Datasets:
sa8
/
zkml-github-repos-truncated

Dataset card Data Studio Files Community
1
text
stringlengths
1
2.05k
return out def ReLUInt(nRows, nCols, nChannels, input): out = [[[str(max(int(input[i][j][k]), 0)) for k in range(nChannels)] for j in range(nCols)] for i in range(nRows)] return out def ArgMaxInt(input): return [input.index(str(max(int(input[i]) for i in range(len(input)))))]
""" Transpile a Keras model to a CIRCOM circuit. Usage: main.py <model.h5> [-o <output>] [--raw] [-d <decimals>] main.py (-h | --help) Options: -h --help Show this screen. -o <output> --output=<output> Output directory [default: output]. --raw Output raw model outputs instead of the argmax of outputs [default: False]. -d <decimals> --decimals=<decimals> Number of decimals for model precision [default: 18]. """ from docopt import docopt from keras2circom import circom, transpiler def main(): """ Main entry point of the app """ args = docopt(__doc__) circom.dir_parse('node_modules/circomlib-ml/circuits/', skips=['util.circom', 'circomlib-matrix', 'circomlib', 'crypto']) transpiler.transpile(args['<model.h5>'], args['--output'], args['--raw'], args['--decimals']) if __name__ == "__main__": """ This is executed when run from the command line """ main()
{ "cells": [ { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ " "from tensorflow.keras.layers
import (\n", " Input,\n", " Activation,\n", " AveragePooling2D,\n", " BatchNormalization,\n", " Conv2D,\n", " Dense,\n", " Dropout,\n", " Flatten,\n", " GlobalAveragePooling2D,\n", " GlobalMaxPooling2D,\n", " MaxPooling2D,\n", " ReLU,\n", " Softmax,\n", " )\n", "from tensorflow.keras
import Model\n", "from tensorflow.keras.datasets
import mnist\n", "from tensorflow.keras.utils
import to_categorical\n", "
import numpy as np\n", "
import matplotlib.pyplot as plt\n", "
import tensorflow as tf" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ " "(X_train, y_train), (X_test, y_test) = mnist.load_data()" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ " "y_train = to_categorical(y_train)\n", "y_test = to_categorical(y_test)" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ " "X_train = X_train.reshape(X_train.shape[0], 28, 28, 1)\n", "X_test = X_test.reshape(X_test.shape[0], 28, 28, 1)\n", "\n", " "X_train = X_train.astype('float32')\n", "X_test = X_test.astype('float32')\n", "X_train /= 255.0\n", "X_test /= 255.0" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "inputs = Input(shape=(28,28,1))\n", "out = Conv2D(4, 3, use_bias=False)(inputs)\n", "out = BatchNormalization()(out)\n", "out = Activation('relu')(out)\n", "out = MaxPooling2D()(out)\n", "out = Conv2D(8, 3, use_bias=True, strides=2)(out)\n", "out = ReLU()(out)\n", "out = AveragePooling2D()(out)\n", "out = Flatten()(out)\n", " "out = Dense(10, activation=\"softmax\")(out)\n", "model = Model(inputs, out)" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Model: \"model\"\n", "_________________________________________________________________\n", " Layer (type) Output Shape Param "=================================================================\n", " input_1 (InputLayer) [(None, 28, 28, 1)] 0 \n", " \n", " conv2d (Conv2D) (None, 26, 26, 4) 36 \n", "
\n", " batch_normalization (BatchN (None, 26, 26, 4) 16 \n", " ormalization) \n", " \n", " activation (Activation) (None, 26, 26, 4) 0 \n", " \n", " max_pooling2d (MaxPooling2D (None, 13, 13, 4) 0 \n", " ) \n", " \n", " conv2d_1 (Conv2D) (None, 6, 6, 8) 296 \n", " \n", " re_lu (ReLU) (None, 6, 6, 8) 0 \n", " \n", " average_pooling2d (AverageP (None, 3, 3, 8) 0 \n", " ooling2D) \n", " \n", " flatten (Flatten) (None, 72) 0 \n", " \n", " dense (Dense) (None, 10) 730 \n", " \n", "=================================================================\n", "Total params: 1,078\n", "Trainable params: 1,070\n", "Non-trainable params: 8\n", "_________________________________________________________________\n" ] } ], "source": [ "model.summary()" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "model.compile(\n", " loss='categorical_crossentropy',\n", " optimizer='adam',\
