[codex] Use a named Python import package

.github/workflows/ci.yml

@@ -49,6 +49,20 @@ jobs:
       - name: Build wheel and source distribution
         run: python -m build
 
+      - name: Verify wheel package layout
+        run: |
+          python - <<'PY'
+          import glob
+          import zipfile
+
+          wheel, = glob.glob("dist/*.whl")
+          with zipfile.ZipFile(wheel) as archive:
+              names = set(archive.namelist())
+
+          assert "aep_load_forecasting/__init__.py" in names
+          assert "src/__init__.py" not in names
+          PY
+
       - name: Smoke-test installed console commands
         run: |
           aep-demo --help

README.md

@@ -37,14 +37,16 @@ out-of-time test set:
 ## Approach
 
 - **Target:** hourly load `AEP_MW`.
-- **Features** (`src/make_features.py`):
+- **Features** (`src/aep_load_forecasting/make_features.py`):
   - Calendar: `hour`, `dayofweek`, `month`, `is_weekend`
   - Lags: `lag_1` (1 h), `lag_24` (1 day), `lag_168` (1 week)
   - Rolling means (shifted, past-only): `roll_24_mean`, `roll_168_mean`
-- **Split** (`src/xgb_eval.py`): disjoint, time-based partitions — train
+- **Split** (`src/aep_load_forecasting/xgb_eval.py`): disjoint, time-based
+  partitions — train
   (everything before the final 60 days), validation (720 h), and test (720 h).
 - **Model:** `XGBRegressor(n_estimators=800, learning_rate=0.05, max_depth=6, subsample=0.8, colsample_bytree=0.8)`.
-- **24-hour forecast** (`src/forecasting.py`, `src/forecast_24h.py`): the
+- **24-hour forecast** (`src/aep_load_forecasting/forecasting.py`,
+  `src/aep_load_forecasting/forecast_24h.py`): the
   final model is fit on all data, then steps hour-by-hour, feeding each
   prediction back in as the next `lag_1`. The reusable forecast core validates
   the hourly history and model feature contract before prediction.
@@ -65,16 +67,17 @@ out-of-time test set:
 ```text
 load-forecasting-xgboost/
 ├── src/
-│   ├── plot_load.py        # EDA: plot the last 14 days of load
-│   ├── make_features.py    # build calendar + lag + rolling features
-│   ├── sample_data.py      # deterministic dataset for a no-download demo
-│   ├── evaluation.py       # leakage-safe chronological split utilities
-│   ├── reporting.py        # reusable metrics and CSV report persistence
-│   ├── forecasting.py      # validated recursive forecast utilities
-│   ├── demo_data.py        # forecast CSV validation for the demo
-│   ├── baseline_eval.py    # naive baselines (yesterday / last week / blend)
-│   ├── xgb_eval.py         # train + evaluate XGBoost vs. baseline (last 30 days)
-│   └── forecast_24h.py     # final model + recursive next-24h forecast export
+│   └── aep_load_forecasting/
+│       ├── plot_load.py        # EDA: plot the last 14 days of load
+│       ├── make_features.py    # build calendar + lag + rolling features
+│       ├── sample_data.py      # deterministic no-download dataset
+│       ├── evaluation.py       # chronological split utilities
+│       ├── reporting.py        # metrics and CSV report persistence
+│       ├── forecasting.py      # validated recursive forecast utilities
+│       ├── demo_data.py        # forecast CSV validation for the demo
+│       ├── baseline_eval.py    # yesterday / last-week / blend baselines
+│       ├── xgb_eval.py         # train and evaluate XGBoost
+│       └── forecast_24h.py     # export a recursive future forecast
 ├── streamlit_app.py        # interactive demo (reads assets/forecast_next24h.csv)
 ├── assets/                 # plots + a sample forecast CSV
 ├── requirements.txt
@@ -102,8 +105,8 @@ This single command creates the sample data, features, baseline and XGBoost
 metrics, evaluation plots, trained model, and next-24h forecast under the
 existing ignored `data/`, `reports/`, and `models/` directories. Use
 `--output-dir` to keep every artifact below a different directory. The module
-form, `python -m src.demo_pipeline`, remains available when working directly
-from a source checkout.
+form, `python -m aep_load_forecasting.demo_pipeline`, remains available after
+installing the project.
 
 Each successful run also writes `reports/sample_run_manifest.json`. The
 manifest records the effective pipeline parameters, Python and dependency
@@ -160,30 +163,30 @@ mkdir data
 #    -> place AEP_hourly.csv in .\data\
 
 # 2) build the validated feature table
-python -m src.make_features
+aep-make-features
 
 # 3) evaluate baselines and XGBoost (test = last 30 days)
-python -m src.baseline_eval
-python -m src.xgb_eval
+aep-baseline-eval
+aep-xgb-eval
 
 # 4) train the final model and export the next-24h forecast
-python -m src.forecast_24h
+aep-forecast
 ```
 
