{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"name": "GAP_cnn_tf2.ipynb",
"provenance": [],
"collapsed_sections": [],
"toc_visible": true,
"include_colab_link": true
},
"kernelspec": {
"name": "python3",
"display_name": "Python 3"
},
"accelerator": "GPU"
},
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "view-in-github",
"colab_type": "text"
},
"source": [
""
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "CkGh79ZDOG31",
"colab_type": "text"
},
"source": [
"## ‘কনভলিউশনাল নিউরাল নেটওয়ার্ক’ এবং গ্লোবাল অ্যাভারেজ পুলিং"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "aBd60ARu5iiB",
"colab_type": "text"
},
"source": [
"আমরা যখন ‘কনভলিউশনাল নিউরাল নেটওয়ার্ক’ নিয়ে কাজ করলাম, তখন একটা জিনিস বোঝা গেল আমাদের শেষ লেয়ারগুলোতে আমরা ‘ফুললি কানেক্টেড লেয়ার’গুলোর কিছু ভ্যারাইটি পেয়েছি। মানে শেষ লেয়ারে আমরা একটা স্ট্যান্ডার্ড নিউরাল নেটওয়ার্ককে যোগ করে দিতে পেরেছি। ‘সিএনএন’ অথবা কনভলিউশনাল নিউরাল নেটওয়ার্কের যে কয়েকটা এলিমেন্ট আছে সেগুলো নিয়ে আমরা দেখিয়েছি আগের চ্যাপ্টারগুলোতে। আপনি যদি ভালোভাবে দেখেন তাহলে মনে হবে এই কনভলিউশন মানে একটা ‘ইমেজ প্রসেসর’ - এর পরে একটা ‘স্ট্যান্ডার্ড নিউরাল নেটওয়ার্ক ক্লাসিফায়ার’কে যোগ করে দিয়েছি। আমাদের এই কনভলিউশনাল নিউরাল নেটওয়ার্ক বেশ কয়েকটা কনভলিউশনাল লেয়ার দিয়ে শুরু হয় যার মধ্যে ‘কার্নাল কনভলিউশন’ এবং ‘ম্যাক্স কুলিং’ অনেকগুলো ফিচার ম্যাপ তৈরি করে দেয় যেটা ছবির বিভিন্ন কম্পোনেন্ট এর রিপ্রেজেন্টেশন নিয়ে আসে। \n",
"\n",
"সবশেষে এই ‘ফুললি কানেক্টেড লেয়ার’ ওই ফিচার ম্যাপগুলোকে এমনভাবে ইন্টারপ্রেট করে যাতে তারা দরকার মতো ‘ক্যাটেগরি’ প্রেডিকশন করতে পারে। ভালো কথা, জিনিসটা এভাবেই চলছিল অনেকদিন - তবে এই ডিপ লার্নিং রিসার্চের যুগে এই জিনিসটাকে পাশ কাটিয়ে চলে আসে আরেকটা কনসেপ্ট যেটাকে আমরা বলছি ‘গ্লোবাল অ্যাভারেজ পুলিং’। এই ‘গ্লোবাল এভারেজ পুলিং’ করতে গেলে আমাদেরকে একটা ‘সিএনএন’ নেটওয়ার্ক তৈরি করে দেখাতে হবে। এরপর, আমরা সেখানে কিছুটা এডজাস্টমেন্ট করব তখনই বুঝে যাবেন কেন দরকার পড়ছে এই ‘গ্লোবাল অ্যাভারেজ পুলিং’। দুটো ‘প্রি-ট্রেইনড’ উদাহরণ দেখি। প্রথমটা VGG-16 মডেল। এটার একটা ছবি দেখি। \n",
" চিত্রঃ VGG-16 মডেল\n"
]
},
{
"cell_type": "code",
"metadata": {
"id": "yKtffSO-Sy8c",
"colab_type": "code",
"outputId": "ad8ba61f-c8be-4bd7-8a5e-05f1963b4865",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 34
}
},
"source": [
"try:\n",
" # শুধুমাত্র টেন্সর-ফ্লো ২.x ব্যবহার করবো \n",
" %tensorflow_version 2.x\n",
"except Exception:\n",
" pass\n",
"\n",
"import tensorflow as tf"
],
"execution_count": 0,
"outputs": [
{
"output_type": "stream",
"text": [
"TensorFlow 2.x selected.\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "gnioTikw5LSx",
"colab_type": "text"
},
"source": [
"## প্রি-ট্রেইনড মডেল VGG16 এর লেয়ার"
]
},
{
"cell_type": "code",
"metadata": {
"id": "6KePwncpRcwW",
"colab_type": "code",
"outputId": "c0b2c6a4-d042-47f4-cb78-60c95a3dd29e",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 976
}
},
"source": [
"# from tensorflow.keras import layers\n",
"\n",
"from tensorflow.keras.applications.vgg16 import VGG16; VGG16().summary()"
],
"execution_count": 0,
"outputs": [
{
"output_type": "stream",
"text": [
"Model: \"vgg16\"\n",
"_________________________________________________________________\n",
"Layer (type) Output Shape Param # \n",
"=================================================================\n",
"input_3 (InputLayer) [(None, 224, 224, 3)] 0 \n",
"_________________________________________________________________\n",
"block1_conv1 (Conv2D) (None, 224, 224, 64) 1792 \n",
"_________________________________________________________________\n",
"block1_conv2 (Conv2D) (None, 224, 224, 64) 36928 \n",
"_________________________________________________________________\n",
"block1_pool (MaxPooling2D) (None, 112, 112, 64) 0 \n",
"_________________________________________________________________\n",
"block2_conv1 (Conv2D) (None, 112, 112, 128) 73856 \n",
"_________________________________________________________________\n",
"block2_conv2 (Conv2D) (None, 112, 112, 128) 147584 \n",
"_________________________________________________________________\n",
"block2_pool (MaxPooling2D) (None, 56, 56, 128) 0 \n",
"_________________________________________________________________\n",
"block3_conv1 (Conv2D) (None, 56, 56, 256) 295168 \n",
"_________________________________________________________________\n",
"block3_conv2 (Conv2D) (None, 56, 56, 256) 590080 \n",
"_________________________________________________________________\n",
"block3_conv3 (Conv2D) (None, 56, 56, 256) 590080 \n",
"_________________________________________________________________\n",
"block3_pool (MaxPooling2D) (None, 28, 28, 256) 0 \n",
"_________________________________________________________________\n",
"block4_conv1 (Conv2D) (None, 28, 28, 512) 1180160 \n",