n", " metrics=['acc']\n", " )" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1/15\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "2023-11-26 21:47:52.776729: W tensorflow/core/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "1875/1875 [==============================] - 11s 6ms/step - loss: 0.5203 - acc: 0.8386 - val_loss: 0.2099 - val_acc: 0.9363\n", "Epoch 2/15\n", "1875/1875 [==============================] - 11s 6ms/step - loss: 0.1926 - acc: 0.9419 - val_loss: 0.1497 - val_acc: 0.9543\n", "Epoch 3/15\n", "1875/1875 [==============================] - 10s 5ms/step - loss: 0.1551 - acc: 0.9522 - val_loss: 0.1263 - val_acc: 0.9591\n", "Epoch 4/15\n", "1875/1875 [==============================] - 10s 5ms/step - loss: 0.1361 - acc: 0.9580 - val_loss: 0.1139 - val_acc: 0.9628\n", "Epoch 5/15\n", "1875/1875 [==============================] - 10s 5ms/step - loss: 0.1253 - acc: 0.9617 - val_loss: 0.1031 - val_acc: 0.9679\n", "Epoch 6/15\n", "1875/1875 [==============================] - 11s 6ms/step - loss: 0.1168 - acc: 0.9636 - val_loss: 0.0976 - val_acc: 0.9697\n", "Epoch 7/15\n", "1875/1875 [==============================] - 10s 5ms/step - loss: 0.1113 - acc: 0.9650 - val_loss: 0.0923 - val_acc: 0.9711\n", "Epoch 8/15\n", "1875/1875 [==============================] - 10s 5ms/step - loss: 0.1072 - acc: 0.9673 - val_loss: 0.0884 - val_acc: 0.9732\n", "Epoch 9/15\n", "1875/1875 [==============================] - 12s 7ms/step - loss: 0.1026 - acc: 0.9683 - val_loss: 0.0879 - val_acc: 0.9725\n", "Epoch 10/15\n", "1875/1875 [==============================] - 11s 6ms/step - loss: 0.0999 - acc: 0.
9691 - val_loss: 0.0928 - val_acc: 0.9719\n", "Epoch 11/15\n", "1875/1875 [==============================] - 10s 5ms/step - loss: 0.0968 - acc: 0.9702 - val_loss: 0.0954 - val_acc: 0.9699\n", "Epoch 12/15\n", "1875/1875 [==============================] - 10s 5ms/step - loss: 0.0945 - acc: 0.9706 - val_loss: 0.0841 - val_acc: 0.9740\n", "Epoch 13/15\n", "1875/1875 [==============================] - 10s 5ms/step - loss: 0.0926 - acc: 0.9718 - val_loss: 0.0826 - val_acc: 0.9748\n", "Epoch 14/15\n", "1875/1875 [==============================] - 10s 5ms/step - loss: 0.0893 - acc: 0.9723 - val_loss: 0.0803 - val_acc: 0.9751\n", "Epoch 15/15\n", "1875/1875 [==============================] - 10s 5ms/step - loss: 0.0892 - acc: 0.9723 - val_loss: 0.0767 - val_acc: 0.9757\n" ] }, { "data": { "text/plain": [ "<keras.callbacks.History at 0x177842d60>" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.fit(X_train, y_train, epochs=15, batch_size=32, validation_data=(X_test, y_test))" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [], "source": [ "model.save('model.h5')" ] } ], "metadata": { "kernelspec": { "display_name": "keras2circom", "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": "3.9.13" }, "orig_nbformat": 4, "vscode": { "interpreter": { "hash": "71414dc221f26c27f268040756e42b4f7499507456a67f7434828e3314a20678" } } }, "nbformat": 4, "nbformat_minor": 2 }
{ "cells": [ { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "!cd .. && python main.py models/model.h5" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "
import sys\n", "
import os\n", " "sys.path.append(os.path.dirname((os.getcwd())))\n", "from output.circuit
import inference\n", "
import json" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [], "source": [ "with open('../output/circuit.json') as f:\n", " circuit = json.load(f)\n", "\n", "with open('y_test.json') as f:\n", " y_test = json.load(f)" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " "
" " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " ] } ], "source": [ "correct = 0\n", "for i in range(10000):\n", " with open(f\"X_test/{i}.json\", \"r\") as f:\n", " input = json.load(f)\n", " out, _ = inference(input, circuit)\n", " correct += 1 if out[0] == y_test[i] else 0\n", " print(f\" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "keras2circom", "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": "3.9.13" } }, "nbformat": 4, "nbformat_minor": 2 }
