get_matrix_from_PCA.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import faiss\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# training data\n",
    "xt = np.random.rand(1000, 20).astype('float32')\n",
    "# test data\n",
    "x = np.random.rand(10, 20).astype('float32')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# make the PCA matrix\n",
    "pca = faiss.PCAMatrix(20, 10)\n",
    "pca.train(xt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# apply it to test data\n",
    "yref = pca.apply_py(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# extract matrix + bias from the PCA object\n",
    "# works for any linear transform (OPQ, random rotation, etc.)\n",
    "b = faiss.vector_to_array(pca.b)\n",
    "A = faiss.vector_to_array(pca.A).reshape(pca.d_out, pca.d_in)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# apply transformation\n",
    "ynew = x @ A.T + b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 14,
     "metadata": {
      "bento_obj_id": "140586613660256"
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# are the vectors the same?\n",
    "np.allclose(yref, ynew)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "bento_stylesheets": {
   "bento/extensions/flow/main.css": true,
   "bento/extensions/kernel_selector/main.css": true,
   "bento/extensions/kernel_ui/main.css": true,
   "bento/extensions/new_kernel/main.css": true,
   "bento/extensions/system_usage/main.css": true,
   "bento/extensions/theme/main.css": true
  },
  "disseminate_notebook_id": {
   "notebook_id": "368435637443095"
  },
  "disseminate_notebook_info": {
   "bento_version": "20190826-030256",
   "description": "",
   "hide_code": false,
   "hipster_group": "",
   "kernel_build_info": {
    "error": "The file located at '/data/users/matthijs/fbsource/fbcode/bento/kernels/TARGETS' could not be found."
   },
   "no_uii": true,
   "notebook_number": "138918",
   "others_can_edit": false,
   "reviewers": "",
   "revision_id": "544712462932444",
   "tags": "",
   "tasks": "",
   "title": "get_matrix_from_PCA"
  },
  "kernelspec": {
   "display_name": "pytorch",
   "language": "python",
   "name": "bento_kernel_pytorch"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.3rc1+"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}