Skip to content

Instantly share code, notes, and snippets.

@sparkingdark

sparkingdark/copy-of-indian-dance-type-image-classification-using-vgg1.ipynb

Created June 24, 2020 14:54
Show Gist options
  • Save sparkingdark/4ef2df80b581ca3de2a02bb0cb1ae39a to your computer and use it in GitHub Desktop.
Save sparkingdark/4ef2df80b581ca3de2a02bb0cb1ae39a to your computer and use it in GitHub Desktop.
Download ZIP
Copy of Indian dance type image classification using VGG1.ipynb
Display the source blob
Display the rendered blob
Raw
copy-of-indian-dance-type-image-classification-using-vgg1.ipynb
{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"kernelspec": {
"language": "python",
"display_name": "Python 3",
"name": "python3"
},
"language_info": {
"pygments_lexer": "ipython3",
"nbconvert_exporter": "python",
"version": "3.6.4",
"file_extension": ".py",
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"name": "python",
"mimetype": "text/x-python"
},
"colab": {
"name": "Copy of Indian dance type image classification using VGG1.ipynb",
"provenance": [],
"toc_visible": true,
"include_colab_link": true
},
"accelerator": "GPU"
},
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "view-in-github",
"colab_type": "text"
},
"source": [
"<a href=\"https://colab.research.google.com/gist/sparkingdark/4ef2df80b581ca3de2a02bb0cb1ae39a/copy-of-indian-dance-type-image-classification-using-vgg1.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "DEqLQz-9kGJH",
"colab_type": "text"
},
"source": [
"# HackerEarth Deep Learning challenge: Identify the dance form\n",
"This International Dance Day, an event management company organized an evening of Indian classical dance performances to celebrate the rich, eloquent, and elegant art of dance. Post the event, the company planned to create a microsite to promote and raise awareness among the public about these dance forms. However, identifying them from images is a tough nut to crack.\n",
"You have been appointed as a Machine Learning Engineer for this project. Build an image tagging Deep Learning model that can help the company classify these images into eight categories of Indian classical dance.\n",
"\n",
"### Dataset\n",
"The dataset consists of 364 images belonging to 8 categories, namely manipuri, bharatanatyam, odissi, kathakali, kathak, sattriya, kuchipudi, and mohiniyattam.\n",
"The benefits of practicing this problem by using Machine Learning/Deep Learning techniques are as follows:\n",
"This challenge will encourage you to apply your Machine Learning skills to build models that classify images into multiple categories\n",
"This challenge will help you enhance your knowledge of classification actively. It is one of the basic building blocks of Machine Learning and Deep Learning\n",
"We challenge you to build a model that auto-tags images and classifies them into various categories of Indian classical dance forms.\n",
"\n",
"The data folder consists of two folders and two .csv files. The details are as follows:\n",
"train: Contains 364 images for 8 classes \n",
"* manipuri,\n",
"* bharatanatyam\n",
"* odissi\n",
"* kathakali\n",
"* kathak\n",
"* sattriya\n",
"* kuchipudi\n",
"* mohiniyattam\n",
"\n",
"test: Contains 156 images\n",
"train.csv: 364 x 2\n",
"test.csv: 156 x 1\n",
"\n",
