workbench 0.8.162__py3-none-any.whl → 0.8.202__py3-none-any.whl

workbench/model_scripts/custom_models/uq_models/mapie_xgb.template

@@ -1,203 +0,0 @@
-# Model: HistGradientBoosting with MAPIE Conformalized Quantile Regression
-from mapie.regression import MapieQuantileRegressor
-from sklearn.ensemble import HistGradientBoostingRegressor
-from sklearn.model_selection import train_test_split
-import numpy as np
-
-# Template Placeholders
-TEMPLATE_PARAMS = {
-    "features": "{{feature_list}}",
-    "target": "{{target_column}}",
-    "train_all_data": "{{train_all_data}}"
-}
-
-from io import StringIO
-import json
-import argparse
-import joblib
-import os
-import pandas as pd
-
-
-# Function to check if dataframe is empty
-def check_dataframe(df: pd.DataFrame, df_name: str) -> None:
-    """Check if the DataFrame is empty and raise an error if so."""
-    if df.empty:
-        msg = f"*** The training data {df_name} has 0 rows! ***STOPPING***"
-        print(msg)
-        raise ValueError(msg)
-
-
-def match_features_case_insensitive(df: pd.DataFrame, model_features: list) -> pd.DataFrame:
-    """
-    Matches and renames DataFrame columns to match model feature names (case-insensitive).
-    Prioritizes exact matches, then case-insensitive matches.
-    
-    Raises ValueError if any model features cannot be matched.
-    """
-    df_columns_lower = {col.lower(): col for col in df.columns}
-    rename_dict = {}
-    missing = []
-    for feature in model_features:
-        if feature in df.columns:
-            continue  # Exact match
-        elif feature.lower() in df_columns_lower:
-            rename_dict[df_columns_lower[feature.lower()]] = feature
-        else:
-            missing.append(feature)
-    
-    if missing:
-        raise ValueError(f"Features not found: {missing}")
-
-    # Rename the DataFrame columns to match the model features
-    return df.rename(columns=rename_dict)
-
-
-# TRAINING SECTION
-#
-# This section (__main__) is where SageMaker will execute the training job
-# and save the model artifacts to the model directory.
-#
-if __name__ == "__main__":
-    # Template Parameters
-    features = TEMPLATE_PARAMS["features"]
-    target = TEMPLATE_PARAMS["target"]
-    train_all_data = TEMPLATE_PARAMS["train_all_data"]
-    validation_split = 0.2
-
-    # Script arguments for input/output directories
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--model-dir", type=str, default=os.environ.get("SM_MODEL_DIR", "/opt/ml/model"))
-    parser.add_argument("--train", type=str, default=os.environ.get("SM_CHANNEL_TRAIN", "/opt/ml/input/data/train"))
-    parser.add_argument(
-        "--output-data-dir", type=str, default=os.environ.get("SM_OUTPUT_DATA_DIR", "/opt/ml/output/data")
-    )
-    args = parser.parse_args()
-
-    # Load training data from the specified directory
-    training_files = [
-        os.path.join(args.train, file)
-        for file in os.listdir(args.train) if file.endswith(".csv")
-    ]
-    df = pd.concat([pd.read_csv(file, engine="python") for file in training_files])
-
-    # Check if the DataFrame is empty
-    check_dataframe(df, "training_df")
-
-    # Training data split logic
-    if train_all_data:
-        # Use all data for both training and validation
-        print("Training on all data...")
-        df_train = df.copy()
-        df_val = df.copy()
-    elif "training" in df.columns:
-        # Split data based on a 'training' column if it exists
-        print("Splitting data based on 'training' column...")
-        df_train = df[df["training"]].copy()
-        df_val = df[~df["training"]].copy()
-    else:
-        # Perform a random split if no 'training' column is found
-        print("Splitting data randomly...")
-        df_train, df_val = train_test_split(df, test_size=validation_split, random_state=42)
-
-    # Create HistGradientBoosting base model configured for quantile regression
-    base_estimator = HistGradientBoostingRegressor(
-        loss='quantile',  # Required for MAPIE CQR
-        quantile=0.5,  # Will be overridden by MAPIE for different quantiles
-        max_iter=1000,
-        max_depth=6,
-        learning_rate=0.01,
-        random_state=42
-    )
-
-    # Create MAPIE CQR predictor - it will create quantile versions internally
-    model = MapieQuantileRegressor(
-        estimator=base_estimator,
-        method="quantile",
-        cv="split",
-        alpha=0.05  # For 95% coverage
-    )
-
-    # Prepare features and targets for training
-    X_train = df_train[features]
-    X_val = df_val[features]
-    y_train = df_train[target]
-    y_val = df_val[target]
-
-    # Fit the MAPIE CQR model (train/calibration is handled internally)
-    model.fit(X_train, y_train)
-
-    # Save the trained model and any necessary assets
-    joblib.dump(model, os.path.join(args.model_dir, "model.joblib"))
-
-    # Save the feature list to validate input during predictions
-    with open(os.path.join(args.model_dir, "feature_columns.json"), "w") as fp:
-        json.dump(features, fp)
-
-
-#
-# Inference Section
-#
-def model_fn(model_dir):
-    """Load and return the model from the specified directory."""