const chai = require('chai'); const fs = require('fs'); const wasm_tester = require('circom_tester').wasm; const F1Field = require('ffjavascript').F1Field; const Scalar = require('ffjavascript').Scalar; exports.p = Scalar.fromString('21888242871839275222246405745257275088548364400416034343698204186575808495617'); const Fr = new F1Field(exports.p); const assert = chai.assert; const exec = require('await-exec'); const input = require('../test/X_test/0.json'); describe('keras2circom test', function () { this.timeout(100000000); describe('softmax output', async () => { it('softmax output test', async () => { await exec('python main.py models/model.h5 && python output/circuit.py output/circuit.json test/X_test/0.json'); const model = JSON.parse(fs.readFileSync('./output/circuit.json')); const output = JSON.parse(fs.readFileSync('./output/output.json')); const INPUT = {...model, ...input, ...output}; const circuit = await wasm_tester('./output/circuit.circom'); const witness = await circuit.calculateWitness(INPUT, true); assert(Fr.eq(Fr.e(witness[0]),Fr.e(1))); assert(Fr.eq(Fr.e(witness[1]),Fr.e(7))); }); }); describe('raw output', async () => { it('raw output test', async () => { await exec('python main.py models/model.h5 --raw && python output/circuit.py output/circuit.json test/X_test/0.json'); const model = JSON.parse(fs.readFileSync('./output/circuit.json')); const output = JSON.parse(fs.readFileSync('./output/output.json')); const INPUT = {...model, ...input, ...output}; const circuit = await wasm_tester('./output/circuit.circom'); const witness = await circuit.calculateWitness(INPUT, true); assert(Fr.eq(Fr.e(witness[0]),Fr.e(1))); }); }); });
{ "cells": [ { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "from tensorflow.keras.datasets import mnist\n", "import json\n", "import numpy as np" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "# load MNIST dataset\n", "_, (X_test, y_test) = mnist.load_data()" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "#normalizing\n", "X_test = X_test.astype('float32')\n", "X_test /= 255.0" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "for i in range(len(X_test)):\n", " X = [str(int(x * float(10**18))) for x in X_test[i].flatten().tolist()]\n", " X = np.array(X).reshape(28, 28, 1).tolist()\n", " with open(f'X_test/{i}.json', 'w') as f:\n", " json.dump({\"in\": X}, f)" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "with open('y_test.json', 'w') as f:\n", " json.dump(y_test.tolist(), f)" ] } ], "metadata": { "kernelspec": { "display_name": "keras2circom", "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": "3.9.13" } }, "nbformat": 4, "nbformat_minor": 2 }
// from 0xZKML/zk-mnist pragma circom 2.0.0; include "./circomlib/comparators.circom"; include "./circomlib/switcher.circom"; template ArgMax (n) { signal input in[n]; signal input out; assert (out < n); component gts[n]; // store comparators component switchers[n+1]; // switcher for comparing maxs component aswitchers[n+1]; // switcher for arg max signal maxs[n+1]; signal amaxs[n+1]; maxs[0] <== in[0]; amaxs[0] <== 0; for(var i = 0; i < n; i++) { gts[i] = GreaterThan(252); // changed to 252 (maximum) for better compatibility switchers[i+1] = Switcher(); aswitchers[i+1] = Switcher(); gts[i].in[1] <== maxs[i]; gts[i].in[0] <== in[i]; switchers[i+1].sel <== gts[i].out; switchers[i+1].L <== maxs[i]; switchers[i+1].R <== in[i]; aswitchers[i+1].sel <== gts[i].out; aswitchers[i+1].L <== amaxs[i]; aswitchers[i+1].R <== i; amaxs[i+1] <== aswitchers[i+1].outL; maxs[i+1] <== switchers[i+1].outL; } out === amaxs[n]; } // component main { public [ out ] } = ArgMax(5); /* INPUT = { "in": ["2","3","1","5","4"], "out": "3" } */
pragma circom 2.0.0; include "./SumPooling2D.circom"; // AveragePooling2D layer, poolSize is required to be equal for both dimensions, might lose precision compared to SumPooling2D template AveragePooling2D (nRows, nCols, nChannels, poolSize, strides) { signal input in[nRows][nCols][nChannels]; signal input out[(nRows-poolSize)\strides+1][(nCols-poolSize)\strides+1][nChannels]; signal input remainder[(nRows-poolSize)\strides+1][(nCols-poolSize)\strides+1][nChannels]; component sumPooling2D = SumPooling2D (nRows, nCols, nChannels, poolSize, strides); for (var i=0; i<nRows; i++) { for (var j=0; j<nCols; j++) { for (var k=0; k<nChannels; k++) { sumPooling2D.in[i][j][k] <== in[i][j][k]; } } } for (var i=0; i<(nRows-poolSize)\strides+1; i++) { for (var j=0; j<(nCols-poolSize)\strides+1; j++) { for (var k=0; k<nChannels; k++) { assert(remainder[i][j][k] < poolSize * poolSize); out[i][j][k] * poolSize * poolSize + remainder[i][j][k] === sumPooling2D.out[i][j][k]; } } } }
