Delete init_repo_MLstructureMining.py
Browse files
init_repo_MLstructureMining.py
DELETED
|
@@ -1,86 +0,0 @@
|
|
| 1 |
-
import json
|
| 2 |
-
import os
|
| 3 |
-
import pickle
|
| 4 |
-
from pathlib import Path
|
| 5 |
-
from tempfile import mkdtemp, mkstemp
|
| 6 |
-
from uuid import uuid4
|
| 7 |
-
|
| 8 |
-
import numpy as np
|
| 9 |
-
import xgboost
|
| 10 |
-
from xgboost import XGBClassifier
|
| 11 |
-
|
| 12 |
-
import sklearn
|
| 13 |
-
from huggingface_hub import HfApi
|
| 14 |
-
from sklearn.datasets import load_breast_cancer
|
| 15 |
-
from sklearn.ensemble import HistGradientBoostingClassifier
|
| 16 |
-
from sklearn.experimental import enable_halving_search_cv # noqa
|
| 17 |
-
from sklearn.model_selection import HalvingGridSearchCV, train_test_split
|
| 18 |
-
import shutil
|
| 19 |
-
from skops import card, hub_utils
|
| 20 |
-
from data_loader import get_data_splits_from_clean_data
|
| 21 |
-
# Paths
|
| 22 |
-
model_path = "MLstructureMining_model.bin"
|
| 23 |
-
label_path = "labels.csv"
|
| 24 |
-
data_path = "./cifs_test_s_trained_model"
|
| 25 |
-
|
| 26 |
-
train_tuple = get_data_splits_from_clean_data(
|
| 27 |
-
data_path, label_path, simple_load=True, n_data=-1
|
| 28 |
-
)
|
| 29 |
-
print(train_tuple)
|
| 30 |
-
X_test = train_tuple[0]
|
| 31 |
-
|
| 32 |
-
booster = xgboost.Booster({'nthread': 8})
|
| 33 |
-
booster.load_model(model_path)
|
| 34 |
-
|
| 35 |
-
model = XGBClassifier()
|
| 36 |
-
|
| 37 |
-
# Set the booster
|
| 38 |
-
model._Booster = booster
|
| 39 |
-
|
| 40 |
-
local_repo = mkdtemp(prefix="skops-")
|
| 41 |
-
hub_utils.init(
|
| 42 |
-
model=model_path,
|
| 43 |
-
requirements=[f"xgboost={xgboost.__version__}"],
|
| 44 |
-
dst=local_repo,
|
| 45 |
-
task="tabular-classification",
|
| 46 |
-
data=X_test,
|
| 47 |
-
)
|
| 48 |
-
|
| 49 |
-
shutil.copy(label_path, os.path.join(local_repo, label_path))
|
| 50 |
-
if "__file__" in locals(): # __file__ not defined during docs built
|
| 51 |
-
# Add this script itself to the files to be uploaded for reproducibility
|
| 52 |
-
hub_utils.add_files(__file__, dst=local_repo)
|
| 53 |
-
|
| 54 |
-
print(os.listdir(local_repo))
|
| 55 |
-
print(type(model))
|
| 56 |
-
|
| 57 |
-
card.metadata_from_config(Path(local_repo))["model_type"] = "xgboost"
|
| 58 |
-
model_card = card.Card(model, metadata=card.metadata_from_config(Path(local_repo)))
|
| 59 |
-
model_card.add(**{"model_type": "xgboost"})
|
| 60 |
-
model_card.save(Path(local_repo) / "README.md")
|
| 61 |
-
model_card.save("README.md")
|
| 62 |
-
|
| 63 |
-
with open(os.path.join(local_repo, "config.json"), "r") as file:
|
| 64 |
-
data = json.load(file)
|
| 65 |
-
data["model_type"] = "xgboost"
|
| 66 |
-
with open(os.path.join(local_repo, "config.json"), "w") as file:
|
| 67 |
-
json.dump(data, file, indent=4)
|
| 68 |
-
|
| 69 |
-
# you can put your own token here, or set it as an environment variable before
|
| 70 |
-
# running this script.
|
| 71 |
-
token = os.environ["HF_HUB_TOKEN"]
|
| 72 |
-
|
| 73 |
-
repo_name = f"MLstructureMining"
|
| 74 |
-
user_name = HfApi().whoami(token=token)["name"]
|
| 75 |
-
repo_id = f"{user_name}/{repo_name}"
|
| 76 |
-
print(f"Creating and pushing to repo: {repo_id}")
|
| 77 |
-
|
| 78 |
-
|
| 79 |
-
hub_utils.push(
|
| 80 |
-
repo_id=repo_id,
|
| 81 |
-
source=local_repo,
|
| 82 |
-
token=token,
|
| 83 |
-
commit_message="pushing files to the repo from the example!",
|
| 84 |
-
create_remote=True,
|
| 85 |
-
private=True,
|
| 86 |
-
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|