-    return joblib.load(os.path.join(model_dir, "model.joblib"))
-
-
-def input_fn(input_data, content_type):
-    """Parse input data and return a DataFrame."""
-    if not input_data:
-        raise ValueError("Empty input data is not supported!")
-
-    # Decode bytes to string if necessary
-    if isinstance(input_data, bytes):
-        input_data = input_data.decode("utf-8")
-
-    if "text/csv" in content_type:
-        return pd.read_csv(StringIO(input_data))
-    elif "application/json" in content_type:
-        return pd.DataFrame(json.loads(input_data))  # Assumes JSON array of records
-    else:
-        raise ValueError(f"{content_type} not supported!")
-
-
-def output_fn(output_df, accept_type):
-    """Supports both CSV and JSON output formats."""
-    if "text/csv" in accept_type:
-        csv_output = output_df.fillna("N/A").to_csv(index=False)  # CSV with N/A for missing values
-        return csv_output, "text/csv"
-    elif "application/json" in accept_type:
-        return output_df.to_json(orient="records"), "application/json"  # JSON array of records (NaNs -> null)
-    else:
-        raise RuntimeError(f"{accept_type} accept type is not supported by this script.")
-
-
-def predict_fn(df, model):
-    """Make predictions using MAPIE CQR and return the DataFrame with results."""
-    model_dir = os.environ.get("SM_MODEL_DIR", "/opt/ml/model")
-
-    # Load feature columns from the saved file
-    with open(os.path.join(model_dir, "feature_columns.json")) as fp:
-        model_features = json.load(fp)
-
-    # Match features in a case-insensitive manner
-    matched_df = match_features_case_insensitive(df, model_features)
-
-    # Get CQR predictions - returns point prediction and intervals
-    X_pred = matched_df[model_features]
-    y_pred, y_pis = model.predict(X_pred)
-
-    # Add predictions to dataframe with 95% intervals
-    df["prediction"] = y_pred
-    df["q_025"] = y_pis[:, 0, 0]  # Lower bound (2.5th percentile)
-    df["q_975"] = y_pis[:, 1, 0]  # Upper bound (97.5th percentile)
-
-    # Calculate std estimate from 95% interval
-    interval_width_95 = df["q_975"] - df["q_025"]
-    df["prediction_std"] = interval_width_95 / 3.92  # 95% CI = ±1.96σ, so width = 3.92σ
-
-    # Calculate 50% intervals using normal approximation
-    df["q_25"] = df["prediction"] - 0.674 * df["prediction_std"]
-    df["q_75"] = df["prediction"] + 0.674 * df["prediction_std"]
-
-    # Return the modified DataFrame
-    return df

workbench/model_scripts/ensemble_xgb/generated_model_script.py

@@ -1,279 +0,0 @@
-# Template Placeholders
-TEMPLATE_PARAMS = {
-    "model_type": "ensemble_regressor",
-    "target_column": "solubility",
-    "feature_list": ['molwt', 'mollogp', 'molmr', 'heavyatomcount', 'numhacceptors', 'numhdonors', 'numheteroatoms', 'numrotatablebonds', 'numvalenceelectrons', 'numaromaticrings', 'numsaturatedrings', 'numaliphaticrings', 'ringcount', 'tpsa', 'labuteasa', 'balabanj', 'bertzct'],
-    "model_metrics_s3_path": "s3://sandbox-sageworks-artifacts/models/aqsol-ensemble/training"
-}
-
-# Imports for XGB Model
-import xgboost as xgb
-import awswrangler as wr
-import numpy as np
-
-# Model Performance Scores
-from sklearn.metrics import (
-    mean_absolute_error,
-    r2_score,
-    root_mean_squared_error
-)
-
-from io import StringIO
-import json
-import argparse
-import os
-import pandas as pd
-
-
-# Function to check if dataframe is empty
-def check_dataframe(df: pd.DataFrame, df_name: str) -> None:
-    """
-    Check if the provided dataframe is empty and raise an exception if it is.
-
-    Args:
-        df (pd.DataFrame): DataFrame to check
-        df_name (str): Name of the DataFrame
-    """
-    if df.empty:
-        msg = f"*** The training data {df_name} has 0 rows! ***STOPPING***"
-        print(msg)
-        raise ValueError(msg)
-
-def match_features_case_insensitive(df: pd.DataFrame, model_features: list) -> pd.DataFrame:
-    """
-    Matches and renames the DataFrame's column names to match the model's feature names (case-insensitive).
