{ "cells": [ { "attachments": {}, "cell_type": "markdown", "metadata": { "id": "hRTa3Ee15WsJ" }, "source": [ "# Load and Save a model\n", "\n", "This notebook loads a pretrained model and saves it.\n", "The goal is to upload the model later to huggingface and test the model online.\n", "\n" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## Import Libraries" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "execution": { "iopub.execute_input": "2022-12-14T02:23:01.733281Z", "iopub.status.busy": "2022-12-14T02:23:01.732884Z", "iopub.status.idle": "2022-12-14T02:23:04.483275Z", "shell.execute_reply": "2022-12-14T02:23:04.482452Z" }, "id": "TqOt6Sv7AsMi" }, "outputs": [], "source": [ "import matplotlib.pyplot as plt\n", "import numpy as np\n", "import os\n", "import tensorflow as tf\n", "from pathlib import Path\n", "import imghdr" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "execution": { "iopub.execute_input": "2022-12-14T02:23:17.402521Z", "iopub.status.busy": "2022-12-14T02:23:17.401942Z", "iopub.status.idle": "2022-12-14T02:23:17.405507Z", "shell.execute_reply": "2022-12-14T02:23:17.404891Z" }, "id": "cO0HM9JAQUFq" }, "outputs": [ { "data": { "text/plain": [ "array([-1., 1.], dtype=float32)" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "preprocess_input = tf.keras.applications.resnet_v2.preprocess_input\n", "\n", "# example\n", "preprocess_input(np.asarray([0, 255]))" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": { "id": "xnr81qRMzcs5" }, "source": [ "Note: Alternatively, you could rescale pixel values from `[0, 255]` to `[-1, 1]` using `tf.keras.layers.Rescaling`.
\n", "It is recommended to use preprocess_input, because when you change an other base application, you don't have to change the scaling ratio." ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "execution": { "iopub.execute_input": "2022-12-14T02:23:17.408530Z", "iopub.status.busy": "2022-12-14T02:23:17.408086Z", "iopub.status.idle": "2022-12-14T02:23:17.411907Z", "shell.execute_reply": "2022-12-14T02:23:17.411338Z" }, "id": "R2NyJn4KQMux" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Metal device set to: Apple M1 Max\n", "\n", "systemMemory: 32.00 GB\n", "maxCacheSize: 10.67 GB\n", "\n" ] }, { "data": { "text/plain": [ "" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "rescale = tf.keras.layers.Rescaling(1./127.5, offset=-1)\n", "\n", "# example\n", "rescale(np.asarray([0, 255]))" ] }, { "cell_type": "markdown", "metadata": { "id": "Wz7qgImhTxw4" }, "source": [ "Note: If using other `tf.keras.applications`, be sure to check the API doc to determine if they expect pixels in `[-1, 1]` or `[0, 1]`, or use the included `preprocess_input` function." ] }, { "attachments": {}, "cell_type": "markdown", "metadata": { "id": "OkH-kazQecHB" }, "source": [ "## Download the model from the pre-trained EfficientNet\n", "You will create the base model from the **EfficientNetB4** model. This is pre-trained on the ImageNet dataset, a large dataset consisting of 1.4M images and 1000 classes. ImageNet is a research training dataset with a wide variety of categories like `jackfruit` and `syringe`. This base of knowledge will help us classify cats and dogs from our specific dataset.\n", "\n", "First, instantiate a EfficientNetB4 model pre-loaded with weights trained on ImageNet. By specifying the **include_top=True** argument, you load a network that does include the classification layers at the top, which is ideal for feature extraction." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "IMG_SHAPE = (224, 224, 3)" ] }, { "cell_type": "code", "execution_count": 15, "metadata": { "execution": { "iopub.execute_input": "2022-12-14T02:23:17.415060Z", "iopub.status.busy": "2022-12-14T02:23:17.414630Z", "iopub.status.idle": "2022-12-14T02:23:18.869000Z", "shell.execute_reply": "2022-12-14T02:23:18.868159Z" }, "id": "19IQ2gqneqmS" }, "outputs": [], "source": [ "# Create the base model from the pre-trained model EfficientNetB4\n", "model = tf.keras.applications.MobileNetV3Large(input_shape=IMG_SHAPE,\n", " \n", " # include_top=True :Includes the fully connected layers for predictions \n", " include_top=True,\n", "\n", " # weights from the imagenet challenge.\n", " weights='imagenet')" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [], "source": [ "model.build(input_shape=IMG_SHAPE)" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Model: \"MobilenetV3large\"\n", "__________________________________________________________________________________________________\n", " Layer (type) Output Shape Param # Connected to \n", "==================================================================================================\n", " input_3 (InputLayer) [(None, 224, 224, 3 0 [] \n", " )] \n", " \n", " rescaling_1 (Rescaling) (None, 224, 224, 3) 0 ['input_3[0][0]'] \n", " \n", " Conv (Conv2D) (None, 112, 112, 16 432 ['rescaling_1[0][0]'] \n", " ) \n", " \n", " Conv/BatchNorm (BatchNormaliza (None, 112, 112, 16 64 ['Conv[0][0]'] \n", " tion) ) \n", " \n", " tf.