"_________________________________________________________________\n",
"block4_conv2 (Conv2D) (None, 28, 28, 512) 2359808 \n",
"_________________________________________________________________\n",
"block4_conv3 (Conv2D) (None, 28, 28, 512) 2359808 \n",
"_________________________________________________________________\n",
"block4_pool (MaxPooling2D) (None, 14, 14, 512) 0 \n",
"_________________________________________________________________\n",
"block5_conv1 (Conv2D) (None, 14, 14, 512) 2359808 \n",
"_________________________________________________________________\n",
"block5_conv2 (Conv2D) (None, 14, 14, 512) 2359808 \n",
"_________________________________________________________________\n",
"block5_conv3 (Conv2D) (None, 14, 14, 512) 2359808 \n",
"_________________________________________________________________\n",
"block5_pool (MaxPooling2D) (None, 7, 7, 512) 0 \n",
"_________________________________________________________________\n",
"flatten (Flatten) (None, 25088) 0 \n",
"_________________________________________________________________\n",
"fc1 (Dense) (None, 4096) 102764544 \n",
"_________________________________________________________________\n",
"fc2 (Dense) (None, 4096) 16781312 \n",
"_________________________________________________________________\n",
"predictions (Dense) (None, 1000) 4097000 \n",
"=================================================================\n",
"Total params: 138,357,544\n",
"Trainable params: 138,357,544\n",
"Non-trainable params: 0\n",
"_________________________________________________________________\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "3Rcr0FwU4jOc",
"colab_type": "text"
},
"source": [
"## প্রি-ট্রেইনড মডেল ResNet-50 এর লেয়ার\n",
"\n",
"কি দেখলাম? অনেকগুুলো কনভলিউশন নেটওয়ার্ক, ম্যাক্স পুলিং, আর শেষে তিনটা ডেন্স কানেক্টেড লেয়ার। বাকি আর কিছু বলছিনা এখন। দেখি আরেকটা নেটওয়ার্ক। \n",
"\n",
"পরেরটা ResNet-50, কি পার্থক্য দেখলাম?\n"
]
},
{
"cell_type": "code",
"metadata": {
"id": "QC01bvx665rq",
"colab_type": "code",
"outputId": "e3fbaf1d-0db6-4087-9051-7e6105ee1f44",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 1000
}
},
"source": [
"from tensorflow.keras.applications.resnet50 import ResNet50; ResNet50().summary()"
],
"execution_count": 0,
"outputs": [
{
"output_type": "stream",
"text": [
"Model: \"resnet50\"\n",
"__________________________________________________________________________________________________\n",
"Layer (type) Output Shape Param # Connected to \n",
"==================================================================================================\n",
"input_4 (InputLayer) [(None, 224, 224, 3) 0 \n",
"__________________________________________________________________________________________________\n",
"conv1_pad (ZeroPadding2D) (None, 230, 230, 3) 0 input_4[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv1_conv (Conv2D) (None, 112, 112, 64) 9472 conv1_pad[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv1_bn (BatchNormalization) (None, 112, 112, 64) 256 conv1_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv1_relu (Activation) (None, 112, 112, 64) 0 conv1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"pool1_pad (ZeroPadding2D) (None, 114, 114, 64) 0 conv1_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"pool1_pool (MaxPooling2D) (None, 56, 56, 64) 0 pool1_pad[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block1_1_conv (Conv2D) (None, 56, 56, 64) 4160 pool1_pool[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block1_1_bn (BatchNormali (None, 56, 56, 64) 256 conv2_block1_1_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block1_1_relu (Activation (None, 56, 56, 64) 0 conv2_block1_1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block1_2_conv (Conv2D) (None, 56, 56, 64) 36928 conv2_block1_1_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block1_2_bn (BatchNormali (None, 56, 56, 64) 256 conv2_block1_2_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block1_2_relu (Activation (None, 56, 56, 64) 0 conv2_block1_2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block1_0_conv (Conv2D) (None, 56, 56, 256) 16640 pool1_pool[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block1_3_conv (Conv2D) (None, 56, 56, 256) 16640 conv2_block1_2_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block1_0_bn (BatchNormali (None, 56, 56, 256) 1024 conv2_block1_0_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block1_3_bn (BatchNormali (None, 56, 56, 256) 1024 conv2_block1_3_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block1_add (Add) (None, 56, 56, 256) 0 conv2_block1_0_bn[0][0] \n",