-    Prioritizes exact case matches first, then falls back to case-insensitive matching if no exact match exists.
-
-    Args:
-        df (pd.DataFrame): The DataFrame with the original columns.
-        model_features (list): The desired list of feature names (mixed case allowed).
-
-    Returns:
-        pd.DataFrame: The DataFrame with renamed columns to match the model's feature names.
-    """
-    # Create a mapping for exact and case-insensitive matching
-    exact_match_set = set(df.columns)
-    column_map = {}
-
-    # Build the case-insensitive map (if we have any duplicate columns, the first one wins)
-    for col in df.columns:
-        lower_col = col.lower()
-        if lower_col not in column_map:
-            column_map[lower_col] = col
-
-    # Create a dictionary for renaming
-    rename_dict = {}
-    for feature in model_features:
-        # Check for an exact match first
-        if feature in exact_match_set:
-            rename_dict[feature] = feature
-
-        # If not an exact match, fall back to case-insensitive matching
-        elif feature.lower() in column_map:
-            rename_dict[column_map[feature.lower()]] = feature
-
-    # Rename the columns in the DataFrame to match the model's feature names
-    return df.rename(columns=rename_dict)
-
-
-if __name__ == "__main__":
-    """The main function is for training the XGBoost Quantile Regression models"""
-
-    # Harness Template Parameters
-    target = TEMPLATE_PARAMS["target_column"]
-    feature_list = TEMPLATE_PARAMS["feature_list"]
-    model_metrics_s3_path = TEMPLATE_PARAMS["model_metrics_s3_path"]
-    models = {}
-
-    # Script arguments for input/output directories
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--model-dir", type=str, default=os.environ.get("SM_MODEL_DIR", "/opt/ml/model"))
-    parser.add_argument("--train", type=str, default=os.environ.get("SM_CHANNEL_TRAIN", "/opt/ml/input/data/train"))
-    parser.add_argument(
-        "--output-data-dir", type=str, default=os.environ.get("SM_OUTPUT_DATA_DIR", "/opt/ml/output/data")
-    )
-    args = parser.parse_args()
-
-    # Read the training data into DataFrames
-    training_files = [
-        os.path.join(args.train, file)
-        for file in os.listdir(args.train)
-        if file.endswith(".csv")
-    ]
-    print(f"Training Files: {training_files}")
-
-    # Combine files and read them all into a single pandas dataframe
-    df = pd.concat([pd.read_csv(file, engine="python") for file in training_files])
-
-    # Check if the dataframe is empty
-    check_dataframe(df, "training_df")
-
-    # Features/Target output
-    print(f"Target: {target}")
-    print(f"Features: {str(feature_list)}")
-    print(f"Data Shape: {df.shape}")
-
-    # Grab our Features and Target with traditional X, y handles
-    y = df[target]
-    X = df[feature_list]
-
-    # Train 50 models with random 70/30 splits of the data
-    for model_id in range(50):
-        # Model Name
-        model_name = f"m_{model_id:02}"
-
-        # Bootstrap sample (50% with replacement)
-        sample_size = int(0.5 * len(X))
-        bootstrap_indices = np.random.choice(len(X), size=sample_size, replace=True)
-        X_train, y_train = X.iloc[bootstrap_indices], y.iloc[bootstrap_indices]
-        print(f"Training Model {model_name} with {len(X_train)} rows")
-        model = xgb.XGBRegressor(reg_alpha=0.1, reg_lambda=1.0)
-        model.fit(X_train, y_train)
-
-        # Store the model
-        models[model_name] = model
-
-    # Run predictions for each model
-    all_predictions = {model_name: model.predict(X) for model_name, model in models.items()}
-
-    # Create a copy of the provided DataFrame and add the new columns
-    result_df = df[[target]].copy()
-
-    # Add the model predictions to the DataFrame
-    for name, preds in all_predictions.items():
-        result_df[name] = preds
-
-    # Add the main prediction to the DataFrame (mean of all models)
-    result_df["prediction"] = result_df[[name for name in result_df.columns if name.startswith("m_")]].mean(axis=1)
-
-    # Now compute residuals on the rmse prediction
-    result_df["residual"] = result_df[target] - result_df["prediction"]
-    result_df["residual_abs"] = result_df["residual"].abs()
-
-
-    # Save the results dataframe to S3
-    wr.s3.to_csv(
-        result_df,
-        path=f"{model_metrics_s3_path}/validation_predictions.csv",
-        index=False,
-    )
-
-    # Report Performance Metrics
-    rmse = root_mean_squared_error(result_df[target], result_df["prediction"])
-    mae = mean_absolute_error(result_df[target], result_df["prediction"])
-    r2 = r2_score(result_df[target], result_df["prediction"])
-    print(f"RMSE: {rmse:.3f}")
-    print(f"MAE: {mae:.3f}")
-    print(f"R2: {r2:.3f}")
-    print(f"NumRows: {len(result_df)}")
-
-    # Now save the models
-    for name, model in models.items():
-        model_path = os.path.join(args.model_dir, f"{name}.json")
-        print(f"Saving model:  {model_path}")
-        model.save_model(model_path)
-
-    # Also save the features (this will validate input during predictions)
-    with open(os.path.join(args.model_dir, "feature_columns.json"), "w") as fp:
-        json.dump(feature_list, fp)
-
-
-def model_fn(model_dir) -> dict:
-    """Deserialized and return all the fitted models from the model directory.
