example notebook for batch training execution (#245)

* Create example_batch_training.ipynb * Update example_batch_training.ipynb * Update example_batch_training.ipynb * Update example_batch_training.ipynb * Update notebooks/examples/example_batch_training.ipynb Co-authored-by: Hannah Eslinger <[email protected]> * Update notebooks/examples/example_batch_training.ipynb Co-authored-by: Hannah Eslinger <[email protected]> * Update notebooks/examples/example_batch_training.ipynb Co-authored-by: Hannah Eslinger <[email protected]> * Update example_batch_training.ipynb * Update example_batch_training.ipynb Co-authored-by: Hannah Eslinger <[email protected]>
NREL · Dec 5, 2022 · 2bf6cbe · 2bf6cbe
1 parent 227fc34
commit 2bf6cbe
Showing 1 changed file with 154 additions and 0 deletions.
diff --git a/notebooks/examples/example_batch_training.ipynb b/notebooks/examples/example_batch_training.ipynb
@@ -0,0 +1,154 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import json\n",
+    "import logging\n",
+    "import os\n",
+    "from pathlib import Path\n",
+    "\n",
+    "from wattile.buildings_processing import prep_for_rnn\n",
+    "from wattile.data_reading import read_dataset_from_file\n",
+    "from wattile.entry_point import init_logging\n",
+    "from wattile.models import ModelFactory\n",
+    "\n",
+    "PROJECT_DIRECTORY = Path().resolve().parents[1]\n",
+    "logger = logging.getLogger(str(os.getpid()))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "\"\"\"\n",
+    "define parametric space\n",
+    "\"\"\"\n",
+    "parametric_space = [\n",
+    "    {\n",
+    "        \"building\": \"Synthetic Site\",\n",
+    "        \"targets\": [\n",
+    "            \"Synthetic Site Electricity Main Total Power\"\n",
+    "            # add more targets if necessary\n",
+    "        ],\n",
+    "        \"dir_data\": \"tests/data/Synthetic Site\",\n",
+    "    }\n",
+    "    # add more entries if necessary\n",
+    "]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# read configs file for training\n",
+    "with open(\n",
+    "    PROJECT_DIRECTORY / \"wattile\" / \"configs\" / \"configs.json\", \"r\"\n",
+    ") as f:\n",
+    "    configs = json.load(f)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "\"\"\"\n",
+    "run model training and validation for the entire parametric space\n",
+    "\"\"\"\n",
+    "for entry in parametric_space:\n",
+    "\n",
+    "    # set building name\n",
+    "    building = entry[\"building\"]\n",
+    "\n",
+    "    # set data directory\n",
+    "    data_dir = entry[\"dir_data\"]\n",
+    "\n",
+    "    for target in entry[\"targets\"]:\n",
+    "\n",
+    "        \"\"\"\n",
+    "        model training\n",
+    "        \"\"\"\n",
+    "        logger.info(\n",
+    "            \"training for building = {} and target = {}\".format(building, target)\n",
+    "        )\n",
+    "\n",
+    "        # override configs params based on the parametric space information\n",
+    "        configs[\"data_input\"][\"data_dir\"] = str(PROJECT_DIRECTORY / data_dir)\n",
+    "        configs[\"data_input\"][\"data_config\"] = \"{} Config.json\".format(building)\n",
+    "        configs[\"data_input\"][\"target_var\"] = target\n",
+    "        exp_dir = PROJECT_DIRECTORY / building / target\n",
+    "        configs[\"data_output\"][\"exp_dir\"] = str(exp_dir)\n",
+    "\n",
+    "        # create results folder\n",
+    "        init_logging(local_results_dir=configs[\"data_output\"][\"exp_dir\"])\n",
+    "\n",
+    "        # read data\n",
+    "        data = read_dataset_from_file(configs)\n",
+    "\n",
+    "        # prepare data for training\n",
+    "        train_df, val_df = prep_for_rnn(configs, data)\n",
+    "        \n",
+    "        # create model\n",
+    "        model = ModelFactory.create_model(configs)\n",
+    "        \n",
+    "        # train model\n",
+    "        results = model.train(train_df, val_df)\n",
+    "\n",
+    "        \"\"\"\n",
+    "        model validation (original data used)\n",
+    "        \"\"\"\n",
+    "        logger.info(\n",
+    "            \"validating for building = {} and target = {}\".format(building, target)\n",
+    "        )\n",
+    "\n",
+    "        # reset config param for validation\n",
+    "        configs[\"data_output\"][\"plot_comparison_portion_start\"] = 0.0\n",
+    "        configs[\"data_output\"][\"plot_comparison_portion_end\"] = 1.0\n",
+    "        configs[\"learning_algorithm\"][\"use_case\"] = \"validation\"\n",
+    "\n",
+    "        # prepare data for validation\n",
+    "        _, val_df = prep_for_rnn(configs, data)\n",
+    "\n",
+    "        # validate model\n",
+    "        results = model.validate(val_df)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "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": "3.8.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}