" conv2_block1_3_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block1_out (Activation) (None, 56, 56, 256) 0 conv2_block1_add[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block2_1_conv (Conv2D) (None, 56, 56, 64) 16448 conv2_block1_out[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block2_1_bn (BatchNormali (None, 56, 56, 64) 256 conv2_block2_1_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block2_1_relu (Activation (None, 56, 56, 64) 0 conv2_block2_1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block2_2_conv (Conv2D) (None, 56, 56, 64) 36928 conv2_block2_1_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block2_2_bn (BatchNormali (None, 56, 56, 64) 256 conv2_block2_2_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block2_2_relu (Activation (None, 56, 56, 64) 0 conv2_block2_2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block2_3_conv (Conv2D) (None, 56, 56, 256) 16640 conv2_block2_2_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block2_3_bn (BatchNormali (None, 56, 56, 256) 1024 conv2_block2_3_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block2_add (Add) (None, 56, 56, 256) 0 conv2_block1_out[0][0] \n",
" conv2_block2_3_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block2_out (Activation) (None, 56, 56, 256) 0 conv2_block2_add[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block3_1_conv (Conv2D) (None, 56, 56, 64) 16448 conv2_block2_out[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block3_1_bn (BatchNormali (None, 56, 56, 64) 256 conv2_block3_1_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block3_1_relu (Activation (None, 56, 56, 64) 0 conv2_block3_1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block3_2_conv (Conv2D) (None, 56, 56, 64) 36928 conv2_block3_1_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block3_2_bn (BatchNormali (None, 56, 56, 64) 256 conv2_block3_2_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block3_2_relu (Activation (None, 56, 56, 64) 0 conv2_block3_2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block3_3_conv (Conv2D) (None, 56, 56, 256) 16640 conv2_block3_2_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block3_3_bn (BatchNormali (None, 56, 56, 256) 1024 conv2_block3_3_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block3_add (Add) (None, 56, 56, 256) 0 conv2_block2_out[0][0] \n",
" conv2_block3_3_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2_block3_out (Activation) (None, 56, 56, 256) 0 conv2_block3_add[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block1_1_conv (Conv2D) (None, 28, 28, 128) 32896 conv2_block3_out[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block1_1_bn (BatchNormali (None, 28, 28, 128) 512 conv3_block1_1_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block1_1_relu (Activation (None, 28, 28, 128) 0 conv3_block1_1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block1_2_conv (Conv2D) (None, 28, 28, 128) 147584 conv3_block1_1_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block1_2_bn (BatchNormali (None, 28, 28, 128) 512 conv3_block1_2_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block1_2_relu (Activation (None, 28, 28, 128) 0 conv3_block1_2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block1_0_conv (Conv2D) (None, 28, 28, 512) 131584 conv2_block3_out[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block1_3_conv (Conv2D) (None, 28, 28, 512) 66048 conv3_block1_2_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block1_0_bn (BatchNormali (None, 28, 28, 512) 2048 conv3_block1_0_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block1_3_bn (BatchNormali (None, 28, 28, 512) 2048 conv3_block1_3_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block1_add (Add) (None, 28, 28, 512) 0 conv3_block1_0_bn[0][0] \n",
" conv3_block1_3_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block1_out (Activation) (None, 28, 28, 512) 0 conv3_block1_add[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block2_1_conv (Conv2D) (None, 28, 28, 128) 65664 conv3_block1_out[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block2_1_bn (BatchNormali (None, 28, 28, 128) 512 conv3_block2_1_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block2_1_relu (Activation (None, 28, 28, 128) 0 conv3_block2_1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block2_2_conv (Conv2D) (None, 28, 28, 128) 147584 conv3_block2_1_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block2_2_bn (BatchNormali (None, 28, 28, 128) 512 conv3_block2_2_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block2_2_relu (Activation (None, 28, 28, 128) 0 conv3_block2_2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block2_3_conv (Conv2D) (None, 28, 28, 512) 66048 conv3_block2_2_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block2_3_bn (BatchNormali (None, 28, 28, 512) 2048 conv3_block2_3_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block2_add (Add) (None, 28, 28, 512) 0 conv3_block1_out[0][0] \n",