pragma circom 2.0.0; // BatchNormalization layer for 2D inputs // a = gamma/(moving_var+epsilon)**.5 // b = beta-gamma*moving_mean/(moving_var+epsilon)**.5 // n = 10 to the power of the number of decimal places template BatchNormalization2D(nRows, nCols, nChannels, n) { signal input in[nRows][nCols][nChannels]; signal input a[nChannels]; signal input b[nChannels]; signal input out[nRows][nCols][nChannels]; signal input remainder[nRows][nCols][nChannels]; for (var i=0; i<nRows; i++) { for (var j=0; j<nCols; j++) { for (var k=0; k<nChannels; k++) { assert(remainder[i][j][k] < n); out[i][j][k] * n + remainder[i][j][k] === a[k]*in[i][j][k]+b[k]; } } } } // component main { public [ out ] } = BatchNormalization2D(1, 1, 1, 1000); /* INPUT = { "in": ["123"], "a": ["234"], "b": ["345678"], "out": ["374"], "remainder": ["460"] } */
pragma circom 2.0.0; include "./circomlib-matrix/matElemMul.circom"; include "./circomlib-matrix/matElemSum.circom"; include "./util.circom"; // Conv1D layer with valid padding // n = 10 to the power of the number of decimal places template Conv1D (nInputs, nChannels, nFilters, kernelSize, strides, n) { signal input in[nInputs][nChannels]; signal input weights[kernelSize][nChannels][nFilters]; signal input bias[nFilters]; signal input out[(nInputs-kernelSize)\strides+1][nFilters]; signal input remainder[(nInputs-kernelSize)\strides+1][nFilters]; component mul[(nInputs-kernelSize)\strides+1][nChannels][nFilters]; component elemSum[(nInputs-kernelSize)\strides+1][nChannels][nFilters]; component sum[(nInputs-kernelSize)\strides+1][nFilters]; for (var i=0; i<(nInputs-kernelSize)\strides+1; i++) { for (var j=0; j<nChannels; j++) { for (var k=0; k<nFilters; k++) { mul[i][j][k] = matElemMul(kernelSize,1); for (var x=0; x<kernelSize; x++) { mul[i][j][k].a[x][0] <== in[i*strides+x][j]; mul[i][j][k].b[x][0] <== weights[x][j][k]; } elemSum[i][j][k] = matElemSum(kernelSize,1); for (var x=0; x<kernelSize; x++) { elemSum[i][j][k].a[x][0] <== mul[i][j][k].out[x][0]; } } } for (var k=0; k<nFilters; k++) { assert (remainder[i][k] < n); sum[i][k] = Sum(nChannels); for (var j=0; j<nChannels; j++) { sum[i][k].in[j] <== elemSum[i][j][k].out; } out[i][k] * n + remainder[i][k] === sum[i][k].out + bias[k]; } } }
pragma circom 2.0.0;
include "./circomlib-matrix/matElemMul.circom";
include "./circomlib-matrix/matElemSum.circom";
include "./util.circom"; template Conv2D (nRows, nCols, nChannels, nFilters, kernelSize, strides, n) { signal input in[nRows][nCols][nChannels]; signal input weights[kernelSize][kernelSize][nChannels][nFilters]; signal input bias[nFilters]; signal input out[(nRows-kernelSize)\strides+1][(nCols-kernelSize)\strides+1][nFilters]; signal input remainder[(nRows-kernelSize)\strides+1][(nCols-kernelSize)\strides+1][nFilters]; component mul[(nRows-kernelSize)\strides+1][(nCols-kernelSize)\strides+1][nChannels][nFilters]; component elemSum[(nRows-kernelSize)\strides+1][(nCols-kernelSize)\strides+1][nChannels][nFilters]; component sum[(nRows-kernelSize)\strides+1][(nCols-kernelSize)\strides+1][nFilters]; for (var i=0; i<(nRows-kernelSize)\strides+1; i++) { for (var j=0; j<(nCols-kernelSize)\strides+1; j++) { for (var k=0; k<nChannels; k++) { for (var m=0; m<nFilters; m++) { mul[i][j][k][m] = matElemMul(kernelSize,kernelSize); for (var x=0; x<kernelSize; x++) { for (var y=0; y<kernelSize; y++) { mul[i][j][k][m].a[x][y] <== in[i*strides+x][j*strides+y][k]; mul[i][j][k][m].b[x][y] <== weights[x][y][k][m]; } } elemSum[i][j][k][m] = matElemSum(kernelSize,kernelSize); for (var x=0; x<kernelSize; x++) { for (var y=0; y<kernelSize; y++) { elemSum[i][j][k][m].a[x][y] <== mul[i][j][k][m].out[x][y]; } } } } for (var m=0; m<nFilters; m++) { assert (remainder[i][j][m] < n); sum[i][j][m] = Sum(nChannels); for (var k=0; k<nChannels; k++) { sum[i][j][m].in[k] <== elemSum[i][j][k][m].out; } out[i][j][m] * n + remainder[i][j][m] ===
sum[i][j][m].out + bias[m]; } } } }
