Fix first stages of experiments

.gitignore

@@ -5,4 +5,6 @@ errors
 results
 d2v*
 !d2v*.py
-.venv/
+.venv/
+dist/
+wsmf.egg-info/

README.md

@@ -7,4 +7,9 @@ export PYTHONPATH=`pwd`
 ### Loading data
 ```
 python bin/load_data.py
+```
+
+### Release
+```
+python -m build
 ```

bin/load_data/download_data_from_openml.py

@@ -1,4 +1,5 @@
 import json
+import warnings
 from operator import itemgetter
 
 import pytorch_lightning as pl
@@ -12,8 +13,10 @@
 )
 from experiments_engine.paths import paths_provider
 
+warnings.simplefilter("ignore")
 
-def main():
+
+def main() -> None:
     pl.seed_everything(123)
     logger.info("Loading tasks ids")
     with open(paths_provider.tasks_ids_path, "r") as f:

bin/metatrain_preparation/calculate_metafeatures.py

@@ -43,7 +43,7 @@ def calculate_meta_features_record_from_path(
     return ft  # type: ignore
 
 
-def main():
+def main() -> None:
     logger.info("Calculating meta-features")
     paths = list(sorted(paths_provider.datasets_splitted_path.iterdir()))
     with Pool(14) as p:
@@ -57,7 +57,16 @@ def main():
     logger.info("Processing meta-features")
     meta_features_df = pd.concat(meta_features)
     meta_features_df = meta_features_df.drop(
-        columns=["num_to_cat", "sd_ratio", "lh_trace", "roy_root"]
+        columns=[
+            "num_to_cat",
+            "sd_ratio",
+            "lh_trace",
+            "roy_root",
+            "can_cor.mean",
+            "nr_disc",
+            "p_trace",
+            "w_lambda",
+        ]
     )
     meta_features_df.iloc[:, :-1] = meta_features_df.iloc[:, :-1].fillna(
         meta_features_df.iloc[:, :-1].mean(axis=0)

bin/metatrain_preparation/perform_hpo_on_all_datasets.py

@@ -14,7 +14,7 @@
 from experiments_engine.utils import extract_dataset_name_from_path
 
 
-def main():
+def main() -> None:
     objectives_to_evaluate = [XGBoostObjective]
     erroneous_datasets = []
     for dataset_path in (

data/openml/.gitignore

@@ -0,0 +1 @@
+**.parquet

data/openml/prohibited_datasets.json

@@ -0,0 +1,18 @@
+[
+    "Amazon_employee_access",
+    "DRSongsLyrics",
+    "StackOverflow-polarity",
+    "StackOverflow-polarity-test",
+    "SyskillWebert-Bands",
+    "SyskillWebert-BioMedical",
+    "SyskillWebert-Sheep",
+    "Traffic_violations",
+    "breastTumor",
+    "credit-g",
+    "kick",
+    "lymph",
+    "okcupid-stem",
+    "online_shoppers",
+    "primary-tumor",
+    "usp05-ft"
+]

pyproject.toml

@@ -0,0 +1,38 @@
+[build-system]
+requires = ["setuptools", "setuptools-scm"]
+build-backend = "setuptools.build_meta"
+
+[tool.setuptools.packages.find]
+include = ["wsmf"]
+
+[project]
+name = "wsmf"
+version = "0.0.1"
+authors = [
+  { name="Antoni Zajko", email="antoni.zajko.1@gmail.com" }
+]
+description = "Package with encoder-based approaches to warm-starting Bayesian Hyperparameter Optimization."
+readme = "README.md"
+requires-python = ">=3.10"
+classifiers = [
+    "Programming Language :: Python :: 3",
+    "License :: OSI Approved :: MIT License",
+    "Operating System :: OS Independent",
+]
+
+dependencies = [
+    "dataset2vec>=1.0.0",
+    "openml>=0.14.2",
+    "loguru>=0.7.2",
+    "optuna>=3.6.1",
+    "xgboost>=2.0.3",
+    "scikit-learn>=1.4.2",
+    "pytest>=8.2.1",
+    "pymfe>=0.4.3",
+    "seaborn>=0.13.2",
+    "tensorboard>=2.17.0",
+    "numpy>=1.26.4",
+]
+
+[project.urls]
+"Homepage" = "https://github.com/azoz01/wsmf"

requirements_dev.txt

@@ -12,4 +12,6 @@ numpy==1.26.4
 isort==5.13.2
 black==24.8.0
 flake8==7.1.0
-mypy==1.11.1
+mypy==1.11.1
+twine==5.0.0
+build==1.2.1