{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "GAMCR in practice: A brief overview\n",
    "=============================="
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Typical File Structure for GAMCR\n",
    "\n",
    "```plaintext\n",
    "GAMCR/\n",
    "├── experiments/\n",
    "│   ├── check_data.ipynb\n",
    "│   ├── save_data_batch.py\n",
    "│   ├── train_models.py\n",
    "│   ├── compute_statistics.py\n",
    "│   └── mysite/\n",
    "│       ├── results/\n",
    "│       │   ├── detailedresults/\n",
    "│       │   ├── groups.csv\n",
    "│       │   ├── NRF_RRD.csv\n",
    "│       │   └── streamflow.csv\n",
    "│       ├── data/\n",
    "│       │   ├── transfer.npy\n",
    "│       │   ├── lst_transfer.npy\n",
    "│       │   ├── y_0.npy\n",
    "│       │   └── ...\n",
    "│       ├── data_mysite.txt\n",
    "│       └── mysite_best_model.pkl\n",
    "```\n",
    "    \n",
    "\n",
    "### Folder Structure for Using the GAMCR Package\n",
    "\n",
    "To properly use the GAMCR package, you should have a folder for each site with the following structure:\n",
    "\n",
    "- You can name the folder as per your preference. For example: `mysite`.\n",
    "- Inside this folder, you should have a file named `data_{mysite}.txt`.  \n",
    "  This file can be created by running the notebook `check_data.ipynb`.\n",
    "- Two subfolders will be created and used by GAMCR:\n",
    "  - **`data/`**:  \n",
    "    This subfolder will store the preprocessed data, created when running the script `save_data_batch.py`.  \n",
    "    - For simulated data:  \n",
    "      - If the true transfer functions have been computed, they are saved in a matrix in the file `transfer.npy`.  \n",
    "      - The indices corresponding to each precipitation event used to compute the true transfer functions are saved in the file `lst_transfer.npy`.\n",
    "  - **`results/`**:  \n",
    "    This subfolder will store various statistics on the results of a trained model, created when running the script `compute_statistics.py`. Specifically, the following files will be generated:\n",
    "    - `groups.csv`: Describes the precipitation and/or antecedent wetness ranges for each group of data points used to compute averages.\n",
    "    - `NRF_RRD.csv`: Contains NRF, RRD, and weighted average RRD computed for the different groups of data points.\n",
    "    - `streamflow.csv`: Includes observed and predicted streamflow time series.\n",
    "- GAMCR will also save the models you train for that site in this folder.  \n",
    "  By running the script `train_models.py`, a file named `{mysite}_best_model.pkl` will be generated.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![Screen shot of an example of simulated data.](../../_static/gamcr_readme.png)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}