__operators__.add (TFOpLamb (None, 112, 112, 16 0 ['Conv/BatchNorm[0][0]'] \n", " da) ) \n", " \n", " re_lu (ReLU) (None, 112, 112, 16 0 ['tf.__operators__.add[0][0]'] \n", " ) \n", " \n", " tf.math.multiply (TFOpLambda) (None, 112, 112, 16 0 ['re_lu[0][0]'] \n", " ) \n", " \n", " multiply (Multiply) (None, 112, 112, 16 0 ['Conv/BatchNorm[0][0]', \n", " ) 'tf.math.multiply[0][0]'] \n", " \n", " expanded_conv/depthwise (Depth (None, 112, 112, 16 144 ['multiply[0][0]'] \n", " wiseConv2D) ) \n", " \n", " expanded_conv/depthwise/BatchN (None, 112, 112, 16 64 ['expanded_conv/depthwise[0][0]']\n", " orm (BatchNormalization) ) \n", " \n", " re_lu_1 (ReLU) (None, 112, 112, 16 0 ['expanded_conv/depthwise/BatchNo\n", " ) rm[0][0]'] \n", " \n", " expanded_conv/project (Conv2D) (None, 112, 112, 16 256 ['re_lu_1[0][0]'] \n", " ) \n", " \n", " expanded_conv/project/BatchNor (None, 112, 112, 16 64 ['expanded_conv/project[0][0]'] \n", " m (BatchNormalization) ) \n", " \n", " expanded_conv/Add (Add) (None, 112, 112, 16 0 ['multiply[0][0]', \n", " ) 'expanded_conv/project/BatchNorm\n", " [0][0]'] \n", " \n", " expanded_conv_1/expand (Conv2D (None, 112, 112, 64 1024 ['expanded_conv/Add[0][0]'] \n", " ) ) \n", " \n", " expanded_conv_1/expand/BatchNo (None, 112, 112, 64 256 ['expanded_conv_1/expand[0][0]'] \n", " rm (BatchNormalization) ) \n", " \n", " re_lu_2 (ReLU) (None, 112, 112, 64 0 ['expanded_conv_1/expand/BatchNor\n", " ) m[0][0]'] \n", " \n", " expanded_conv_1/depthwise/pad (None, 113, 113, 64 0 ['re_lu_2[0][0]'] \n", " (ZeroPadding2D) ) \n", " \n", " expanded_conv_1/depthwise (Dep (None, 56, 56, 64) 576 ['expanded_conv_1/depthwise/pad[0\n", " thwiseConv2D) ][0]'] \n", " \n", " expanded_conv_1/depthwise/Batc (None, 56, 56, 64) 256 ['expanded_conv_1/depthwise[0][0]\n", " hNorm (BatchNormalization) '] \n", " \n", " re_lu_3 (ReLU) (None, 56, 56, 64) 0 ['expanded_conv_1/depthwise/Batch\n", " Norm[0][0]'] \n", " \n", " expanded_conv_1/project (Conv2 (None, 56, 56, 24) 1536 ['re_lu_3[0][0]'] \n", " D) \n", " \n", " expanded_conv_1/project/BatchN (None, 56, 56, 24) 96 ['expanded_conv_1/project[0][0]']\n", " orm (BatchNormalization) \n", " \n", " expanded_conv_2/expand (Conv2D (None, 56, 56, 72) 1728 ['expanded_conv_1/project/BatchNo\n", " ) rm[0][0]'] \n", " \n", " expanded_conv_2/expand/BatchNo (None, 56, 56, 72) 288 ['expanded_conv_2/expand[0][0]'] \n", " rm (BatchNormalization) \n", " \n", " re_lu_4 (ReLU) (None, 56, 56, 72) 0 ['expanded_conv_2/expand/BatchNor\n", " m[0][0]'] \n", " \n", " expanded_conv_2/depthwise (Dep (None, 56, 56, 72) 648 ['re_lu_4[0][0]'] \n", " thwiseConv2D) \n", " \n", " expanded_conv_2/depthwise/Batc (None, 56, 56, 72) 288 ['expanded_conv_2/depthwise[0][0]\n", " hNorm (BatchNormalization) '] \n", " \n", " re_lu_5 (ReLU) (None, 56, 56, 72) 0 ['expanded_conv_2/depthwise/Batch\n", " Norm[0][0]'] \n", " \n", " expanded_conv_2/project (Conv2 (None, 56, 56, 24) 1728 ['re_lu_5[0][0]'] \n", " D) \n", " \n", " expanded_conv_2/project/BatchN (None, 56, 56, 24) 96 ['expanded_conv_2/project[0][0]']\n", " orm (BatchNormalization) \n", " \n", " expanded_conv_2/Add (Add) (None, 56, 56, 24) 0 ['expanded_conv_1/project/BatchNo\n", " rm[0][0]', \n", " 'expanded_conv_2/project/BatchNo\n", " rm[0][0]'] \n", " \n", " expanded_conv_3/expand (Conv2D (None, 56, 56, 72) 1728 ['expanded_conv_2/Add[0][0]'] \n", " ) \n", " \n", " expanded_conv_3/expand/BatchNo (None, 56, 56, 72) 288 ['expanded_conv_3/expand[0][0]'] \n", " rm (BatchNormalization) \n", " \n", " re_lu_6 (ReLU) (None, 56, 56, 72) 0 ['expanded_conv_3/expand/BatchNor\n", " m[0][0]'] \n", " \n", " expanded_conv_3/depthwise/pad (None, 59, 59, 72) 0 ['re_lu_6[0][0]'] \n", " (ZeroPadding2D) \n", " \n", " expanded_conv_3/depthwise (Dep (None, 28, 28, 72) 1800 ['expanded_conv_3/depthwise/pad[0\n", " thwiseConv2D) ][0]'] \n", " \n", " expanded_conv_3/depthwise/Batc (None, 28, 28, 72) 288 ['expanded_conv_3/depthwise[0][0]\n", " hNorm (BatchNormalization) '] \n", " \n", " re_lu_7 (ReLU) (None, 28, 28, 72) 0 ['expanded_conv_3/depthwise/Batch\n", " Norm[0][0]'] \n", " \n", " expanded_conv_3/squeeze_excite (None, 1, 1, 72) 0 ['re_lu_7[0][0]'] \n", " /AvgPool (GlobalAveragePooling \n", " 2D) \n", " \n", " expanded_conv_3/squeeze_excite (None, 1, 1, 24) 1752 ['expanded_conv_3/squeeze_excite/\n", " /Conv (Conv2D) AvgPool[0][0]'] \n", " \n", " expanded_conv_3/squeeze_excite (None, 1, 1, 24) 0 ['expanded_conv_3/squeeze_excite/\n", " /Relu (ReLU) Conv[0][0]'] \n", " \n", " expanded_conv_3/squeeze_excite (None, 1, 1, 72) 1800 ['expanded_conv_3/squeeze_excite/\n", " /Conv_1 (Conv2D) Relu[0][0]'] \n", " \n", " tf.__operators__.add_1 (TFOpLa (None, 1, 1, 72) 0 ['expanded_conv_3/squeeze_excite/\n", " mbda) Conv_1[0][0]'] \n", " \n", " re_lu_8 (ReLU) (None, 1, 1, 72) 0 ['tf.