" conv3_block2_3_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block2_out (Activation) (None, 28, 28, 512) 0 conv3_block2_add[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block3_1_conv (Conv2D) (None, 28, 28, 128) 65664 conv3_block2_out[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block3_1_bn (BatchNormali (None, 28, 28, 128) 512 conv3_block3_1_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block3_1_relu (Activation (None, 28, 28, 128) 0 conv3_block3_1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block3_2_conv (Conv2D) (None, 28, 28, 128) 147584 conv3_block3_1_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block3_2_bn (BatchNormali (None, 28, 28, 128) 512 conv3_block3_2_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block3_2_relu (Activation (None, 28, 28, 128) 0 conv3_block3_2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block3_3_conv (Conv2D) (None, 28, 28, 512) 66048 conv3_block3_2_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block3_3_bn (BatchNormali (None, 28, 28, 512) 2048 conv3_block3_3_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block3_add (Add) (None, 28, 28, 512) 0 conv3_block2_out[0][0] \n",
" conv3_block3_3_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block3_out (Activation) (None, 28, 28, 512) 0 conv3_block3_add[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block4_1_conv (Conv2D) (None, 28, 28, 128) 65664 conv3_block3_out[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block4_1_bn (BatchNormali (None, 28, 28, 128) 512 conv3_block4_1_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block4_1_relu (Activation (None, 28, 28, 128) 0 conv3_block4_1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block4_2_conv (Conv2D) (None, 28, 28, 128) 147584 conv3_block4_1_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block4_2_bn (BatchNormali (None, 28, 28, 128) 512 conv3_block4_2_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block4_2_relu (Activation (None, 28, 28, 128) 0 conv3_block4_2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block4_3_conv (Conv2D) (None, 28, 28, 512) 66048 conv3_block4_2_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block4_3_bn (BatchNormali (None, 28, 28, 512) 2048 conv3_block4_3_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block4_add (Add) (None, 28, 28, 512) 0 conv3_block3_out[0][0] \n",
" conv3_block4_3_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv3_block4_out (Activation) (None, 28, 28, 512) 0 conv3_block4_add[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block1_1_conv (Conv2D) (None, 14, 14, 256) 131328 conv3_block4_out[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block1_1_bn (BatchNormali (None, 14, 14, 256) 1024 conv4_block1_1_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block1_1_relu (Activation (None, 14, 14, 256) 0 conv4_block1_1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block1_2_conv (Conv2D) (None, 14, 14, 256) 590080 conv4_block1_1_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block1_2_bn (BatchNormali (None, 14, 14, 256) 1024 conv4_block1_2_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block1_2_relu (Activation (None, 14, 14, 256) 0 conv4_block1_2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block1_0_conv (Conv2D) (None, 14, 14, 1024) 525312 conv3_block4_out[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block1_3_conv (Conv2D) (None, 14, 14, 1024) 263168 conv4_block1_2_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block1_0_bn (BatchNormali (None, 14, 14, 1024) 4096 conv4_block1_0_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block1_3_bn (BatchNormali (None, 14, 14, 1024) 4096 conv4_block1_3_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block1_add (Add) (None, 14, 14, 1024) 0 conv4_block1_0_bn[0][0] \n",
" conv4_block1_3_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block1_out (Activation) (None, 14, 14, 1024) 0 conv4_block1_add[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block2_1_conv (Conv2D) (None, 14, 14, 256) 262400 conv4_block1_out[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block2_1_bn (BatchNormali (None, 14, 14, 256) 1024 conv4_block2_1_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block2_1_relu (Activation (None, 14, 14, 256) 0 conv4_block2_1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block2_2_conv (Conv2D) (None, 14, 14, 256) 590080 conv4_block2_1_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block2_2_bn (BatchNormali (None, 14, 14, 256) 1024 conv4_block2_2_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block2_2_relu (Activation (None, 14, 14, 256) 0 conv4_block2_2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block2_3_conv (Conv2D) (None, 14, 14, 1024) 263168 conv4_block2_2_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block2_3_bn (BatchNormali (None, 14, 14, 1024) 4096 conv4_block2_3_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block2_add (Add) (None, 14, 14, 1024) 0 conv4_block1_out[0][0] \n",