"Data description\n",
"This data set consists of the following two columns:\n",
"\n",
"| Column Name | Description |\n",
"|-------------|-------------|\n",
"| Image | Name of Image| \n",
"|target |Category of Image ['manipuri','bharatanatyam','odissi','kathakali','kathak','sattriya','kuchipudi','mohiniyattam'] |\n"
]
},
{
"cell_type": "code",
"metadata": {
"id": "jmmDZipEkME2",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 224
},
"outputId": "7ea21b8f-7acf-49a1-e2a6-60dfb02c55b9"
},
"source": [
"!wget https://he-s3.s3.amazonaws.com/media/hackathon/hackerearth-deep-learning-challenge-identify-dance-form/identify-the-dance-form-deea77f8/0664343c9a8f11ea.zip\n",
"ROOT = '/content/dataset/'\n",
"test_dir = ROOT+'test'\n",
"train_dir = ROOT+'train'"
],
"execution_count": 1,
"outputs": [
{
"output_type": "stream",
"text": [
"--2020-06-24 13:41:07-- https://he-s3.s3.amazonaws.com/media/hackathon/hackerearth-deep-learning-challenge-identify-dance-form/identify-the-dance-form-deea77f8/0664343c9a8f11ea.zip\n",
"Resolving he-s3.s3.amazonaws.com (he-s3.s3.amazonaws.com)... 52.219.125.20\n",
"Connecting to he-s3.s3.amazonaws.com (he-s3.s3.amazonaws.com)|52.219.125.20|:443... connected.\n",
"HTTP request sent, awaiting response... 200 OK\n",
"Length: 27882564 (27M) [application/zip]\n",
"Saving to: ‘0664343c9a8f11ea.zip’\n",
"\n",
"0664343c9a8f11ea.zi 100%[===================>] 26.59M 3.83MB/s in 16s \n",
"\n",
"2020-06-24 13:41:24 (1.71 MB/s) - ‘0664343c9a8f11ea.zip’ saved [27882564/27882564]\n",
"\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "SfqyRjAxkuLv",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 1000
},
"outputId": "8ac23548-11c2-42af-d3bb-a8dc415b5f6d"
},
"source": [
"!unzip test.zip"
],
"execution_count": 2,
"outputs": [
{
"output_type": "stream",
"text": [
"Archive: test.zip\n",
" creating: dataset/\n",
" creating: dataset/test/\n",
" inflating: dataset/test/508.jpg \n",
" inflating: dataset/test/246.jpg \n",
" inflating: dataset/test/473.jpg \n",
" inflating: dataset/test/485.jpg \n",
" inflating: dataset/test/128.jpg \n",
" inflating: dataset/test/410.jpg \n",
" inflating: dataset/test/465.jpg \n",
" inflating: dataset/test/196.jpg \n",
" inflating: dataset/test/340.jpg \n",
" inflating: dataset/test/467.jpg \n",
" inflating: dataset/test/306.jpg \n",
" inflating: dataset/test/171.jpg \n",
" inflating: dataset/test/323.jpg \n",
" inflating: dataset/test/332.jpg \n",
" inflating: dataset/test/330.jpg \n",
" inflating: dataset/test/170.jpg \n",
" inflating: dataset/test/479.jpg \n",
" inflating: dataset/test/9.jpg \n",
" inflating: dataset/test/283.jpg \n",
" inflating: dataset/test/114.jpg \n",
" inflating: dataset/test/507.jpg \n",
" inflating: dataset/test/183.jpg \n",
" inflating: dataset/test/200.jpg \n",
" inflating: dataset/test/457.jpg \n",
" inflating: dataset/test/264.jpg \n",
" inflating: dataset/test/194.jpg \n",
" inflating: dataset/test/518.jpg \n",
" inflating: dataset/test/175.jpg \n",
" inflating: dataset/test/64.jpg \n",
" inflating: dataset/test/152.jpg \n",
" inflating: dataset/test/270.jpg \n",
" inflating: dataset/test/125.jpg \n",
" inflating: dataset/test/38.jpg \n",
" inflating: dataset/test/440.jpg \n",
" inflating: dataset/test/302.jpg \n",
" inflating: dataset/test/394.jpg \n",
