circomlib-ml/models/globalAveragePooling2D.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "p = 21888242871839275222246405745257275088548364400416034343698204186575808495617"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.layers import Input, GlobalAveragePooling2D\n",
    "from tensorflow.keras import Model\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs = Input(shape=(5,5,3))\n",
    "out = GlobalAveragePooling2D()(inputs)\n",
    "model = Model(inputs, out)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " input_1 (InputLayer)        [(None, 5, 5, 3)]         0         \n",
      "                                                                 \n",
      " global_average_pooling2d (G  (None, 3)                0         \n",
      " lobalAveragePooling2D)                                          \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 0\n",
      "Trainable params: 0\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[[0.20867486, 0.38680755, 0.84218508],\n",
       "         [0.23655331, 0.33771458, 0.73516473],\n",
       "         [0.49345271, 0.95094652, 0.25402692],\n",
       "         [0.22771833, 0.97688694, 0.52917136],\n",
       "         [0.5871173 , 0.50441061, 0.97392083]],\n",
       "\n",
       "        [[0.93934312, 0.52666508, 0.31051829],\n",
       "         [0.2163095 , 0.79177499, 0.3108483 ],\n",
       "         [0.53926143, 0.15753146, 0.99773704],\n",
       "         [0.12234007, 0.20568095, 0.11838809],\n",
       "         [0.9248088 , 0.52638782, 0.81404877]],\n",
       "\n",
       "        [[0.01465677, 0.32765939, 0.74282836],\n",
       "         [0.6800781 , 0.40869424, 0.62145002],\n",
       "         [0.67374829, 0.81617885, 0.39987386],\n",
       "         [0.82099264, 0.35918735, 0.47107381],\n",
       "         [0.83104015, 0.83004572, 0.28737773]],\n",
       "\n",
       "        [[0.74027671, 0.85697829, 0.49504698],\n",
       "         [0.94596904, 0.25070827, 0.22236492],\n",
       "         [0.00357426, 0.35882451, 0.32972314],\n",
       "         [0.57254891, 0.86380467, 0.30862848],\n",
       "         [0.93720522, 0.4496124 , 0.74115158]],\n",
       "\n",
       "        [[0.12640468, 0.76330103, 0.35499368],\n",
       "         [0.37773597, 0.016954  , 0.43058637],\n",
       "         [0.94290805, 0.06019639, 0.95692684],\n",
       "         [0.09562172, 0.61791084, 0.47187214],\n",
       "         [0.67092949, 0.27421069, 0.85342606]]]])"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = np.random.rand(1,5,5,3)\n",
    "X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 36ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-10-23 20:10:13.674664: W tensorflow/tsl/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[0.5171708 , 0.5047629 , 0.54293334]], dtype=float32)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y = model.predict(X)\n",
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_in = [[[int(X[0][i][j][k] * 1e36) for k in range(3)] for j in range(5)] for i in range(5)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def GlobalAveragePooling2DInt(nRows, nCols, nChannels, input):\n",
    "    Input = [[[str(input[i][j][k] % p) for k in range(nChannels)] for j in range(nCols)] for i in range(nRows)]\n",
    "    out = [0 for _ in range(nChannels)]\n",
    "    remainder = [None for _ in range(nChannels)]\n",
    "    for k in range(nChannels):\n",
    "        for i in range(nRows):\n",
    "            for j in range(nCols):\n",
    "                out[k] += input[i][j][k]\n",
    "        remainder[k] = str(out[k] % (nRows * nCols))\n",
    "        out[k] = str(out[k] // (nRows * nCols) % p)\n",
    "    return Input, out, remainder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(['517170777415975145178424005003973754',\n",
       "  '504762926219743484887371893374996971',\n",
       "  '542933334573104965421804807186892718'],\n",
       " ['22', '13', '2'])"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_in, out, remainder = GlobalAveragePooling2DInt(5, 5, 3, X_in)\n",
    "out, remainder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "in_json = {\n",
    "    \"in\": X_in,\n",
    "    \"out\": out,\n",
    "    \"remainder\": remainder\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(\"globalAveragePooling2D_input.json\", \"w\") as f:\n",
    "    json.dump(in_json, f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "sklearn",
   "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": "3.9.16"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}