" conv4_block2_3_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block2_out (Activation) (None, 14, 14, 1024) 0 conv4_block2_add[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block3_1_conv (Conv2D) (None, 14, 14, 256) 262400 conv4_block2_out[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block3_1_bn (BatchNormali (None, 14, 14, 256) 1024 conv4_block3_1_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block3_1_relu (Activation (None, 14, 14, 256) 0 conv4_block3_1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block3_2_conv (Conv2D) (None, 14, 14, 256) 590080 conv4_block3_1_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block3_2_bn (BatchNormali (None, 14, 14, 256) 1024 conv4_block3_2_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block3_2_relu (Activation (None, 14, 14, 256) 0 conv4_block3_2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block3_3_conv (Conv2D) (None, 14, 14, 1024) 263168 conv4_block3_2_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block3_3_bn (BatchNormali (None, 14, 14, 1024) 4096 conv4_block3_3_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block3_add (Add) (None, 14, 14, 1024) 0 conv4_block2_out[0][0] \n",
" conv4_block3_3_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block3_out (Activation) (None, 14, 14, 1024) 0 conv4_block3_add[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block4_1_conv (Conv2D) (None, 14, 14, 256) 262400 conv4_block3_out[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block4_1_bn (BatchNormali (None, 14, 14, 256) 1024 conv4_block4_1_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block4_1_relu (Activation (None, 14, 14, 256) 0 conv4_block4_1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block4_2_conv (Conv2D) (None, 14, 14, 256) 590080 conv4_block4_1_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block4_2_bn (BatchNormali (None, 14, 14, 256) 1024 conv4_block4_2_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block4_2_relu (Activation (None, 14, 14, 256) 0 conv4_block4_2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block4_3_conv (Conv2D) (None, 14, 14, 1024) 263168 conv4_block4_2_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block4_3_bn (BatchNormali (None, 14, 14, 1024) 4096 conv4_block4_3_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block4_add (Add) (None, 14, 14, 1024) 0 conv4_block3_out[0][0] \n",
" conv4_block4_3_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block4_out (Activation) (None, 14, 14, 1024) 0 conv4_block4_add[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block5_1_conv (Conv2D) (None, 14, 14, 256) 262400 conv4_block4_out[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block5_1_bn (BatchNormali (None, 14, 14, 256) 1024 conv4_block5_1_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block5_1_relu (Activation (None, 14, 14, 256) 0 conv4_block5_1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block5_2_conv (Conv2D) (None, 14, 14, 256) 590080 conv4_block5_1_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block5_2_bn (BatchNormali (None, 14, 14, 256) 1024 conv4_block5_2_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block5_2_relu (Activation (None, 14, 14, 256) 0 conv4_block5_2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block5_3_conv (Conv2D) (None, 14, 14, 1024) 263168 conv4_block5_2_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block5_3_bn (BatchNormali (None, 14, 14, 1024) 4096 conv4_block5_3_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block5_add (Add) (None, 14, 14, 1024) 0 conv4_block4_out[0][0] \n",