__operators__.add_1[0][0]'] \n", " \n", " tf.math.multiply_1 (TFOpLambda (None, 1, 1, 72) 0 ['re_lu_8[0][0]'] \n", " ) \n", " \n", " expanded_conv_3/squeeze_excite (None, 28, 28, 72) 0 ['re_lu_7[0][0]', \n", " /Mul (Multiply) 'tf.math.multiply_1[0][0]'] \n", " \n", " expanded_conv_3/project (Conv2 (None, 28, 28, 40) 2880 ['expanded_conv_3/squeeze_excite/\n", " D) Mul[0][0]'] \n", " \n", " expanded_conv_3/project/BatchN (None, 28, 28, 40) 160 ['expanded_conv_3/project[0][0]']\n", " orm (BatchNormalization) \n", " \n", " expanded_conv_4/expand (Conv2D (None, 28, 28, 120) 4800 ['expanded_conv_3/project/BatchNo\n", " ) rm[0][0]'] \n", " \n", " expanded_conv_4/expand/BatchNo (None, 28, 28, 120) 480 ['expanded_conv_4/expand[0][0]'] \n", " rm (BatchNormalization) \n", " \n", " re_lu_9 (ReLU) (None, 28, 28, 120) 0 ['expanded_conv_4/expand/BatchNor\n", " m[0][0]'] \n", " \n", " expanded_conv_4/depthwise (Dep (None, 28, 28, 120) 3000 ['re_lu_9[0][0]'] \n", " thwiseConv2D) \n", " \n", " expanded_conv_4/depthwise/Batc (None, 28, 28, 120) 480 ['expanded_conv_4/depthwise[0][0]\n", " hNorm (BatchNormalization) '] \n", " \n", " re_lu_10 (ReLU) (None, 28, 28, 120) 0 ['expanded_conv_4/depthwise/Batch\n", " Norm[0][0]'] \n", " \n", " expanded_conv_4/squeeze_excite (None, 1, 1, 120) 0 ['re_lu_10[0][0]'] \n", " /AvgPool (GlobalAveragePooling \n", " 2D) \n", " \n", " expanded_conv_4/squeeze_excite (None, 1, 1, 32) 3872 ['expanded_conv_4/squeeze_excite/\n", " /Conv (Conv2D) AvgPool[0][0]'] \n", " \n", " expanded_conv_4/squeeze_excite (None, 1, 1, 32) 0 ['expanded_conv_4/squeeze_excite/\n", " /Relu (ReLU) Conv[0][0]'] \n", " \n", " expanded_conv_4/squeeze_excite (None, 1, 1, 120) 3960 ['expanded_conv_4/squeeze_excite/\n", " /Conv_1 (Conv2D) Relu[0][0]'] \n", " \n", " tf.__operators__.add_2 (TFOpLa (None, 1, 1, 120) 0 ['expanded_conv_4/squeeze_excite/\n", " mbda) Conv_1[0][0]'] \n", " \n", " re_lu_11 (ReLU) (None, 1, 1, 120) 0 ['tf.__operators__.add_2[0][0]'] \n", " \n", " tf.math.multiply_2 (TFOpLambda (None, 1, 1, 120) 0 ['re_lu_11[0][0]'] \n", " ) \n", " \n", " expanded_conv_4/squeeze_excite (None, 28, 28, 120) 0 ['re_lu_10[0][0]', \n", " /Mul (Multiply) 'tf.math.multiply_2[0][0]'] \n", " \n", " expanded_conv_4/project (Conv2 (None, 28, 28, 40) 4800 ['expanded_conv_4/squeeze_excite/\n", " D) Mul[0][0]'] \n", " \n", " expanded_conv_4/project/BatchN (None, 28, 28, 40) 160 ['expanded_conv_4/project[0][0]']\n", " orm (BatchNormalization) \n", " \n", " expanded_conv_4/Add (Add) (None, 28, 28, 40) 0 ['expanded_conv_3/project/BatchNo\n", " rm[0][0]', \n", " 'expanded_conv_4/project/BatchNo\n", " rm[0][0]'] \n", " \n", " expanded_conv_5/expand (Conv2D (None, 28, 28, 120) 4800 ['expanded_conv_4/Add[0][0]'] \n", " ) \n", " \n", " expanded_conv_5/expand/BatchNo (None, 28, 28, 120) 480 ['expanded_conv_5/expand[0][0]'] \n", " rm (BatchNormalization) \n", " \n", " re_lu_12 (ReLU) (None, 28, 28, 120) 0 ['expanded_conv_5/expand/BatchNor\n", " m[0][0]'] \n", " \n", " expanded_conv_5/depthwise (Dep (None, 28, 28, 120) 3000 ['re_lu_12[0][0]'] \n", " thwiseConv2D) \n", " \n", " expanded_conv_5/depthwise/Batc (None, 28, 28, 120) 480 ['expanded_conv_5/depthwise[0][0]\n", " hNorm (BatchNormalization) '] \n", " \n", " re_lu_13 (ReLU) (None, 28, 28, 120) 0 ['expanded_conv_5/depthwise/Batch\n", " Norm[0][0]'] \n", " \n", " expanded_conv_5/squeeze_excite (None, 1, 1, 120) 0 ['re_lu_13[0][0]'] \n", " /AvgPool (GlobalAveragePooling \n", " 2D) \n", " \n", " expanded_conv_5/squeeze_excite (None, 1, 1, 32) 3872 ['expanded_conv_5/squeeze_excite/\n", " /Conv (Conv2D) AvgPool[0][0]'] \n", " \n", " expanded_conv_5/squeeze_excite (None, 1, 1, 32) 0 ['expanded_conv_5/squeeze_excite/\n", " /Relu (ReLU) Conv[0][0]'] \n", " \n", " expanded_conv_5/squeeze_excite (None, 1, 1, 120) 3960 ['expanded_conv_5/squeeze_excite/\n", " /Conv_1 (Conv2D) Relu[0][0]'] \n", " \n", " tf.__operators__.add_3 (TFOpLa (None, 1, 1, 120) 0 ['expanded_conv_5/squeeze_excite/\n", " mbda) Conv_1[0][0]'] \n", " \n", " re_lu_14 (ReLU) (None, 1, 1, 120) 0 ['tf.__operators__.add_3[0][0]'] \n", " \n", " tf.math.multiply_3 (TFOpLambda (None, 1, 1, 120) 0 ['re_lu_14[0][0]'] \n", " ) \n", " \n", " expanded_conv_5/squeeze_excite (None, 28, 28, 120) 0 ['re_lu_13[0][0]', \n", " /Mul (Multiply) 'tf.math.multiply_3[0][0]'] \n", " \n", " expanded_conv_5/project (Conv2 (None, 28, 28, 40) 4800 ['expanded_conv_5/squeeze_excite/\n", " D) Mul[0][0]'] \n", " \n", " expanded_conv_5/project/BatchN (None, 28, 28, 40) 160 ['expanded_conv_5/project[0][0]']\n", " orm (BatchNormalization) \n", " \n", " expanded_conv_5/Add (Add) (None, 28, 28, 40) 0 ['expanded_conv_4/Add[0][0]', \n", " 'expanded_conv_5/project/BatchNo\n", " rm[0][0]'] \n", " \n", " expanded_conv_6/expand (Conv2D (None, 28, 28, 240) 9600 ['expanded_conv_5/Add[0][0]'] \n", " ) \n", " \n", " expanded_conv_6/expand/BatchNo (None, 28, 28, 240) 960 ['expanded_conv_6/expand[0][0]'] \n", " rm (BatchNormalization) \n", " \n", " tf.