character. Then we take that character, pass it back in, and get another predicted character. Just keep doing this and you...
def predict(net, char, h=None, top_k=None): ''' Given a character, predict the next character. Returns the predicted character and the hidden state. ''' # tensor inputs x = np.array([[net.char2int[char]]]) x = one_hot_encode(x, len(net.chars)) inputs ...
_____no_output_____
Apache-2.0
Recurrent Neural Networks/Character_Level_RNN_Exercise.ipynb
sayakpaul/Favorite-Execises-from-Udacity-s-Deep-Learning-Course