 The evaluation commands create reproducible artifacts automatically:
 
 | Command | Metrics | Plot |
 |---|---|---|
-| `python -m src.baseline_eval` | `reports/baseline_metrics.csv` | `reports/figures/baseline_evaluation.png` |
-| `python -m src.xgb_eval` | `reports/xgb_evaluation_metrics.csv` | `reports/figures/xgb_evaluation.png` |
-| `python -m src.forecast_24h` | `reports/forecast_next24h.csv` | `reports/figures/forecast_next24h.png` |
+| `aep-baseline-eval` | `reports/baseline_metrics.csv` | `reports/figures/baseline_evaluation.png` |
+| `aep-xgb-eval` | `reports/xgb_evaluation_metrics.csv` | `reports/figures/xgb_evaluation.png` |
+| `aep-forecast` | `reports/forecast_next24h.csv` | `reports/figures/forecast_next24h.png` |
 
 The commands create their output directories automatically. The forecast
 horizon, evaluation windows, boosting rounds, and every input or output path
 can also be configured from the command line:
 
 ```powershell
-python -m src.forecast_24h `
+aep-forecast `
   --horizon 48 `
   --estimators 400 `
   --output reports\forecast_next48h.csv `
@@ -215,7 +218,7 @@ before rendering them.
 Custom paths and target columns can be supplied without editing the source:
 
 ```powershell
-python -m src.make_features `
+aep-make-features `
   --input data\AEP_hourly.csv `
   --output data\features_aep.csv `
   --target AEP_MW

pyproject.toml

@@ -40,19 +40,19 @@ dev = [
 ]
 
 [project.scripts]
-aep-demo = "src.demo_pipeline:main"
-aep-sample-data = "src.sample_data:main"
-aep-make-features = "src.make_features:main"
-aep-baseline-eval = "src.baseline_eval:main"
-aep-xgb-eval = "src.xgb_eval:main"
-aep-forecast = "src.forecast_24h:main"
+aep-demo = "aep_load_forecasting.demo_pipeline:main"
+aep-sample-data = "aep_load_forecasting.sample_data:main"
+aep-make-features = "aep_load_forecasting.make_features:main"
+aep-baseline-eval = "aep_load_forecasting.baseline_eval:main"
+aep-xgb-eval = "aep_load_forecasting.xgb_eval:main"
+aep-forecast = "aep_load_forecasting.forecast_24h:main"
 
 [project.urls]
 Homepage = "https://github.com/amarakramali/load-forecasting-xgboost"
 Issues = "https://github.com/amarakramali/load-forecasting-xgboost/issues"
 
-[tool.setuptools]
-packages = ["src"]
+[tool.setuptools.packages.find]
+where = ["src"]
 
 [tool.ruff]
 target-version = "py310"

src/baseline_eval.py → src/aep_load_forecasting/baseline_eval.py

@@ -9,8 +9,8 @@
 import pandas as pd
 from matplotlib.figure import Figure
 
-from src.evaluation import HOURS_PER_DAY, trailing_window
-from src.reporting import (
+from aep_load_forecasting.evaluation import HOURS_PER_DAY, trailing_window
+from aep_load_forecasting.reporting import (
     EvaluationResult,
     evaluate_predictions,
     format_result,

src/demo_pipeline.py → src/aep_load_forecasting/demo_pipeline.py

@@ -12,29 +12,35 @@
 
 import joblib
 
-from src.baseline_eval import (
+from aep_load_forecasting.baseline_eval import (
     DEFAULT_EVALUATION_DAYS,
     DEFAULT_PLOT_DAYS,
     evaluate_baselines,
     save_baseline_plot,
 )
-from src.evaluation import HOURS_PER_DAY
-from src.forecast_24h import (
+from aep_load_forecasting.evaluation import HOURS_PER_DAY
+from aep_load_forecasting.forecast_24h import (
     DEFAULT_ESTIMATORS,
     save_forecast_plot,
     train_final_model,
 )
-from src.forecasting import recursive_forecast
-from src.make_features import load_hourly_series, make_feature_table
-from src.reporting import save_results
-from src.sample_data import (
+from aep_load_forecasting.forecasting import recursive_forecast
+from aep_load_forecasting.make_features import (
+    load_hourly_series,
+    make_feature_table,
+)
+from aep_load_forecasting.reporting import save_results
+from aep_load_forecasting.sample_data import (
     DEFAULT_DAYS,
     DEFAULT_SEED,
     DEFAULT_START,
     generate_hourly_load,
     save_sample_data,
 )
-from src.xgb_eval import evaluate_xgboost, save_evaluation_plot
+from aep_load_forecasting.xgb_eval import (
+    evaluate_xgboost,
+    save_evaluation_plot,
+)
 
 DEFAULT_OUTPUT_DIR = Path(".")
 DEFAULT_HORIZON = 24

src/forecast_24h.py → src/aep_load_forecasting/forecast_24h.py

@@ -10,12 +10,15 @@
 from matplotlib.figure import Figure
 from xgboost import XGBRegressor
 
