Merge pull request #4 from lincc-frameworks/awo/scaffolding-for-extra…

…ctor-and-classifier

Rough scaffolding for feature extractor and classifier.

Using_RESSPECT_Walkthrough.ipynb

@@ -0,0 +1,145 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "3d07c5ba-09e5-4d50-ad87-07dcd0f481f0",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# https://resspect.readthedocs.io/en/latest/"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "dab55b9a-e3dd-4e87-b849-fa5a62b36989",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from resspect import fit_snpcc"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "88c74cf5-bfe5-4685-9335-bdfe8c63f051",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "path_to_data_dir = 'data/SIMGEN_PUBLIC_DES/'            # raw data directory \n",
+    "features_file = 'results/Malanchev.csv'                     # output file \n",
+    "feature_extractor = 'malanchev'\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "360ecf08-9c52-4075-8617-4deb1959bb94",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "fit_snpcc(path_to_data_dir=path_to_data_dir, features_file=features_file, feature_extractor=feature_extractor)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "9df62bef-2588-4fd4-8039-41cfd9bb91bc",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from resspect.learn_loop import  learn_loop\n",
+    "\n",
+    "nloops = 1000                                  # number of iterations\n",
+    "method = 'malanchev'                               # only option in v1.0\n",
+    "ml = 'RandomForest'                            # classifier\n",
+    "strategy = 'UncSampling'                       # learning strategy\n",
+    "input_file = 'results/Malanchev.csv'               # input features file\n",
+    "metric = 'results/metrics.csv'                 # output metrics file\n",
+    "queried = 'results/queried.csv'                # output query file\n",
+    "train = 'original'                             # initial training\n",
+    "batch = 1                                      # size of batch\n",
+    "\n",
+    "learn_loop(nloops=nloops, features_method=method, classifier=ml,\n",
+    "           strategy=strategy, path_to_features=input_file, output_metrics_file=metric,\n",
+    "           output_queried_file=queried, training=train, batch=batch)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "ef3b8d05-d53c-402b-836c-f5b6a2efc0f4",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "nloops = 1000                                  # number of iterations\n",
+    "method = 'malanchev'                               # only option in v1.0\n",
+    "ml = 'RandomForest'                            # classifier\n",
+    "strategy = 'RandomSampling'                       # learning strategy\n",
+    "input_file = 'results/Malanchev.csv'               # input features file\n",
+    "metric = 'results/metrics_random.csv'                 # output metrics file\n",
+    "queried = 'results/queried_random.csv'                # output query file\n",
+    "train = 'original'                             # initial training\n",
+    "batch = 1                                      # size of batch\n",
+    "\n",
+    "learn_loop(nloops=nloops, features_method=method, classifier=ml,\n",
+    "           strategy=strategy, path_to_features=input_file, output_metrics_file=metric,\n",
+    "           output_queried_file=queried, training=train, batch=batch)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "3ffa34f1-64f5-46ea-93f7-f8935cba1cc6",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from resspect.plot_results import Canvas\n",
+    "\n",
+    "# define parameters\n",
+    "path_to_files = ['results/metrics.csv','results/metrics_random.csv']\n",
+    "strategies_list = ['UncSampling', 'RandomSampling']\n",
+    "output_plot = 'plots/metrics.png'\n",
+    "\n",
+    "#Initiate the Canvas object, read and plot the results for\n",
+    "# each metric and strategy.\n",
+    "cv = Canvas()\n",
+    "cv.load_metrics(path_to_files=path_to_files,\n",
+    "                   strategies_list=strategies_list)\n",
+    "cv.set_plot_dimensions()\n",
+    "cv.plot_metrics(output_plot_file=output_plot,\n",
+    "                   strategies_list=strategies_list)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "1f88d53b-1591-4aed-8ee1-35904d8b23b2",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "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": "3.10.8"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

docs/notebooks.rst

@@ -3,4 +3,5 @@ Notebooks
 
 .. toctree::
 
-    Introducing Jupyter Notebooks <notebooks/intro_notebook>
+    Load pre-trained models <notebooks/use_resspect_to_load_pretrained_model>
+    Use LAISS in RESSPECT <notebooks/use_resspect_learn_loop_with_laiss>