__operators__.add_4 (TFOpLa (None, 28, 28, 240) 0 ['expanded_conv_6/expand/BatchNor\n", " mbda) m[0][0]'] \n", " \n", " re_lu_15 (ReLU) (None, 28, 28, 240) 0 ['tf.__operators__.add_4[0][0]'] \n", " \n", " tf.math.multiply_4 (TFOpLambda (None, 28, 28, 240) 0 ['re_lu_15[0][0]'] \n", " ) \n", " \n", " multiply_1 (Multiply) (None, 28, 28, 240) 0 ['expanded_conv_6/expand/BatchNor\n", " m[0][0]', \n", " 'tf.math.multiply_4[0][0]'] \n", " \n", " expanded_conv_6/depthwise/pad (None, 29, 29, 240) 0 ['multiply_1[0][0]'] \n", " (ZeroPadding2D) \n", " \n", " expanded_conv_6/depthwise (Dep (None, 14, 14, 240) 2160 ['expanded_conv_6/depthwise/pad[0\n", " thwiseConv2D) ][0]'] \n", " \n", " expanded_conv_6/depthwise/Batc (None, 14, 14, 240) 960 ['expanded_conv_6/depthwise[0][0]\n", " hNorm (BatchNormalization) '] \n", " \n", " tf.__operators__.add_5 (TFOpLa (None, 14, 14, 240) 0 ['expanded_conv_6/depthwise/Batch\n", " mbda) Norm[0][0]'] \n", " \n", " re_lu_16 (ReLU) (None, 14, 14, 240) 0 ['tf.__operators__.add_5[0][0]'] \n", " \n", " tf.math.multiply_5 (TFOpLambda (None, 14, 14, 240) 0 ['re_lu_16[0][0]'] \n", " ) \n", " \n", " multiply_2 (Multiply) (None, 14, 14, 240) 0 ['expanded_conv_6/depthwise/Batch\n", " Norm[0][0]', \n", " 'tf.math.multiply_5[0][0]'] \n", " \n", " expanded_conv_6/project (Conv2 (None, 14, 14, 80) 19200 ['multiply_2[0][0]'] \n", " D) \n", " \n", " expanded_conv_6/project/BatchN (None, 14, 14, 80) 320 ['expanded_conv_6/project[0][0]']\n", " orm (BatchNormalization) \n", " \n", " expanded_conv_7/expand (Conv2D (None, 14, 14, 200) 16000 ['expanded_conv_6/project/BatchNo\n", " ) rm[0][0]'] \n", " \n", " expanded_conv_7/expand/BatchNo (None, 14, 14, 200) 800 ['expanded_conv_7/expand[0][0]'] \n", " rm (BatchNormalization) \n", " \n", " tf.__operators__.add_6 (TFOpLa (None, 14, 14, 200) 0 ['expanded_conv_7/expand/BatchNor\n", " mbda) m[0][0]'] \n", " \n", " re_lu_17 (ReLU) (None, 14, 14, 200) 0 ['tf.__operators__.add_6[0][0]'] \n", " \n", " tf.math.multiply_6 (TFOpLambda (None, 14, 14, 200) 0 ['re_lu_17[0][0]'] \n", " ) \n", " \n", " multiply_3 (Multiply) (None, 14, 14, 200) 0 ['expanded_conv_7/expand/BatchNor\n", " m[0][0]', \n", " 'tf.math.multiply_6[0][0]'] \n", " \n", " expanded_conv_7/depthwise (Dep (None, 14, 14, 200) 1800 ['multiply_3[0][0]'] \n", " thwiseConv2D) \n", " \n", " expanded_conv_7/depthwise/Batc (None, 14, 14, 200) 800 ['expanded_conv_7/depthwise[0][0]\n", " hNorm (BatchNormalization) '] \n", " \n", " tf.__operators__.add_7 (TFOpLa (None, 14, 14, 200) 0 ['expanded_conv_7/depthwise/Batch\n", " mbda) Norm[0][0]'] \n", " \n", " re_lu_18 (ReLU) (None, 14, 14, 200) 0 ['tf.__operators__.add_7[0][0]'] \n", " \n", " tf.math.multiply_7 (TFOpLambda (None, 14, 14, 200) 0 ['re_lu_18[0][0]'] \n", " ) \n", " \n", " multiply_4 (Multiply) (None, 14, 14, 200) 0 ['expanded_conv_7/depthwise/Batch\n", " Norm[0][0]', \n", " 'tf.math.multiply_7[0][0]'] \n", " \n", " expanded_conv_7/project (Conv2 (None, 14, 14, 80) 16000 ['multiply_4[0][0]'] \n", " D) \n", " \n", " expanded_conv_7/project/BatchN (None, 14, 14, 80) 320 ['expanded_conv_7/project[0][0]']\n", " orm (BatchNormalization) \n", " \n", " expanded_conv_7/Add (Add) (None, 14, 14, 80) 0 ['expanded_conv_6/project/BatchNo\n", " rm[0][0]', \n", " 'expanded_conv_7/project/BatchNo\n", " rm[0][0]'] \n", " \n", " expanded_conv_8/expand (Conv2D (None, 14, 14, 184) 14720 ['expanded_conv_7/Add[0][0]'] \n", " ) \n", " \n", " expanded_conv_8/expand/BatchNo (None, 14, 14, 184) 736 ['expanded_conv_8/expand[0][0]'] \n", " rm (BatchNormalization) \n", " \n", " tf.__operators__.add_8 (TFOpLa (None, 14, 14, 184) 0 ['expanded_conv_8/expand/BatchNor\n", " mbda) m[0][0]'] \n", " \n", " re_lu_19 (ReLU) (None, 14, 14, 184) 0 ['tf.__operators__.add_8[0][0]'] \n", " \n", " tf.math.multiply_8 (TFOpLambda (None, 14, 14, 184) 0 ['re_lu_19[0][0]'] \n", " ) \n", " \n", " multiply_5 (Multiply) (None, 14, 14, 184) 0 ['expanded_conv_8/expand/BatchNor\n", " m[0][0]', \n", " 'tf.math.multiply_8[0][0]'] \n", " \n", " expanded_conv_8/depthwise (Dep (None, 14, 14, 184) 1656 ['multiply_5[0][0]'] \n", " thwiseConv2D) \n", " \n", " expanded_conv_8/depthwise/Batc (None, 14, 14, 184) 736 ['expanded_conv_8/depthwise[0][0]\n", " hNorm (BatchNormalization) '] \n", " \n", " tf.__operators__.add_9 (TFOpLa (None, 14, 14, 184) 0 ['expanded_conv_8/depthwise/Batch\n", " mbda) Norm[0][0]'] \n", " \n", " re_lu_20 (ReLU) (None, 14, 14, 184) 0 ['tf.