circomlib-ml/models/globalMaxPooling2D.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, GlobalMaxPooling2D\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",
    "x = GlobalMaxPooling2D()(inputs)\n",
    "model = Model(inputs, x)"
   ]
  },
  {
   "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_max_pooling2d (Globa  (None, 3)                0         \n",
      " lMaxPooling2D)                                                  \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([[[[0.91518641, 0.66718006, 0.92376279],\n",
       "         [0.97385149, 0.35848662, 0.25065166],\n",
       "         [0.02434808, 0.66270045, 0.51526436],\n",
       "         [0.97110905, 0.49335089, 0.27623285],\n",
       "         [0.88055993, 0.91070856, 0.89195416]],\n",
       "\n",
       "        [[0.7950447 , 0.42411655, 0.66516519],\n",
       "         [0.18674268, 0.3046312 , 0.77807526],\n",
       "         [0.13333453, 0.68076544, 0.64069414],\n",
       "         [0.63039814, 0.71725918, 0.74384312],\n",
       "         [0.48789065, 0.68079997, 0.25869622]],\n",
       "\n",
       "        [[0.55852658, 0.78138444, 0.0772444 ],\n",
       "         [0.71960766, 0.01860611, 0.63859032],\n",
       "         [0.04100894, 0.007163  , 0.28648401],\n",
       "         [0.70371242, 0.8565901 , 0.73254654],\n",
       "         [0.35201173, 0.3338802 , 0.83269692]],\n",
       "\n",
       "        [[0.31146493, 0.11242401, 0.46909255],\n",
       "         [0.785379  , 0.69905536, 0.99196427],\n",
       "         [0.29254832, 0.04347593, 0.40404928],\n",
       "         [0.64393514, 0.6579046 , 0.44890337],\n",
       "         [0.25879095, 0.64296721, 0.65792656]],\n",
       "\n",
       "        [[0.7972691 , 0.77522241, 0.02028976],\n",
       "         [0.71408815, 0.2214879 , 0.07804482],\n",
       "         [0.65261239, 0.62851164, 0.12214903],\n",
       "         [0.31611407, 0.18022595, 0.97735959],\n",
       "         [0.57391523, 0.8818251 , 0.06020382]]]])"
      ]
     },
     "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 31ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-10-23 20:16:54.369715: W tensorflow/tsl/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[0.9738515 , 0.91070855, 0.9919643 ]], 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 GlobalMaxPooling2DInt(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 = [max(input[i][j][k] for i in range(nRows) for j in range(nCols)) for k in range(nChannels)]\n",
    "    return Input, out"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[973851490313537338484516198430015488,\n",
       " 910708561343324144695836121136889856,\n",
       " 991964273065568131927416012428804096]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_in, out = GlobalMaxPooling2DInt(5,5,3,X_in)\n",
    "out"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "in_json = {\n",
    "    \"in\": X_in,\n",
    "    \"out\": out\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(\"globalMaxPooling2D_input.json\", \"w\") as f:\n",
    "    json.dump(in_json, f)"
   ]
  }
 ],
 "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
}