" conv4_block5_3_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block5_out (Activation) (None, 14, 14, 1024) 0 conv4_block5_add[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block6_1_conv (Conv2D) (None, 14, 14, 256) 262400 conv4_block5_out[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block6_1_bn (BatchNormali (None, 14, 14, 256) 1024 conv4_block6_1_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block6_1_relu (Activation (None, 14, 14, 256) 0 conv4_block6_1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block6_2_conv (Conv2D) (None, 14, 14, 256) 590080 conv4_block6_1_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block6_2_bn (BatchNormali (None, 14, 14, 256) 1024 conv4_block6_2_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block6_2_relu (Activation (None, 14, 14, 256) 0 conv4_block6_2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block6_3_conv (Conv2D) (None, 14, 14, 1024) 263168 conv4_block6_2_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block6_3_bn (BatchNormali (None, 14, 14, 1024) 4096 conv4_block6_3_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block6_add (Add) (None, 14, 14, 1024) 0 conv4_block5_out[0][0] \n",
" conv4_block6_3_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv4_block6_out (Activation) (None, 14, 14, 1024) 0 conv4_block6_add[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block1_1_conv (Conv2D) (None, 7, 7, 512) 524800 conv4_block6_out[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block1_1_bn (BatchNormali (None, 7, 7, 512) 2048 conv5_block1_1_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block1_1_relu (Activation (None, 7, 7, 512) 0 conv5_block1_1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block1_2_conv (Conv2D) (None, 7, 7, 512) 2359808 conv5_block1_1_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block1_2_bn (BatchNormali (None, 7, 7, 512) 2048 conv5_block1_2_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block1_2_relu (Activation (None, 7, 7, 512) 0 conv5_block1_2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block1_0_conv (Conv2D) (None, 7, 7, 2048) 2099200 conv4_block6_out[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block1_3_conv (Conv2D) (None, 7, 7, 2048) 1050624 conv5_block1_2_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block1_0_bn (BatchNormali (None, 7, 7, 2048) 8192 conv5_block1_0_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block1_3_bn (BatchNormali (None, 7, 7, 2048) 8192 conv5_block1_3_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block1_add (Add) (None, 7, 7, 2048) 0 conv5_block1_0_bn[0][0] \n",
" conv5_block1_3_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block1_out (Activation) (None, 7, 7, 2048) 0 conv5_block1_add[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block2_1_conv (Conv2D) (None, 7, 7, 512) 1049088 conv5_block1_out[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block2_1_bn (BatchNormali (None, 7, 7, 512) 2048 conv5_block2_1_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block2_1_relu (Activation (None, 7, 7, 512) 0 conv5_block2_1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block2_2_conv (Conv2D) (None, 7, 7, 512) 2359808 conv5_block2_1_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block2_2_bn (BatchNormali (None, 7, 7, 512) 2048 conv5_block2_2_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block2_2_relu (Activation (None, 7, 7, 512) 0 conv5_block2_2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block2_3_conv (Conv2D) (None, 7, 7, 2048) 1050624 conv5_block2_2_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block2_3_bn (BatchNormali (None, 7, 7, 2048) 8192 conv5_block2_3_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block2_add (Add) (None, 7, 7, 2048) 0 conv5_block1_out[0][0] \n",
" conv5_block2_3_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block2_out (Activation) (None, 7, 7, 2048) 0 conv5_block2_add[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block3_1_conv (Conv2D) (None, 7, 7, 512) 1049088 conv5_block2_out[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block3_1_bn (BatchNormali (None, 7, 7, 512) 2048 conv5_block3_1_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block3_1_relu (Activation (None, 7, 7, 512) 0 conv5_block3_1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block3_2_conv (Conv2D) (None, 7, 7, 512) 2359808 conv5_block3_1_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block3_2_bn (BatchNormali (None, 7, 7, 512) 2048 conv5_block3_2_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block3_2_relu (Activation (None, 7, 7, 512) 0 conv5_block3_2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block3_3_conv (Conv2D) (None, 7, 7, 2048) 1050624 conv5_block3_2_relu[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block3_3_bn (BatchNormali (None, 7, 7, 2048) 8192 conv5_block3_3_conv[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block3_add (Add) (None, 7, 7, 2048) 0 conv5_block2_out[0][0] \n",
" conv5_block3_3_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv5_block3_out (Activation) (None, 7, 7, 2048) 0 conv5_block3_add[0][0] \n",