__operators__.add_9[0][0]'] \n", " \n", " tf.math.multiply_9 (TFOpLambda (None, 14, 14, 184) 0 ['re_lu_20[0][0]'] \n", " ) \n", " \n", " multiply_6 (Multiply) (None, 14, 14, 184) 0 ['expanded_conv_8/depthwise/Batch\n", " Norm[0][0]', \n", " 'tf.math.multiply_9[0][0]'] \n", " \n", " expanded_conv_8/project (Conv2 (None, 14, 14, 80) 14720 ['multiply_6[0][0]'] \n", " D) \n", " \n", " expanded_conv_8/project/BatchN (None, 14, 14, 80) 320 ['expanded_conv_8/project[0][0]']\n", " orm (BatchNormalization) \n", " \n", " expanded_conv_8/Add (Add) (None, 14, 14, 80) 0 ['expanded_conv_7/Add[0][0]', \n", " 'expanded_conv_8/project/BatchNo\n", " rm[0][0]'] \n", " \n", " expanded_conv_9/expand (Conv2D (None, 14, 14, 184) 14720 ['expanded_conv_8/Add[0][0]'] \n", " ) \n", " \n", " expanded_conv_9/expand/BatchNo (None, 14, 14, 184) 736 ['expanded_conv_9/expand[0][0]'] \n", " rm (BatchNormalization) \n", " \n", " tf.__operators__.add_10 (TFOpL (None, 14, 14, 184) 0 ['expanded_conv_9/expand/BatchNor\n", " ambda) m[0][0]'] \n", " \n", " re_lu_21 (ReLU) (None, 14, 14, 184) 0 ['tf.__operators__.add_10[0][0]']\n", " \n", " tf.math.multiply_10 (TFOpLambd (None, 14, 14, 184) 0 ['re_lu_21[0][0]'] \n", " a) \n", " \n", " multiply_7 (Multiply) (None, 14, 14, 184) 0 ['expanded_conv_9/expand/BatchNor\n", " m[0][0]', \n", " 'tf.math.multiply_10[0][0]'] \n", " \n", " expanded_conv_9/depthwise (Dep (None, 14, 14, 184) 1656 ['multiply_7[0][0]'] \n", " thwiseConv2D) \n", " \n", " expanded_conv_9/depthwise/Batc (None, 14, 14, 184) 736 ['expanded_conv_9/depthwise[0][0]\n", " hNorm (BatchNormalization) '] \n", " \n", " tf.__operators__.add_11 (TFOpL (None, 14, 14, 184) 0 ['expanded_conv_9/depthwise/Batch\n", " ambda) Norm[0][0]'] \n", " \n", " re_lu_22 (ReLU) (None, 14, 14, 184) 0 ['tf.__operators__.add_11[0][0]']\n", " \n", " tf.math.multiply_11 (TFOpLambd (None, 14, 14, 184) 0 ['re_lu_22[0][0]'] \n", " a) \n", " \n", " multiply_8 (Multiply) (None, 14, 14, 184) 0 ['expanded_conv_9/depthwise/Batch\n", " Norm[0][0]', \n", " 'tf.math.multiply_11[0][0]'] \n", " \n", " expanded_conv_9/project (Conv2 (None, 14, 14, 80) 14720 ['multiply_8[0][0]'] \n", " D) \n", " \n", " expanded_conv_9/project/BatchN (None, 14, 14, 80) 320 ['expanded_conv_9/project[0][0]']\n", " orm (BatchNormalization) \n", " \n", " expanded_conv_9/Add (Add) (None, 14, 14, 80) 0 ['expanded_conv_8/Add[0][0]', \n", " 'expanded_conv_9/project/BatchNo\n", " rm[0][0]'] \n", " \n", " expanded_conv_10/expand (Conv2 (None, 14, 14, 480) 38400 ['expanded_conv_9/Add[0][0]'] \n", " D) \n", " \n", " expanded_conv_10/expand/BatchN (None, 14, 14, 480) 1920 ['expanded_conv_10/expand[0][0]']\n", " orm (BatchNormalization) \n", " \n", " tf.__operators__.add_12 (TFOpL (None, 14, 14, 480) 0 ['expanded_conv_10/expand/BatchNo\n", " ambda) rm[0][0]'] \n", " \n", " re_lu_23 (ReLU) (None, 14, 14, 480) 0 ['tf.__operators__.add_12[0][0]']\n", " \n", " tf.math.multiply_12 (TFOpLambd (None, 14, 14, 480) 0 ['re_lu_23[0][0]'] \n", " a) \n", " \n", " multiply_9 (Multiply) (None, 14, 14, 480) 0 ['expanded_conv_10/expand/BatchNo\n", " rm[0][0]', \n", " 'tf.math.multiply_12[0][0]'] \n", " \n", " expanded_conv_10/depthwise (De (None, 14, 14, 480) 4320 ['multiply_9[0][0]'] \n", " pthwiseConv2D) \n", " \n", " expanded_conv_10/depthwise/Bat (None, 14, 14, 480) 1920 ['expanded_conv_10/depthwise[0][0\n", " chNorm (BatchNormalization) ]'] \n", " \n", " tf.__operators__.add_13 (TFOpL (None, 14, 14, 480) 0 ['expanded_conv_10/depthwise/Batc\n", " ambda) hNorm[0][0]'] \n", " \n", " re_lu_24 (ReLU) (None, 14, 14, 480) 0 ['tf.__operators__.add_13[0][0]']\n", " \n", " tf.math.multiply_13 (TFOpLambd (None, 14, 14, 480) 0 ['re_lu_24[0][0]'] \n", " a) \n", " \n", " multiply_10 (Multiply) (None, 14, 14, 480) 0 ['expanded_conv_10/depthwise/Batc\n", " hNorm[0][0]', \n", " 'tf.math.multiply_13[0][0]'] \n", " \n", " expanded_conv_10/squeeze_excit (None, 1, 1, 480) 0 ['multiply_10[0][0]'] \n", " e/AvgPool (GlobalAveragePoolin \n", " g2D) \n", " \n", " expanded_conv_10/squeeze_excit (None, 1, 1, 120) 57720 ['expanded_conv_10/squeeze_excite\n", " e/Conv (Conv2D) /AvgPool[0][0]'] \n", " \n", " expanded_conv_10/squeeze_excit (None, 1, 1, 120) 0 ['expanded_conv_10/squeeze_excite\n", " e/Relu (ReLU) /Conv[0][0]'] \n", " \n", " expanded_conv_10/squeeze_excit (None, 1, 1, 480) 58080 ['expanded_conv_10/squeeze_excite\n", " e/Conv_1 (Conv2D) /Relu[0][0]'] \n", " \n", " tf.__operators__.add_14 (TFOpL (None, 1, 1, 480) 0 ['expanded_conv_10/squeeze_excite\n", " ambda) /Conv_1[0][0]'] \n", " \n", " re_lu_25 (ReLU) (None, 1, 1, 480) 0 ['tf.__operators__.add_14[0][0]']\n", " \n", " tf.math.multiply_14 (TFOpLambd (None, 1, 1, 480) 0 ['re_lu_25[0][0]'] \n", " a) \n", " \n", " expanded_conv_10/squeeze_excit (None, 14, 14, 480) 0 ['multiply_10[0][0]', \n", " e/Mul (Multiply) 'tf.math.multiply_14[0][0]'] \n", " \n", " expanded_conv_10/project (Conv (None, 14, 14, 112) 53760 ['expanded_conv_10/squeeze_excite\n", " 2D) /Mul[0][0]'] \n", " \n", " expanded_conv_10/project/Batch (None, 14, 14, 112) 448 ['expanded_conv_10/project[0][0]'\n", " Norm (BatchNormalization) ] \n", " \n", " expanded_conv_11/expand (Conv2 (None, 14, 14, 672) 75264 ['expanded_conv_10/project/BatchN\n", " D) orm[0][0]'] \n", " \n", " expanded_conv_11/expand/BatchN (None, 14, 14, 672) 2688 ['expanded_conv_11/expand[0][0]']\n", " orm (BatchNormalization) \n", " \n", " tf.