Open
Conversation
…g for scikit probe - Add --scikit_model_args flag to skip hyperparameter tuning with known params - Add _calculate_metrics() in ScikitProbe returning accuracy, f1, mcc, precision, recall, roc_auc, pr_auc (classification) and mse, rmse, r_squared, mae, spearman_rho, pearson_rho (regression) - Fix log_metrics() call missing from run_scikit_scheme so plots are generated - Matches metrics output of neural probe for consistency
…mpute_regression_metrics in scikit probe
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Changes
main.py:
When --scikit_model_name is specified, skips LazyPredict and goes directly to training
Added --scikit_model_args flag to pass pre-specified hyperparameters as JSON (skips hyperparameter tuning entirely when provided)
Added log_metrics() call in run_scikit_scheme so results are written to TSV and plots are generated
lazy_predict.py:
Precompute preprocessing once instead of refitting StandardScaler/Imputer per model
Added n_jobs=-1 to parallelizable models (RandomForest, etc.)
Removed slow models from LazyPredict (SVC, NuSVC, AdaBoost, KNeighbors, DecisionTree, LDA/QDA, etc.)
Added XGBoost/LightGBM to model dictionaries correctly
scikit_classes.py:
Fixed --scikit_model_name CLI arg mapping to model_name
Refactored _calculate_metrics() to use the shared metrics.py functions via EvalPrediction, returning: accuracy, f1, precision, recall, mcc, roc_auc, pr_auc (classification) and mse, rmse, r_squared, mae, spearman_rho, pearson_rho (regression)
All code paths (run_specific_model, find_best_classifier, find_best_regressor) now use _calculate_metrics
New Usage
Full pipeline (LazyPredict → best model → hyperparameter tuning):
python main.py --model_names ESMC-600 --data_names gold-ppi --use_scikit --n_jobs -1
Skip LazyPredict, tune XGBoost:
python main.py --model_names ESMC-600 --data_names gold-ppi --use_scikit --scikit_model_name XGBClassifier --scikit_n_iter 10
Skip everything — use known optimal hyperparameters directly:
python main.py --model_names ESMC-600 --data_names gold-ppi --use_scikit --scikit_model_name XGBClassifier --scikit_model_args '{"n_estimators": 500, "max_depth": 7, "learning_rate": 0.1}'