circomlib-ml/models/dense.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, Dense\n",
    "from tensorflow.keras import Model\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs = Input(shape=(20,))\n",
    "out = Dense(10)(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, 20)]              0         \n",
      "                                                                 \n",
      " dense (Dense)               (None, 10)                210       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 210\n",
      "Trainable params: 210\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.91709806, 0.82830839, 0.93914066, 0.57037095, 0.04271652,\n",
       "        0.35534695, 0.29179199, 0.80227101, 0.65690956, 0.59359125,\n",
       "        0.57083799, 0.64906287, 0.08615951, 0.20494363, 0.98687436,\n",
       "        0.70022373, 0.8282763 , 0.38845018, 0.1025627 , 0.46584396]])"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = np.random.rand(1,20)\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 20:08:46.369862: W tensorflow/tsl/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[-0.9478299 , -0.2901961 ,  1.2173429 ,  0.9856129 ,  0.44817972,\n",
       "         0.8500049 ,  1.2243729 ,  1.1230452 , -0.8100219 ,  0.65824366]],\n",
       "      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*1e36) for x in X[0]]\n",
    "weights = [[None for _ in range(10)] for _ in range(20)]\n",
    "for i in range(20):\n",
    "    for j in range(10):\n",
    "        weights[i][j] = int(model.get_weights()[0][i][j]*1e36)\n",
    "bias = [int(b*1e72) for b in model.get_weights()[1]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def DenseInt(nInputs, nOutputs, n, input, weights, bias):\n",
    "    Input = [str(input[i] % p) for i in range(nInputs)]\n",
    "    Weights = [[str(weights[i][j] % p) for j in range(nOutputs)] for i in range(nInputs)]\n",
    "    Bias = [str(bias[i] % p) for i in range(nOutputs)]\n",
    "    out = [0 for _ in range(nOutputs)]\n",
    "    remainder = [None for _ in range(nOutputs)]\n",
    "    for j in range(nOutputs):\n",
    "        for i in range(nInputs):\n",
    "            out[j] += input[i] * weights[i][j]\n",
    "        out[j] += bias[j]\n",
    "        remainder[j] = str(out[j] % n)\n",
    "        out[j] = str(out[j] // n % p)\n",
    "    return Input, Weights, Bias, out, remainder\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(['21888242871839275222246405745257275088547416570344934017690618098050994599407',\n",
       "  '21888242871839275222246405745257275088548074204322728883645171911568575253269',\n",
       "  '1217342773824182708964164664298235747',\n",
       "  '985612918881516198369842470059289901',\n",
       "  '448179765935532159653151439285979820',\n",
       "  '850005025728080684762697042708611430',\n",
       "  '1224373017260726380443809993782617513',\n",
       "  '1123045202502242457956309189904120564',\n",
       "  '21888242871839275222246405745257275088547554378545433689485043918282349046950',\n",
       "  '658243650814707862463466520227993190'],\n",
       " ['51492294878586576180273547728388096',\n",
       "  '777789140836427041572477556551057408',\n",
       "  '890475291130181473765283908913463296',\n",
       "  '963682024802736049277914862344732672',\n",
       "  '180838407593997560156976766263492608',\n",
       "  '458215330546393498330258578539020288',\n",
       "  '738904904497555276279419263936102400',\n",
       "  '770453107465054933435045502301241344',\n",
       "  '616225701137188004915996184419500032',\n",
       "  '106851476518473341575934978717384704'])"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_in, weights, bias, out, remainder = DenseInt(20, 10, 10**36, X_in, weights, bias)\n",
    "out, remainder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "in_json = {\n",
    "    \"in\": X_in,\n",
    "    \"weights\": weights,\n",
    "    \"bias\": bias,\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(\"dense_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"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}