__operators__.add_15 (TFOpL (None, 14, 14, 672) 0 ['expanded_conv_11/expand/BatchNo\n", " ambda) rm[0][0]'] \n", " \n", " re_lu_26 (ReLU) (None, 14, 14, 672) 0 ['tf.__operators__.add_15[0][0]']\n", " \n", " tf.math.multiply_15 (TFOpLambd (None, 14, 14, 672) 0 ['re_lu_26[0][0]'] \n", " a) \n", " \n", " multiply_11 (Multiply) (None, 14, 14, 672) 0 ['expanded_conv_11/expand/BatchNo\n", " rm[0][0]', \n", " 'tf.math.multiply_15[0][0]'] \n", " \n", " expanded_conv_11/depthwise (De (None, 14, 14, 672) 6048 ['multiply_11[0][0]'] \n", " pthwiseConv2D) \n", " \n", " expanded_conv_11/depthwise/Bat (None, 14, 14, 672) 2688 ['expanded_conv_11/depthwise[0][0\n", " chNorm (BatchNormalization) ]'] \n", " \n", " tf.__operators__.add_16 (TFOpL (None, 14, 14, 672) 0 ['expanded_conv_11/depthwise/Batc\n", " ambda) hNorm[0][0]'] \n", " \n", " re_lu_27 (ReLU) (None, 14, 14, 672) 0 ['tf.__operators__.add_16[0][0]']\n", " \n", " tf.math.multiply_16 (TFOpLambd (None, 14, 14, 672) 0 ['re_lu_27[0][0]'] \n", " a) \n", " \n", " multiply_12 (Multiply) (None, 14, 14, 672) 0 ['expanded_conv_11/depthwise/Batc\n", " hNorm[0][0]', \n", " 'tf.math.multiply_16[0][0]'] \n", " \n", " expanded_conv_11/squeeze_excit (None, 1, 1, 672) 0 ['multiply_12[0][0]'] \n", " e/AvgPool (GlobalAveragePoolin \n", " g2D) \n", " \n", " expanded_conv_11/squeeze_excit (None, 1, 1, 168) 113064 ['expanded_conv_11/squeeze_excite\n", " e/Conv (Conv2D) /AvgPool[0][0]'] \n", " \n", " expanded_conv_11/squeeze_excit (None, 1, 1, 168) 0 ['expanded_conv_11/squeeze_excite\n", " e/Relu (ReLU) /Conv[0][0]'] \n", " \n", " expanded_conv_11/squeeze_excit (None, 1, 1, 672) 113568 ['expanded_conv_11/squeeze_excite\n", " e/Conv_1 (Conv2D) /Relu[0][0]'] \n", " \n", " tf.__operators__.add_17 (TFOpL (None, 1, 1, 672) 0 ['expanded_conv_11/squeeze_excite\n", " ambda) /Conv_1[0][0]'] \n", " \n", " re_lu_28 (ReLU) (None, 1, 1, 672) 0 ['tf.__operators__.add_17[0][0]']\n", " \n", " tf.math.multiply_17 (TFOpLambd (None, 1, 1, 672) 0 ['re_lu_28[0][0]'] \n", " a) \n", " \n", " expanded_conv_11/squeeze_excit (None, 14, 14, 672) 0 ['multiply_12[0][0]', \n", " e/Mul (Multiply) 'tf.math.multiply_17[0][0]'] \n", " \n", " expanded_conv_11/project (Conv (None, 14, 14, 112) 75264 ['expanded_conv_11/squeeze_excite\n", " 2D) /Mul[0][0]'] \n", " \n", " expanded_conv_11/project/Batch (None, 14, 14, 112) 448 ['expanded_conv_11/project[0][0]'\n", " Norm (BatchNormalization) ] \n", " \n", " expanded_conv_11/Add (Add) (None, 14, 14, 112) 0 ['expanded_conv_10/project/BatchN\n", " orm[0][0]', \n", " 'expanded_conv_11/project/BatchN\n", " orm[0][0]'] \n", " \n", " expanded_conv_12/expand (Conv2 (None, 14, 14, 672) 75264 ['expanded_conv_11/Add[0][0]'] \n", " D) \n", " \n", " expanded_conv_12/expand/BatchN (None, 14, 14, 672) 2688 ['expanded_conv_12/expand[0][0]']\n", " orm (BatchNormalization) \n", " \n", " tf.__operators__.add_18 (TFOpL (None, 14, 14, 672) 0 ['expanded_conv_12/expand/BatchNo\n", " ambda) rm[0][0]'] \n", " \n", " re_lu_29 (ReLU) (None, 14, 14, 672) 0 ['tf.__operators__.add_18[0][0]']\n", " \n", " tf.math.multiply_18 (TFOpLambd (None, 14, 14, 672) 0 ['re_lu_29[0][0]'] \n", " a) \n", " \n", " multiply_13 (Multiply) (None, 14, 14, 672) 0 ['expanded_conv_12/expand/BatchNo\n", " rm[0][0]', \n", " 'tf.math.multiply_18[0][0]'] \n", " \n", " expanded_conv_12/depthwise/pad (None, 17, 17, 672) 0 ['multiply_13[0][0]'] \n", " (ZeroPadding2D) \n", " \n", " expanded_conv_12/depthwise (De (None, 7, 7, 672) 16800 ['expanded_conv_12/depthwise/pad[\n", " pthwiseConv2D) 0][0]'] \n", " \n", " expanded_conv_12/depthwise/Bat (None, 7, 7, 672) 2688 ['expanded_conv_12/depthwise[0][0\n", " chNorm (BatchNormalization) ]'] \n", " \n", " tf.__operators__.add_19 (TFOpL (None, 7, 7, 672) 0 ['expanded_conv_12/depthwise/Batc\n", " ambda) hNorm[0][0]'] \n", " \n", " re_lu_30 (ReLU) (None, 7, 7, 672) 0 ['tf.__operators__.add_19[0][0]']\n", " \n", " tf.math.multiply_19 (TFOpLambd (None, 7, 7, 672) 0 ['re_lu_30[0][0]'] \n", " a) \n", " \n", " multiply_14 (Multiply) (None, 7, 7, 672) 0 ['expanded_conv_12/depthwise/Batc\n", " hNorm[0][0]', \n", " 'tf.math.multiply_19[0][0]'] \n", " \n", " expanded_conv_12/squeeze_excit (None, 1, 1, 672) 0 ['multiply_14[0][0]'] \n", " e/AvgPool (GlobalAveragePoolin \n", " g2D) \n", " \n", " expanded_conv_12/squeeze_excit (None, 1, 1, 168) 113064 ['expanded_conv_12/squeeze_excite\n", " e/Conv (Conv2D) /AvgPool[0][0]'] \n", " \n", " expanded_conv_12/squeeze_excit (None, 1, 1, 168) 0 ['expanded_conv_12/squeeze_excite\n", " e/Relu (ReLU) /Conv[0][0]'] \n", " \n", " expanded_conv_12/squeeze_excit (None, 1, 1, 672) 113568 ['expanded_conv_12/squeeze_excite\n", " e/Conv_1 (Conv2D) /Relu[0][0]'] \n", " \n", " tf.__operators__.add_20 (TFOpL (None, 1, 1, 672) 0 ['expanded_conv_12/squeeze_excite\n", " ambda) /Conv_1[0][0]'] \n", " \n", " re_lu_31 (ReLU) (None, 1, 1, 672) 0 ['tf.