pragma circom 2.0.0;
include "./Conv2D.circom"; template Conv2Dsame (nRows, nCols, nChannels, nFilters, kernelSize, strides, n) { signal input in[nRows][nCols][nChannels]; signal input weights[kernelSize][kernelSize][nChannels][nFilters]; signal input bias[nFilters]; var rowPadding, colPadding; if (nRows % strides == 0) { rowPadding = (kernelSize - strides) > 0 ? (kernelSize - strides) : 0; } else { rowPadding = (kernelSize - (nRows % strides)) > 0 ? (kernelSize - (nRows % strides)) : 0; } if (nCols % strides == 0) { colPadding = (kernelSize - strides) > 0 ? (kernelSize - strides) : 0; } else { colPadding = (kernelSize - (nCols % strides)) > 0 ? (kernelSize - (nCols % strides)) : 0; } signal input out[(nRows+rowPadding-kernelSize)\strides+1][(nCols+colPadding-kernelSize)\strides+1][nFilters]; signal input remainder[(nRows+rowPadding-kernelSize)\strides+1][(nCols+colPadding-kernelSize)\strides+1][nFilters]; component conv2d = Conv2D(nRows+rowPadding, nCols+colPadding, nChannels, nFilters, kernelSize, strides, n); for (var i = rowPadding\2; i < rowPadding\2+nRows; i++) { for (var j = colPadding\2; j < colPadding\2+nCols; j++) { for (var k = 0; k < nChannels; k++) { conv2d.in[i][j][k] <== in[i-rowPadding\2][j-colPadding\2][k]; } } } for (var i = 0; i< rowPadding\2; i++) { for (var j = 0; j < nCols+colPadding; j++) { for (var k = 0; k < nChannels; k++) { conv2d.in[i][j][k] <== 0; } } } for (var i = nRows+rowPadding\2; i < nRows+rowPadding; i++) { for (var j = 0; j < nCols+colPadding; j++) { for (var k = 0; k < nChannels; k++) { conv2d.in[i][j][k] <== 0; } } } for (var i = rowPadding\2; i < nRows+rowPadding\2; i++) { for (var j = 0; j < colPadding\2; j++) { for (var k = 0; k < nChannels; k++) { conv2d.in[i][j][k] <== 0
; } } } for (var i = rowPadding\2; i < nRows+rowPadding\2; i++) { for (var j = nCols+colPadding\2; j < nCols+colPadding; j++) { for (var k = 0; k < nChannels; k++) { conv2d.in[i][j][k] <== 0; } } } for (var i = 0; i < kernelSize; i++) { for (var j = 0; j < kernelSize; j++) { for (var k = 0; k < nChannels; k++) { for (var l = 0; l < nFilters; l++) { conv2d.weights[i][j][k][l] <== weights[i][j][k][l]; } } } } for (var i = 0; i < nFilters; i++) { conv2d.bias[i] <== bias[i]; } for (var i = 0; i < (nRows+rowPadding-kernelSize)\strides+1; i++) { for (var j = 0; j < (nCols+colPadding-kernelSize)\strides+1; j++) { for (var k = 0; k < nFilters; k++) { conv2d.out[i][j][k] <== out[i][j][k]; conv2d.remainder[i][j][k] <== remainder[i][j][k]; } } } }
pragma circom 2.0.0; include "./circomlib-matrix/matMul.circom"; // Dense layer // n = 10 to the power of the number of decimal places template Dense (nInputs, nOutputs, n) { signal input in[nInputs]; signal input weights[nInputs][nOutputs]; signal input bias[nOutputs]; signal input out[nOutputs]; signal input remainder[nOutputs]; component dot[nOutputs]; for (var i=0; i<nOutputs; i++) { assert (remainder[i] < n); dot[i] = matMul(1,nInputs,1); for (var j=0; j<nInputs; j++) { dot[i].a[0][j] <== in[j]; dot[i].b[j][0] <== weights[j][i]; } out[i] * n + remainder[i] === dot[i].out[0][0] + bias[i]; } }
pragma circom 2.1.1;
include "./circomlib/sign.circom";
include "./circomlib/bitify.circom";
include "./circomlib/comparators.circom";
include "./circomlib-matrix/matElemMul.circom";
include "./circomlib-matrix/matElemSum.circom";
include "./util.circom"; template DepthwiseConv2D (nRows, nCols, nChannels, nFilters, kernelSize, strides, n) { var outRows = (nRows-kernelSize)\strides+1; var outCols = (nCols-kernelSize)\strides+1; signal input in[nRows][nCols][nChannels]; signal input weights[kernelSize][kernelSize][nFilters]; signal input bias[nFilters]; signal input remainder[outRows][outCols][nFilters]; signal input out[outRows][outCols][nFilters]; component mul[outRows][outCols][nFilters]; component elemSum[outRows][outCols][nFilters]; var valid_groups = nFilters % nChannels; var filtersPerChannel = nFilters / nChannels; signal groups; groups <== valid_groups; component is_zero = IsZero(); is_zero.in <== groups; is_zero.out === 1; for (var row=0; row<outRows; row++) { for (var col=0; col<outCols; col++) { for (var filterMultiplier=1; filterMultiplier<=filtersPerChannel; filterMultiplier++) { for (var channel=0; channel<nChannels; channel++) { var filter = filterMultiplier*channel; mul[row][col][filter] = matElemMul(kernelSize,kernelSize); for (var x=0; x<kernelSize; x++) { for (var y=0; y<kernelSize; y++) { mul[row][col][filter].a[x][y] <== in[row*strides+x][col*strides+y][channel]; mul[row][col][filter].b[x][y] <== weights[x][y][filter]; } } elemSum[row][col][filter] = matElemSum(kernelSize,kernelSize); for (var x=0; x<kernelSize; x++) { for (var y=0; y<kernelSize; y++) { elemSum[row][col][filter].a[x][y] <== mul[row][col][filter].out[x][y]; } } out[row][col][filter] * n + remainder[row][col][filter] === elemSum[row][col][filter].out + bias[filter]; } }
} } }
pragma circom 2.0.0; // Flatten layer with that accepts a 2D input template Flatten2D (nRows, nCols, nChannels) { signal input in[nRows][nCols][nChannels]; signal input out[nRows*nCols*nChannels]; var idx = 0; for (var i=0; i<nRows; i++) { for (var j=0; j<nCols; j++) { for (var k=0; k<nChannels; k++) { out[idx] === in[i][j][k]; idx++; } } } }
pragma circom 2.0.0; include "./GlobalSumPooling2D.circom"; // GlobalAveragePooling2D layer, might lose precision compared to GlobalSumPooling2D template GlobalAveragePooling2D (nRows, nCols, nChannels) { signal input in[nRows][nCols][nChannels]; signal input out[nChannels]; signal input remainder[nChannels]; component globalSumPooling2D = GlobalSumPooling2D (nRows, nCols, nChannels); for (var i=0; i<nRows; i++) { for (var j=0; j<nCols; j++) { for (var k=0; k<nChannels; k++) { globalSumPooling2D.in[i][j][k] <== in[i][j][k]; } } } for (var k=0; k<nChannels; k++) { assert (remainder[k] < nRows*nCols); out[k] * nRows * nCols + remainder[k] === globalSumPooling2D.out[k]; } }
pragma circom 2.0.0; include "./util.circom"; // GlobalMaxPooling2D layer template GlobalMaxPooling2D (nRows, nCols, nChannels) { signal input in[nRows][nCols][nChannels]; signal input out[nChannels]; component max[nChannels]; for (var k=0; k<nChannels; k++) { max[k] = Max(nRows*nCols); for (var i=0; i<nRows; i++) { for (var j=0; j<nCols; j++) { max[k].in[i*nCols+j] <== in[i][j][k]; } } out[k] === max[k].out; } }
pragma circom 2.0.0; include "./circomlib-matrix/matElemSum.circom"; include "./util.circom"; // GlobalSumPooling2D layer, basically GlobalAveragePooling2D layer with a constant scaling, more optimized for circom template GlobalSumPooling2D (nRows, nCols, nChannels) { signal input in[nRows][nCols][nChannels]; signal output out[nChannels]; component elemSum[nChannels]; for (var k=0; k<nChannels; k++) { elemSum[k] = matElemSum(nRows,nCols); for (var i=0; i<nRows; i++) { for (var j=0; j<nCols; j++) { elemSum[k].a[i][j] <== in[i][j][k]; } } out[k] <== elemSum[k].out; } }
pragma circom 2.0.0; include "./util.circom"; // LeakyReLU layer template LeakyReLU (alpha) { // alpha is 10 times the actual alpha, since usual alpha is 0.2, 0.3, etc. signal input in; signal input out; signal input remainder; component isNegative = IsNegative(); isNegative.in <== in; signal neg; neg <== isNegative.out * alpha * in; out * 10 + remainder === neg + 10 * in * (1 - isNegative.out); }
pragma circom 2.0.0; include "./util.circom"; // MaxPooling2D layer template MaxPooling2D (nRows, nCols, nChannels, poolSize, strides) { signal input in[nRows][nCols][nChannels]; signal input out[(nRows-poolSize)\strides+1][(nCols-poolSize)\strides+1][nChannels]; component max[(nRows-poolSize)\strides+1][(nCols-poolSize)\strides+1][nChannels]; for (var i=0; i<(nRows-poolSize)\strides+1; i++) { for (var j=0; j<(nCols-poolSize)\strides+1; j++) { for (var k=0; k<nChannels; k++) { max[i][j][k] = Max(poolSize*poolSize); for (var x=0; x<poolSize; x++) { for (var y=0; y<poolSize; y++) { max[i][j][k].in[x*poolSize+y] <== in[i*strides+x][j*strides+y][k]; } } out[i][j][k] === max[i][j][k].out; } } } }
pragma circom 2.0.0;