"__________________________________________________________________________________________________\n",
"avg_pool (GlobalAveragePooling2 (None, 2048) 0 conv5_block3_out[0][0] \n",
"__________________________________________________________________________________________________\n",
"probs (Dense) (None, 1000) 2049000 avg_pool[0][0] \n",
"==================================================================================================\n",
"Total params: 25,636,712\n",
"Trainable params: 25,583,592\n",
"Non-trainable params: 53,120\n",
"__________________________________________________________________________________________________\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "5lA4HADZ7O9A",
"colab_type": "text"
},
"source": [
"কি দেখলেন? শেষ আউটপুট লেয়ারের আগে একটা নতুন এন্ট্রি 'avg_pool (GlobalAveragePooling2D)', চলুন আলাপ করি। শুরুতেই আরেকটা ডেটাসেট, সিএনএন দিয়ে। "
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "y3gF-hFNBtKg",
"colab_type": "text"
},
"source": [
"এই নতুন নেটওয়ার্ক তৈরি করতে আজকে আমরা ব্যবহার করব আরেকটা নতুন ডাটাসেট,যা বেশিরভাগ ডিপ লার্নিং এক্সপার্ট ব্যবহার করেছেন তার শেখার শুরুতে। বেড়াল এবং কুকুরের ছবিকে ঠিকমতো ক্লাসিফাই করার জন্য এই ডাটাসেটটা ব্যবহার শুরু হয় সেই ২০১৩ সালে। এর শুরুটা হয় একটা ক্যাগল মেশিন লার্নিং কম্পিটিশনে। এই কাজের পেছনে ছিল মাইক্রোসফট এবং পেট ফাইন্ডার বলে একটা কোম্পানির পার্টনারশিপ। আমরা যে ডাটাসেটটা ব্যবহার করব সেটা প্রায় ৩০ লক্ষ ছবির একটা ম্যানুয়াল ‘লেবেলড’ ডাটাবেজ। মানুষ নিজে হাতে করেছে। তবে আমরা সেটার একটা ‘সাবসেট’ ব্যবহার করব যার জন্য এত ছবি না হলেও চলে। শুরুতে আমরা একটা স্ট্যান্ডার্ড ‘সিএনএন’ তৈরি করলেও তার শেষে গ্লোবাল অ্যাভারেজ পুলিং অংশটুকু জুড়ে দেব আপনাদের বোঝার জন্য। আমাদেরকে একটা ছবি দিয়ে বলতে হবে সেটা আসলে একটা বিড়ালের না কুকুরের ছবি? এবং সেটার অ্যাক্যুরেসি সাধারণ ‘সিএনএন’ থেকে কতো ভালো?"
]
},
{
"cell_type": "code",
"metadata": {
"id": "wyuDd_3a_wYl",
"colab_type": "code",
"colab": {}
},
"source": [
"# লোড করে নেই হেল্পার লাইব্রেরি, টেন্সর-ফ্লো ডেটাসেট এপিআই\n",
"\n",
"import tensorflow as tf\n",
"from tensorflow.keras import layers\n",
"import tensorflow_datasets as tfds\n",
"import matplotlib.pylab as plt\n",
"import datetime as dt"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "kQjvvHB8DBs5",
"colab_type": "text"
},
"source": [
"## গ্লোবাল অ্যাভারেজ পুলিং\n",
"\n",
"আমরা যখন এক লেয়ার থেকে আরেক লেয়ারে ডাটা পাঠাই, তখন লেয়ার ভিত্তিক ক্যালকুলেশনে যদি এমন কিছু করতে পারি যাতে সামনের দিকের ডাটাগুলো কমে আসে। এটা ভালো হবে না? এমনিতেই একেকটা নিউরাল নেটওয়ার্ককে ট্রেইন করতে যে সময় লাগে, তার পাশাপাশি সেই কম্পিউটিং রিসোর্স জোগাড় করা যেরকম কষ্টকর - সেই দিক থেকে বিবেচনা করতে গেলে ‘গ্লোবাল অ্যাভারেজ পুলিং’ আগের লেয়ারের প্রতিটা ফিচার ম্যাপের আউটপুটের গড়কে ক্যালকুলেট করে পাঠায় সামনের লেয়ারে। এই সহজ কাজটা আমাদের ডাটা ফ্লোকে অনেকটাই কমিয়ে নিয়ে আসে আমাদের শেষ ক্লাসিফিকেশন লেয়ারের জন্য। এতে ট্রেনিং দেবার মতো আর কোন প্যারামিটার থাকেনা, কিছুটা ম্যাক্স পুলিং এর মতো। আমরা এখানে একটা ছবি দেখতে পারি।\n",
"\n",
" চিত্রঃ ‘গ্লোবাল অ্যাভারেজ পুলিং’ লেয়ার চলে আসছে আউটপুটের আগের লেয়ারে"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "3R-EYlZuO9qQ",
"colab_type": "text"
},
"source": [
"আপনারা দেখুন - সবশেষ লেয়ারে যেখানে আমরা জিনিসপত্র ক্লাসিফাই করছি তার আগেই এই গ্লোবাল এভারেজ পুলিং লেয়ারটা দেখতে পাচ্ছি। সেটা থেকে আউটপুট চলে যাচ্ছে আগের মত সফটম্যাক্স লেয়ারে। আমাদের এই আর্কিটেকচারে ৬৪টা গড় ক্যালকুলেশন আসছে সেই ৬৪ এবং ৭ x ৭ চ্যানেল থেকে - যেটা আমরা দেখছি দ্বিতীয় কনভলিউশনাল লেয়ারের আউটপুট থেকে আসছে। আমাদের এই ‘গ্যাপ’ মানে ‘গ্লোবাল অ্যাভারেজ পুলিং’ লেয়ার আগের ৭ x ৭, ৬৪ ডাইমেনশন থেকে ১ x ১, ৬৪ ডাইমেনশনে যা আসলে আসছে ৭ x ৭ চ্যানেলের গড় আউটপুট থেকে। অনেকে বলতে পারেন শেষের ‘ফুললি কানেক্টেড লেয়ারগুলো থেকে গ্লোবাল অ্যাভারেজ পুলিং লেয়ার কেন ভালো? \n",
"\n",
"১. সবচেয়ে বড় কথা হচ্ছে আমরা মডেলের শেষ লেয়ারগুলোতে ট্রেনিং করার মত প্যারামিটারগুলোকে বাদ দিয়ে দিচ্ছি। একটা ‘ফুললি কানেক্টেড’ অথবা ‘ডেন্স’ লেয়ারে প্রচুর প্যারামিটার থাকে সেখানে আমাদের আগের ছবিতে একটা ৭ x ৭, ৬৪ ডাইমেনশনের ‘সিএনএন’ আউটপুটকে প্ল্যান করে সেটাকে ঢুকিয়ে দেয়া হচ্ছে ৫০০ নোডের একটা ডেন্স লেয়ারে যেটা প্রায় ১৫ কোটি ‘ওয়েট’ বের করে দিচ্ছে আমাদের ট্রেনিংয়ের জন্য। এটা একটা বিশাল কাজের লোড। এই জিনিসগুলোকে বাদ দেয়াতে আমাদের ট্রেনিং স্পিড বেড়ে যাবে অনেক গুন। \n",
"\n",
"২. যেহেতু আমরা এতগুলো ট্রেনিং দেবার মতো প্যারামিটারগুলোকে ফেলে দিচ্ছি সেটা অনেকটাই কমিয়ে দেবে ‘ওভার ফিটিং’। সেখানে আমরা ‘ফুললি কানেক্টেড’ লেয়ারে ড্রপ আউট ব্যবহার করা যেতো। ‘ড্রপ আউট’ নিয়ে আলোচনা করা যাবে সামনে। \n",
"\n",
"৩. ‘গ্লোবাল অ্যাভারেজ পুলিং’ নিয়ে প্রচুর আলাপ হয়েছে রিসার্চে। দেখা গেছে আমরা যদি এই ‘ফুললি কানেক্টেড ক্লাসিফিকেশন লেয়ার ফেলে দেই তাহলে ফিচার ম্যাপগুলো খুব কাছাকাছি কাজ করতে পারে তাদের ক্লাসিফিকেশন ক্যাটেগরি বানানোর জন্য। কিছুটা ডাইরেক্ট কানেকশন। এখানে আমরা বলতে পারি আমাদের প্রতিটা ফিচার ম্যাপ একেকটা ক্যাটেগরির ‘কনফিডেন্স ম্যাপ’ হিসেবে দাঁড়িয়ে গেছে।\n",