-
-    Args:
-        model_dir (str): The directory where the models are stored.
-
-    Returns:
-        dict: A dictionary of the models.
-    """
-
-    # Load ALL the models from the model directory
-    models = {}
-    for file in os.listdir(model_dir):
-        if file.startswith("m_") and file.endswith(".json"):  # The Quantile models
-            # Load the model
-            model_path = os.path.join(model_dir, file)
-            print(f"Loading model: {model_path}")
-            model = xgb.XGBRegressor()
-            model.load_model(model_path)
-
-            # Store the model
-            m_name = os.path.splitext(file)[0]
-            models[m_name] = model
-
-    # Return all the models
-    return models
-
-
-def input_fn(input_data, content_type):
-    """Parse input data and return a DataFrame."""
-    if not input_data:
-        raise ValueError("Empty input data is not supported!")
-    
-    # Decode bytes to string if necessary
-    if isinstance(input_data, bytes):
-        input_data = input_data.decode("utf-8")
-
-    if "text/csv" in content_type:
-        return pd.read_csv(StringIO(input_data))
-    elif "application/json" in content_type:
-        return pd.DataFrame(json.loads(input_data))  # Assumes JSON array of records
-    else:
-        raise ValueError(f"{content_type} not supported!")
-
-
-def output_fn(output_df, accept_type):
-    """Supports both CSV and JSON output formats."""
-    if "text/csv" in accept_type:
-        csv_output = output_df.fillna("N/A").to_csv(index=False)  # CSV with N/A for missing values
-        return csv_output, "text/csv"
-    elif "application/json" in accept_type:
-        return output_df.to_json(orient="records"), "application/json"  # JSON array of records (NaNs -> null)
-    else:
-        raise RuntimeError(f"{accept_type} accept type is not supported by this script.")
-
-
-def predict_fn(df, models) -> pd.DataFrame:
-    """Make Predictions with our XGB Quantile Regression Model
-
-    Args:
-        df (pd.DataFrame): The input DataFrame
-        models (dict): The dictionary of models to use for predictions
-
-    Returns:
-        pd.DataFrame: The DataFrame with the predictions added
-    """
-
-    # Grab our feature columns (from training)
-    model_dir = os.environ.get("SM_MODEL_DIR", "/opt/ml/model")
-    with open(os.path.join(model_dir, "feature_columns.json")) as fp:
-        model_features = json.load(fp)
-    print(f"Model Features: {model_features}")
-
-    # We're going match features in a case-insensitive manner, accounting for all the permutations
-    # - Model has a feature list that's any case ("Id", "taCos", "cOunT", "likes_tacos")
-    # - Incoming data has columns that are mixed case ("ID", "Tacos", "Count", "Likes_Tacos")
-    matched_df = match_features_case_insensitive(df, model_features)
-
-    # Predict the features against all the models
-    for name, model in models.items():
-        df[name] = model.predict(matched_df[model_features])
-
-    # Add quantiles for consistency with other UQ models
-    df["q_025"] = df[[name for name in df.columns if name.startswith("m_")]].quantile(0.025, axis=1)
-    df["q_975"] = df[[name for name in df.columns if name.startswith("m_")]].quantile(0.975, axis=1)
-    df["q_25"] = df[[name for name in df.columns if name.startswith("m_")]].quantile(0.25, axis=1)
-    df["q_75"] = df[[name for name in df.columns if name.startswith("m_")]].quantile(0.75, axis=1)
-
-    # Compute the mean, min, max and stddev of the predictions
-    df["prediction"] = df[[name for name in df.columns if name.startswith("m_")]].mean(axis=1)
-    df["p_min"] = df[[name for name in df.columns if name.startswith("m_")]].min(axis=1)
-    df["p_max"] = df[[name for name in df.columns if name.startswith("m_")]].max(axis=1)
-    df["prediction_std"] = df[[name for name in df.columns if name.startswith("m_")]].std(axis=1)
-
-    # Reorganize the columns so they are in alphabetical order
-    df = df.reindex(sorted(df.columns), axis=1)
-
-    # All done, return the DataFrame
-    return df