include "./MaxPooling2D.circom"; template MaxPooling2Dsame (nRows, nCols, nChannels, poolSize, strides) { signal input in[nRows][nCols][nChannels]; var rowPadding, colPadding; if (nRows % strides == 0) { rowPadding = (poolSize - strides) > 0 ? (poolSize - strides) : 0; } else { rowPadding = (poolSize - (nRows % strides)) > 0 ? (poolSize - (nRows % strides)) : 0; } if (nCols % strides == 0) { colPadding = (poolSize - strides) > 0 ? (poolSize - strides) : 0; } else { colPadding = (poolSize - (nCols % strides)) > 0 ? (poolSize - (nCols % strides)) : 0; } signal input out[(nRows+rowPadding-poolSize)\strides+1][(nCols+colPadding-poolSize)\strides+1][nChannels]; component max2d = MaxPooling2D(nRows+rowPadding, nCols+colPadding, nChannels, poolSize, strides); for (var i = rowPadding\2; i < rowPadding\2+nRows; i++) { for (var j = colPadding\2; j < colPadding\2+nCols; j++) { for (var k = 0; k < nChannels; k++) { max2d.in[i][j][k] <== in[i-rowPadding\2][j-colPadding\2][k]; } } } for (var i = 0; i< rowPadding\2; i++) { for (var j = 0; j < nCols+colPadding; j++) { for (var k = 0; k < nChannels; k++) { max2d.in[i][j][k] <== 0; } } } for (var i = nRows+rowPadding\2; i< nRows+rowPadding; i++) { for (var j = 0; j < nCols+colPadding; j++) { for (var k = 0; k < nChannels; k++) { max2d.in[i][j][k] <== 0; } } } for (var i = rowPadding\2; i < nRows+rowPadding\2; i++) { for (var j = 0; j < colPadding\2; j++) { for (var k = 0; k < nChannels; k++) { max2d.in[i][j][k] <== 0; } } } for (var i = rowPadding\2; i < nRows+rowPadding\2; i++) { for (var j = nCols+colPadding\2; j < nCols+colPadding; j++) { for (var k = 0; k < nChannels; k++) { max2d.in[i][j][k] <== 0;
} } } for (var i = 0; i < (nRows+rowPadding-poolSize)\strides+1; i++) { for (var j = 0; j < (nCols+colPadding-poolSize)\strides+1; j++) { for (var k = 0; k < nChannels; k++) { max2d.out[i][j][k] <== out[i][j][k]; } } } }
pragma circom 2.1.1; // include "./Conv2D.circom"; include "./circomlib/sign.circom"; include "./circomlib/bitify.circom"; include "./circomlib/comparators.circom"; include "./circomlib-matrix/matElemMul.circom"; include "./circomlib-matrix/matElemSum.circom"; include "./util.circom"; // Pointwise Convolution layer // Note that nFilters must be a multiple of nChannels template PointwiseConv2D (nRows, nCols, nChannels, nFilters, n) { var outRows = nRows; // kernel size and strides are 1 var outCols = nCols; signal input in[nRows][nCols][nChannels]; signal input weights[nChannels][nFilters]; // weights are 2d because kernel_size is 1 signal input bias[nFilters]; signal input out[outRows][outCols][nFilters]; signal input remainder[outRows][outCols][nFilters]; component sum[outRows][outCols][nFilters]; for (var row=0; row<outRows; row++) { for (var col=0; col<outCols; col++) { for (var filter=0; filter<nFilters; filter++) { sum[row][col][filter] = Sum(nChannels); for (var channel=0; channel<nChannels; channel++) { sum[row][col][filter].in[channel] <== in[row][col][channel] * weights[channel][filter]; } out[row][col][filter] * n + remainder[row][col][filter] === sum[row][col][filter].out + bias[filter]; } } } } // component main = PointwiseConv2D(32, 32, 8, 16);
pragma circom 2.0.0; include "./util.circom"; // ReLU layer template ReLU () { signal input in; signal input out; component isPositive = IsPositive(); isPositive.in <== in; out === in * isPositive.out; }
pragma circom 2.0.0; // Reshape layer with that accepts a 1D input template Reshape2D (nRows, nCols, nChannels) { signal input in[nRows*nCols*nChannels]; signal input out[nRows][nCols][nChannels]; for (var i=0; i<nRows; i++) { for (var j=0; j<nCols; j++) { for (var k=0; k<nChannels; k++) { out[i][j][k] === in[i*nCols*nChannels + j*nChannels + k]; } } } }
pragma circom 2.1.1;
include "./PointwiseConv2D.circom";
include "./DepthwiseConv2D.circom"; template SeparableConv2D (nRows, nCols, nChannels, nDepthFilters, nPointFilters, depthKernelSize, strides, n) { var outRows = (nRows-depthKernelSize)\strides+1; var outCols = (nCols-depthKernelSize)\strides+1; signal input in[nRows][nCols][nChannels]; signal input depthWeights[depthKernelSize][depthKernelSize][nDepthFilters]; signal input depthBias[nDepthFilters]; signal input depthRemainder[outRows][outCols][nDepthFilters]; signal input depthOut[outRows][outCols][nDepthFilters]; signal input pointWeights[nChannels][nPointFilters]; signal input pointBias[nPointFilters]; signal input pointRemainder[outRows][outCols][nPointFilters]; signal input pointOut[outRows][outCols][nPointFilters]; component