"\n",
"৪. সবচেয়ে বড় কথা হচ্ছে ফিচার ম্যাপগুলোর গড় অপারেশন আমাদের মডেলকে আরো শক্তিশালী করে তুলছে ডাটার ভেতরের ইন্টারনাল ট্রান্সলেশন কাছাকাছি হবার কারণে। যতক্ষণ দরকারি ফিচারগুলোকে যোগ করা বা এক্টিভেট করা হচ্ছে ফিচার ম্যাপে, সেই জিনিসগুলো চলে আসছে সেই গড় অপারেশনে।\n",
"\n",
"আমাদের একটা সাধারণ ‘কনভলিউশনাল নিউরাল নেটওয়ার্ক’ এবং ‘গ্লোবাল এভারেজ পুলিং’ এর মধ্যে লস এবং অ্যাক্যুরেসি বের করে দেখব কোনটা ভালো কাজ করে।"
]
},
{
"cell_type": "code",
"metadata": {
"id": "IlKXX2DuJK6D",
"colab_type": "code",
"outputId": "028bb0f5-fceb-45f9-a2df-56c3732dd710",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 52
}
},
"source": [
"# দরকার টেন্সর-ফ্লো কলব্যাক ফিচার\n",
"%load_ext tensorboard"
],
"execution_count": 0,
"outputs": [
{
"output_type": "stream",
"text": [
"The tensorboard extension is already loaded. To reload it, use:\n",
" %reload_ext tensorboard\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "A_bx27_1DrkS",
"colab_type": "code",
"colab": {}
},
"source": [
"from tensorflow.keras.callbacks import TensorBoard\n",
"log_dir='log/'"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "M59ag156CDoc",
"colab_type": "code",
"colab": {}
},
"source": [
"# আগের সব ট্রেনিং এর লগ ফেলে দিচ্ছি \n",
"!rm -rf log/ "
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "ENeoSPCJRSVm",
"colab_type": "text"
},
"source": [
"## বিড়াল এবং কুকুরের ডাটাসেটের উপর ‘গ্লোবাল এভারেজ পুলিং’\n",
"\n",
"শুরুতেই আমরা সাহায্য নেব ‘টেন্সর-ফ্লো’ ডাটাসেট এপিআই থেকে। কল করছি ডাটাসেট অবজেক্টের ভেতর থেকে। তবে, এটাকে দরকার মতো তিনটা টুপলে ভাগ করছি। ৮০, ১০, ১০ যার থেকে ৮০% ট্রেনিং ১০% ভ্যালিডেশন আর বাকি ১০% টেস্টসেট হিসেবে রাখবো। শুরুতে আমরা ডাটা সেটের ইনফরমেশন দেখি। আমাদের স্প্লিট অবজেক্ট ডাটাসেট লোডারকে বলছে কিভাবে ডাটাসেটকে সে তিন ভাগে ভাগ করবে। আমরা তাদেরকে ‘সুপারভাইজড লার্নিং’ মডেলের জন্য ব্যবহার করবো কিনা সেখানে ‘হ্যাঁ’ করে দিচ্ছি। এখানে আমরা ডাটা নিয়ে আসছি লেবেলসহ। কি ডাটা নিয়ে এলাম সেটা দেখি একটু?\n"
]
},
{
"cell_type": "code",
"metadata": {
"id": "mJM3-v4s_1Yf",
"colab_type": "code",
"colab": {}
},
"source": [
"split = (80, 10, 10)\n",
"cat_dog_spl = tfds.Split.TRAIN.subsplit(weighted=split)\n",
"\n",
"(cat_train, cat_valid, cat_test), info = tfds.load('cats_vs_dogs', split=list(cat_dog_spl), with_info=True, as_supervised=True)"
],
"execution_count": 0,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "UMeuSM1pRxH5",
"colab_type": "code",
"outputId": "930d6bb0-ee11-4944-a1f4-2f3d7b990924",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 514
}
},
"source": [
"# ডেটাসেটের ইনফো পড়ে দেখি \n",
"\n",
"print(info)"
],
"execution_count": 0,
"outputs": [
{
"output_type": "stream",
"text": [
"tfds.core.DatasetInfo(\n",
" name='cats_vs_dogs',\n",
" version=2.0.1,\n",
" description='A large set of images of cats and dogs.There are 1738 corrupted images that are dropped.',\n",
" urls=['https://www.microsoft.com/en-us/download/details.aspx?id=54765'],\n",
" features=FeaturesDict({\n",
" 'image': Image(shape=(None, None, 3), dtype=tf.uint8),\n",
" 'image/filename': Text(shape=(), dtype=tf.string),\n",
" 'label': ClassLabel(shape=(), dtype=tf.int64, num_classes=2),\n",
" }),\n",
" total_num_examples=23262,\n",
" splits={\n",
" 'train': 23262,\n",
" },\n",
" supervised_keys=('image', 'label'),\n",
" citation=\"\"\"@Inproceedings (Conference){asirra-a-captcha-that-exploits-interest-aligned-manual-image-categorization,\n",
" author = {Elson, Jeremy and Douceur, John (JD) and Howell, Jon and Saul, Jared},\n",
" title = {Asirra: A CAPTCHA that Exploits Interest-Aligned Manual Image Categorization},\n",
" booktitle = {Proceedings of 14th ACM Conference on Computer and Communications Security (CCS)},\n",
" year = {2007},\n",
" month = {October},\n",
" publisher = {Association for Computing Machinery, Inc.},\n",
" url = {https://www.microsoft.com/en-us/research/publication/asirra-a-captcha-that-exploits-interest-aligned-manual-image-categorization/},\n",
" edition = {Proceedings of 14th ACM Conference on Computer and Communications Security (CCS)},\n",
" }\"\"\",\n",
" redistribution_info=,\n",
")\n",
"\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "c8Rl4kG42AXm",
"colab_type": "code",
"outputId": "833f778c-b0af-4228-bb92-73a71d5d1b87",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 521
}
},
"source": [
"# for image, label in cat_train.take(2):\n",
"for image, label in cat_test.take(2):\n",
" plt.figure()\n",
" plt.imshow(image)"
],
"execution_count": 0,
"outputs": [
{
"output_type": "display_data",
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAANcAAAD8CAYAAADkFjFAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
"text/plain": [
"