__operators__.add_20[0][0]']\n", " \n", " tf.math.multiply_20 (TFOpLambd (None, 1, 1, 672) 0 ['re_lu_31[0][0]'] \n", " a) \n", " \n", " expanded_conv_12/squeeze_excit (None, 7, 7, 672) 0 ['multiply_14[0][0]', \n", " e/Mul (Multiply) 'tf.math.multiply_20[0][0]'] \n", " \n", " expanded_conv_12/project (Conv (None, 7, 7, 160) 107520 ['expanded_conv_12/squeeze_excite\n", " 2D) /Mul[0][0]'] \n", " \n", " expanded_conv_12/project/Batch (None, 7, 7, 160) 640 ['expanded_conv_12/project[0][0]'\n", " Norm (BatchNormalization) ] \n", " \n", " expanded_conv_13/expand (Conv2 (None, 7, 7, 960) 153600 ['expanded_conv_12/project/BatchN\n", " D) orm[0][0]'] \n", " \n", " expanded_conv_13/expand/BatchN (None, 7, 7, 960) 3840 ['expanded_conv_13/expand[0][0]']\n", " orm (BatchNormalization) \n", " \n", " tf.__operators__.add_21 (TFOpL (None, 7, 7, 960) 0 ['expanded_conv_13/expand/BatchNo\n", " ambda) rm[0][0]'] \n", " \n", " re_lu_32 (ReLU) (None, 7, 7, 960) 0 ['tf.__operators__.add_21[0][0]']\n", " \n", " tf.math.multiply_21 (TFOpLambd (None, 7, 7, 960) 0 ['re_lu_32[0][0]'] \n", " a) \n", " \n", " multiply_15 (Multiply) (None, 7, 7, 960) 0 ['expanded_conv_13/expand/BatchNo\n", " rm[0][0]', \n", " 'tf.math.multiply_21[0][0]'] \n", " \n", " expanded_conv_13/depthwise (De (None, 7, 7, 960) 24000 ['multiply_15[0][0]'] \n", " pthwiseConv2D) \n", " \n", " expanded_conv_13/depthwise/Bat (None, 7, 7, 960) 3840 ['expanded_conv_13/depthwise[0][0\n", " chNorm (BatchNormalization) ]'] \n", " \n", " tf.__operators__.add_22 (TFOpL (None, 7, 7, 960) 0 ['expanded_conv_13/depthwise/Batc\n", " ambda) hNorm[0][0]'] \n", " \n", " re_lu_33 (ReLU) (None, 7, 7, 960) 0 ['tf.__operators__.add_22[0][0]']\n", " \n", " tf.math.multiply_22 (TFOpLambd (None, 7, 7, 960) 0 ['re_lu_33[0][0]'] \n", " a) \n", " \n", " multiply_16 (Multiply) (None, 7, 7, 960) 0 ['expanded_conv_13/depthwise/Batc\n", " hNorm[0][0]', \n", " 'tf.math.multiply_22[0][0]'] \n", " \n", " expanded_conv_13/squeeze_excit (None, 1, 1, 960) 0 ['multiply_16[0][0]'] \n", " e/AvgPool (GlobalAveragePoolin \n", " g2D) \n", " \n", " expanded_conv_13/squeeze_excit (None, 1, 1, 240) 230640 ['expanded_conv_13/squeeze_excite\n", " e/Conv (Conv2D) /AvgPool[0][0]'] \n", " \n", " expanded_conv_13/squeeze_excit (None, 1, 1, 240) 0 ['expanded_conv_13/squeeze_excite\n", " e/Relu (ReLU) /Conv[0][0]'] \n", " \n", " expanded_conv_13/squeeze_excit (None, 1, 1, 960) 231360 ['expanded_conv_13/squeeze_excite\n", " e/Conv_1 (Conv2D) /Relu[0][0]'] \n", " \n", " tf.__operators__.add_23 (TFOpL (None, 1, 1, 960) 0 ['expanded_conv_13/squeeze_excite\n", " ambda) /Conv_1[0][0]'] \n", " \n", " re_lu_34 (ReLU) (None, 1, 1, 960) 0 ['tf.__operators__.add_23[0][0]']\n", " \n", " tf.math.multiply_23 (TFOpLambd (None, 1, 1, 960) 0 ['re_lu_34[0][0]'] \n", " a) \n", " \n", " expanded_conv_13/squeeze_excit (None, 7, 7, 960) 0 ['multiply_16[0][0]', \n", " e/Mul (Multiply) 'tf.math.multiply_23[0][0]'] \n", " \n", " expanded_conv_13/project (Conv (None, 7, 7, 160) 153600 ['expanded_conv_13/squeeze_excite\n", " 2D) /Mul[0][0]'] \n", " \n", " expanded_conv_13/project/Batch (None, 7, 7, 160) 640 ['expanded_conv_13/project[0][0]'\n", " Norm (BatchNormalization) ] \n", " \n", " expanded_conv_13/Add (Add) (None, 7, 7, 160) 0 ['expanded_conv_12/project/BatchN\n", " orm[0][0]', \n", " 'expanded_conv_13/project/BatchN\n", " orm[0][0]'] \n", " \n", " expanded_conv_14/expand (Conv2 (None, 7, 7, 960) 153600 ['expanded_conv_13/Add[0][0]'] \n", " D) \n", " \n", " expanded_conv_14/expand/BatchN (None, 7, 7, 960) 3840 ['expanded_conv_14/expand[0][0]']\n", " orm (BatchNormalization) \n", " \n", " tf.__operators__.add_24 (TFOpL (None, 7, 7, 960) 0 ['expanded_conv_14/expand/BatchNo\n", " ambda) rm[0][0]'] \n", " \n", " re_lu_35 (ReLU) (None, 7, 7, 960) 0 ['tf.__operators__.add_24[0][0]']\n", " \n", " tf.math.multiply_24 (TFOpLambd (None, 7, 7, 960) 0 ['re_lu_35[0][0]'] \n", " a) \n", " \n", " multiply_17 (Multiply) (None, 7, 7, 960) 0 ['expanded_conv_14/expand/BatchNo\n", " rm[0][0]', \n", " 'tf.math.multiply_24[0][0]'] \n", " \n", " expanded_conv_14/depthwise (De (None, 7, 7, 960) 24000 ['multiply_17[0][0]'] \n", " pthwiseConv2D) \n", " \n", " expanded_conv_14/depthwise/Bat (None, 7, 7, 960) 3840 ['expanded_conv_14/depthwise[0][0\n", " chNorm (BatchNormalization) ]'] \n", " \n", " tf.__operators__.add_25 (TFOpL (None, 7, 7, 960) 0 ['expanded_conv_14/depthwise/Batc\n", " ambda) hNorm[0][0]'] \n", " \n", " re_lu_36 (ReLU) (None, 7, 7, 960) 0 ['tf.