depthConv = DepthwiseConv2D(nRows, nCols, nChannels, nDepthFilters, depthKernelSize, strides, n); component pointConv = PointwiseConv2D(outRows, outCols, nDepthFilters, nPointFilters, n); for (var filter=0; filter<nDepthFilters; filter++) { for (var x=0; x<depthKernelSize; x++) { for (var y=0; y<depthKernelSize; y++) { depthConv.weights[x][y][filter] <== depthWeights[x][y][filter]; } } depthConv.bias[filter] <== depthBias[filter]; } for (var row=0; row < nRows; row++) { for (var col=0; col < nCols; col++) { for (var channel=0; channel < nChannels; channel++) { depthConv.in[row][col][channel] <== in[row][col][channel]; } } } for (var row=0; row < outRows; row++) { for (var col=0; col < outCols; col++) { for (var filter=0; filter < nDepthFilters; filter++) { depthConv.remainder[row][col][filter] <== depthRemainder[row][col][filter]; depthConv.out[row][col][filter] <== depthOut[row][col][filter]; pointConv.in[row][col][filter] <== depthOut[row][col][filter]; } } } for (var filter=0; filter < nPo
intFilters; filter++) { for (var channel=0; channel < nChannels; channel++) { pointConv.weights[channel][filter] <== pointWeights[channel][filter]; } pointConv.bias[filter] <== pointBias[filter]; } for (var row=0; row < outRows; row++) { for (var col=0; col < outCols; col++) { for (var filter=0; filter < nPointFilters; filter++) { pointConv.remainder[row][col][filter] <== pointRemainder[row][col][filter]; pointConv.out[row][col][filter] <== pointOut[row][col][filter]; } } } }
pragma circom 2.0.0; include "./circomlib-matrix/matElemSum.circom"; include "./util.circom"; // SumPooling2D layer, basically AveragePooling2D layer with a constant scaling, more optimized for circom template SumPooling2D (nRows, nCols, nChannels, poolSize, strides) { signal input in[nRows][nCols][nChannels]; signal output out[(nRows-poolSize)\strides+1][(nCols-poolSize)\strides+1][nChannels]; component elemSum[(nRows-poolSize)\strides+1][(nCols-poolSize)\strides+1][nChannels]; for (var i=0; i<(nRows-poolSize)\strides+1; i++) { for (var j=0; j<(nCols-poolSize)\strides+1; j++) { for (var k=0; k<nChannels; k++) { elemSum[i][j][k] = matElemSum(poolSize,poolSize); for (var x=0; x<poolSize; x++) { for (var y=0; y<poolSize; y++) { elemSum[i][j][k].a[x][y] <== in[i*strides+x][j*strides+y][k]; } } out[i][j][k] <== elemSum[i][j][k].out; } } } }
pragma circom 2.0.0; template UpSampling2D(nRows, nCols, nChannels, size) { signal input in[nRows][nCols][nChannels]; signal input out[nRows * size][nCols * size][nChannels]; // nearest neighbor interpolation for (var i = 0; i < nRows; i++) { for (var j = 0; j < nCols; j++) { for (var c = 0; c < nChannels; c++) { for (var k = 0; k < size; k++) { for (var l = 0; l < size; l++) { out[i * size + k][j * size + l][c] === in[i][j][c]; } } } } } }
pragma circom 2.0.0; // Polynomial approximation for the tanh layer // 0.006769816 + 0.554670504 * x - 0.009411195 * x**2 - 0.014187547 * x**3 // 6769816 + 554670504 * x - 9411195 * x**2 - 14187547 * x**3 // 6769816 + x * (554670504 - 9411195 * x - 14187547 * x**2) // n = 10 to the power of the number of decimal places // out and remainder are from division by n**2 so out has the same number of decimal places as in template Zanh (n) { signal input in; signal input out; signal input remainder; assert(remainder < (n**2) * (10**9)); signal tmp; tmp <== 554670504 * n**2 - 9411195 * n * in - 14187547 * in * in; // log(6769816 * n**3 + in * tmp); out * (n**2) * (10**9) + remainder === 6769816 * n**3 + in * tmp; } // component main { public [ out ] } = Zanh(10**9); /* INPUT = { "in": "123456789", "out": "750775175", "remainder": "35162208611038149371400257" } */
pragma circom 2.0.0; // Poly activation layer: https://arxiv.org/abs/2011.05530 // n = 10 to the power of the number of decimal places // out and remainder are from division by n so out has the same number of decimal places as in template ZeLU (n) { signal input in; signal input out; signal input remainder; assert(remainder < n); signal tmp; tmp <== in * in + n*in; out * n + remainder === tmp; } // component main { public [ out ] } = ZeLU(10**9); /* INPUT = { "in": "123456789", "out": "138698367", "remainder": "750190521" } */