-from src.forecasting import (
+from aep_load_forecasting.forecasting import (
     FORECAST_FEATURES,
     recursive_forecast,
     validate_feature_columns,
 )
-from src.make_features import DEFAULT_INPUT, load_hourly_series
+from aep_load_forecasting.make_features import (
+    DEFAULT_INPUT,
+    load_hourly_series,
+)
 
 DEFAULT_FEATURES = Path("data") / "features_aep.csv"
 DEFAULT_FORECAST = Path("reports") / "forecast_next24h.csv"

src/xgb_eval.py → src/aep_load_forecasting/xgb_eval.py

@@ -12,9 +12,16 @@
 from matplotlib.figure import Figure
 from xgboost import XGBRegressor
 
-from src.evaluation import HOURS_PER_DAY, chronological_split, trailing_window
-from src.forecasting import FORECAST_FEATURES, validate_feature_columns
-from src.reporting import (
+from aep_load_forecasting.evaluation import (
+    HOURS_PER_DAY,
+    chronological_split,
+    trailing_window,
+)
+from aep_load_forecasting.forecasting import (
+    FORECAST_FEATURES,
+    validate_feature_columns,
+)
+from aep_load_forecasting.reporting import (
     EvaluationResult,
     evaluate_predictions,
     format_result,

streamlit_app.py

@@ -4,7 +4,7 @@
 
 import streamlit as st
 
-from src.demo_data import (
+from aep_load_forecasting.demo_data import (
     FORECAST_COLUMN,
     ForecastDataError,
     forecast_plot_columns,

tests/test_baseline_eval.py

@@ -4,7 +4,7 @@
 import pandas as pd
 import pytest
 
-from src.baseline_eval import (
+from aep_load_forecasting.baseline_eval import (
     evaluate_baselines,
     load_baseline_features,
     main,

tests/test_demo_data.py

@@ -5,7 +5,7 @@
 import pandas as pd
 import pytest
 
-from src.demo_data import (
+from aep_load_forecasting.demo_data import (
     BASELINE_COLUMN,
     FORECAST_COLUMN,
     ForecastDataError,

tests/test_demo_pipeline.py

@@ -7,8 +7,8 @@
 import pandas as pd
 import pytest
 
-from src.demo_pipeline import main, run_demo_pipeline
-from src.forecasting import FORECAST_FEATURES
+from aep_load_forecasting.demo_pipeline import main, run_demo_pipeline
+from aep_load_forecasting.forecasting import FORECAST_FEATURES
 
 
 def test_command_writes_complete_reproducible_demo(tmp_path) -> None:

tests/test_evaluation.py

@@ -3,7 +3,10 @@
 import pandas as pd
 import pytest
 
-from src.evaluation import chronological_split, trailing_window
+from aep_load_forecasting.evaluation import (
+    chronological_split,
+    trailing_window,
+)
 
 
 def hourly_frame(periods: int = 2_000) -> pd.DataFrame:

tests/test_forecast_24h.py

@@ -4,8 +4,8 @@
 
 import pandas as pd
 
-import src.forecast_24h as command
-from src.forecasting import FORECAST_FEATURES
+import aep_load_forecasting.forecast_24h as command
+from aep_load_forecasting.forecasting import FORECAST_FEATURES
 
 
 def test_command_forwards_estimator_count_and_writes_outputs(

tests/test_forecasting.py

@@ -4,7 +4,7 @@
 import pandas as pd
 import pytest
 
-from src.forecasting import (
+from aep_load_forecasting.forecasting import (
     FORECAST_FEATURES,
     make_forecast_row,
     recursive_forecast,

tests/test_make_features.py

@@ -3,7 +3,11 @@
 import pandas as pd
 import pytest
 
-from src.make_features import build_features, load_hourly_series, make_feature_table
+from aep_load_forecasting.make_features import (
+    build_features,
+    load_hourly_series,
+    make_feature_table,
+)
 
 
 def synthetic_load(periods: int = 200) -> pd.Series:

tests/test_reporting.py

@@ -4,7 +4,7 @@
 import pandas as pd
 import pytest
 
-from src.reporting import (
+from aep_load_forecasting.reporting import (
     EvaluationResult,
     evaluate_predictions,
     format_result,

tests/test_sample_data.py

@@ -3,8 +3,11 @@
 import pandas as pd
 import pytest
 
-from src.make_features import build_features, load_hourly_series
-from src.sample_data import generate_hourly_load, main
+from aep_load_forecasting.make_features import (
+    build_features,
+    load_hourly_series,
+)
+from aep_load_forecasting.sample_data import generate_hourly_load, main
 
 
 def test_generate_hourly_load_is_deterministic_and_hourly() -> None:

tests/test_xgb_eval.py

@@ -4,8 +4,8 @@
 import pandas as pd
 import pytest
 
-from src.forecasting import FORECAST_FEATURES
-from src.xgb_eval import (
+from aep_load_forecasting.forecasting import FORECAST_FEATURES
+from aep_load_forecasting.xgb_eval import (
     evaluate_xgboost,
     load_evaluation_features,
     main,