__operators__.add_25[0][0]']\n", " \n", " tf.math.multiply_25 (TFOpLambd (None, 7, 7, 960) 0 ['re_lu_36[0][0]'] \n", " a) \n", " \n", " multiply_18 (Multiply) (None, 7, 7, 960) 0 ['expanded_conv_14/depthwise/Batc\n", " hNorm[0][0]', \n", " 'tf.math.multiply_25[0][0]'] \n", " \n", " expanded_conv_14/squeeze_excit (None, 1, 1, 960) 0 ['multiply_18[0][0]'] \n", " e/AvgPool (GlobalAveragePoolin \n", " g2D) \n", " \n", " expanded_conv_14/squeeze_excit (None, 1, 1, 240) 230640 ['expanded_conv_14/squeeze_excite\n", " e/Conv (Conv2D) /AvgPool[0][0]'] \n", " \n", " expanded_conv_14/squeeze_excit (None, 1, 1, 240) 0 ['expanded_conv_14/squeeze_excite\n", " e/Relu (ReLU) /Conv[0][0]'] \n", " \n", " expanded_conv_14/squeeze_excit (None, 1, 1, 960) 231360 ['expanded_conv_14/squeeze_excite\n", " e/Conv_1 (Conv2D) /Relu[0][0]'] \n", " \n", " tf.__operators__.add_26 (TFOpL (None, 1, 1, 960) 0 ['expanded_conv_14/squeeze_excite\n", " ambda) /Conv_1[0][0]'] \n", " \n", " re_lu_37 (ReLU) (None, 1, 1, 960) 0 ['tf.__operators__.add_26[0][0]']\n", " \n", " tf.math.multiply_26 (TFOpLambd (None, 1, 1, 960) 0 ['re_lu_37[0][0]'] \n", " a) \n", " \n", " expanded_conv_14/squeeze_excit (None, 7, 7, 960) 0 ['multiply_18[0][0]', \n", " e/Mul (Multiply) 'tf.math.multiply_26[0][0]'] \n", " \n", " expanded_conv_14/project (Conv (None, 7, 7, 160) 153600 ['expanded_conv_14/squeeze_excite\n", " 2D) /Mul[0][0]'] \n", " \n", " expanded_conv_14/project/Batch (None, 7, 7, 160) 640 ['expanded_conv_14/project[0][0]'\n", " Norm (BatchNormalization) ] \n", " \n", " expanded_conv_14/Add (Add) (None, 7, 7, 160) 0 ['expanded_conv_13/Add[0][0]', \n", " 'expanded_conv_14/project/BatchN\n", " orm[0][0]'] \n", " \n", " Conv_1 (Conv2D) (None, 7, 7, 960) 153600 ['expanded_conv_14/Add[0][0]'] \n", " \n", " Conv_1/BatchNorm (BatchNormali (None, 7, 7, 960) 3840 ['Conv_1[0][0]'] \n", " zation) \n", " \n", " tf.__operators__.add_27 (TFOpL (None, 7, 7, 960) 0 ['Conv_1/BatchNorm[0][0]'] \n", " ambda) \n", " \n", " re_lu_38 (ReLU) (None, 7, 7, 960) 0 ['tf.__operators__.add_27[0][0]']\n", " \n", " tf.math.multiply_27 (TFOpLambd (None, 7, 7, 960) 0 ['re_lu_38[0][0]'] \n", " a) \n", " \n", " multiply_19 (Multiply) (None, 7, 7, 960) 0 ['Conv_1/BatchNorm[0][0]', \n", " 'tf.math.multiply_27[0][0]'] \n", " \n", " global_average_pooling2d (Glob (None, 1, 1, 960) 0 ['multiply_19[0][0]'] \n", " alAveragePooling2D) \n", " \n", " Conv_2 (Conv2D) (None, 1, 1, 1280) 1230080 ['global_average_pooling2d[0][0]'\n", " ] \n", " \n", " tf.__operators__.add_28 (TFOpL (None, 1, 1, 1280) 0 ['Conv_2[0][0]'] \n", " ambda) \n", " \n", " re_lu_39 (ReLU) (None, 1, 1, 1280) 0 ['tf.__operators__.add_28[0][0]']\n", " \n", " tf.math.multiply_28 (TFOpLambd (None, 1, 1, 1280) 0 ['re_lu_39[0][0]'] \n", " a) \n", " \n", " multiply_20 (Multiply) (None, 1, 1, 1280) 0 ['Conv_2[0][0]', \n", " 'tf.math.multiply_28[0][0]'] \n", " \n", " dropout (Dropout) (None, 1, 1, 1280) 0 ['multiply_20[0][0]'] \n", " \n", " Logits (Conv2D) (None, 1, 1, 1000) 1281000 ['dropout[0][0]'] \n", " \n", " flatten (Flatten) (None, 1000) 0 ['Logits[0][0]'] \n", " \n", " Predictions (Activation) (None, 1000) 0 ['flatten[0][0]'] \n", " \n", "==================================================================================================\n", "Total params: 5,507,432\n", "Trainable params: 5,483,032\n", "Non-trainable params: 24,400\n", "__________________________________________________________________________________________________\n" ] } ], "source": [ "model.summary()" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## Test the Model with a single image" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "1/1 [==============================] - 2s 2s/step\n", "289\n", "[[('n02128757', 'snow_leopard', 0.84875476), ('n02128385', 'leopard', 0.0349158), ('n02130308', 'cheetah', 0.021457493), ('n02124075', 'Egyptian_cat', 0.006221723), ('n02127052', 'lynx', 0.0051457486)]]\n" ] }, { "data": { "text/plain": [ "" ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "singel_tmp_image = tf.keras.utils.load_img('leopard.jpg',\n", " target_size=IMG_SHAPE,\n", " interpolation=\"nearest\")\n", "\n", "\n", "input_image = np.reshape(np.asarray(singel_tmp_image), ((1,) + IMG_SHAPE))\n", "predictions = model.predict(input_image)\n", "print(np.argmax(predictions))\n", "print(tf.keras.applications.mobilenet_v3.decode_predictions(predictions))\n", "\n", "plt.imshow(singel_tmp_image)\n" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## Save Model" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 155). These functions will not be directly callable after loading.\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "INFO:tensorflow:Assets written to: models/mobilenetv3-imagenet/assets\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "INFO:tensorflow:Assets written to: models/mobilenetv3-imagenet/assets\n" ] } ], "source": [ "#Saving a Keras model:\n", "model.save('models/mobilenetv3-imagenet')" ] } ], "metadata": { "accelerator": "GPU", "colab": { "collapsed_sections": [], "name": "transfer_learning.ipynb", "provenance": [], "toc_visible": true }, "kernelspec": { "display_name": "Python 3", "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": "3.10.9" } }, "nbformat": 4, "nbformat_minor": 0 }