" inflating: dataset/test/448.jpg \n",
" inflating: dataset/test/463.jpg \n",
" inflating: dataset/test/20.jpg \n",
" inflating: dataset/test/301.jpg \n",
" inflating: dataset/test/28.jpg \n",
" inflating: dataset/test/61.jpg \n",
" inflating: dataset/test/398.jpg \n",
" inflating: dataset/test/349.jpg \n",
" inflating: dataset/test/271.jpg \n",
" inflating: dataset/test/286.jpg \n",
" inflating: dataset/test/431.jpg \n",
" inflating: dataset/test/316.jpg \n",
" inflating: dataset/test/224.jpg \n",
" inflating: dataset/test/411.jpg \n",
" inflating: dataset/test/278.jpg \n",
" inflating: dataset/test/456.jpg \n",
" inflating: dataset/test/45.jpg \n",
" inflating: dataset/test/324.jpg \n",
" inflating: dataset/test/6.jpg \n",
" inflating: dataset/test/59.jpg \n",
" inflating: dataset/test/86.jpg \n",
" inflating: dataset/test/470.jpg \n",
" inflating: dataset/test/374.jpg \n",
" inflating: dataset/test/132.jpg \n",
" inflating: dataset/test/216.jpg \n",
" inflating: dataset/test/95.jpg \n",
" inflating: dataset/test/17.jpg \n",
" inflating: dataset/test/162.jpg \n",
" inflating: dataset/test/346.jpg \n",
" inflating: dataset/test/439.jpg \n",
" inflating: dataset/test/232.jpg \n",
" inflating: dataset/test/150.jpg \n",
" inflating: dataset/test/401.jpg \n",
" inflating: dataset/test/145.jpg \n",
" inflating: dataset/test/23.jpg \n",
" inflating: dataset/test/514.jpg \n",
" inflating: dataset/test/193.jpg \n",
" inflating: dataset/test/416.jpg \n",
" inflating: dataset/test/447.jpg \n",
" inflating: dataset/test/220.jpg \n",
" inflating: dataset/test/505.jpg \n",
" inflating: dataset/test/147.jpg \n",
" inflating: dataset/test/210.jpg \n",
" inflating: dataset/test/40.jpg \n",
" inflating: dataset/test/284.jpg \n",
" inflating: dataset/test/55.jpg \n",
" inflating: dataset/test/378.jpg \n",
" inflating: dataset/test/520.jpg \n",
" inflating: dataset/test/484.jpg \n",
" inflating: dataset/test/112.jpg \n",
" inflating: dataset/test/360.jpg \n",
" inflating: dataset/test/317.jpg \n",
" inflating: dataset/test/71.jpg \n",
" inflating: dataset/test/202.jpg \n",
" inflating: dataset/test/92.jpg \n",
" inflating: dataset/test/11.jpg \n",
" inflating: dataset/test/67.jpg \n",
" inflating: dataset/test/138.jpg \n",
" inflating: dataset/test/506.jpg \n",
" inflating: dataset/test/146.jpg \n",
" inflating: dataset/test/318.jpg \n",
" inflating: dataset/test/198.jpg \n",
" inflating: dataset/test/375.jpg \n",
" inflating: dataset/test/41.jpg \n",
" inflating: dataset/test/421.jpg \n",
" inflating: dataset/test/129.jpg \n",
" inflating: dataset/test/14.jpg \n",
" inflating: dataset/test/460.jpg \n",
" inflating: dataset/test/34.jpg \n",
" inflating: dataset/test/380.jpg \n",
" inflating: dataset/test/238.jpg \n",
" inflating: dataset/test/392.jpg \n",
" inflating: dataset/test/80.jpg \n",
" inflating: dataset/test/517.jpg \n",
" inflating: dataset/test/13.jpg \n",
" inflating: dataset/test/189.jpg \n",
" inflating: dataset/test/390.jpg \n",
" inflating: dataset/test/487.jpg \n",
" inflating: dataset/test/438.jpg \n",
" inflating: dataset/test/482.jpg \n",
" inflating: dataset/test/206.jpg \n",
" inflating: dataset/test/217.jpg \n",
" inflating: dataset/test/285.jpg \n",
" inflating: dataset/test/385.jpg \n",
" inflating: dataset/test/495.jpg \n",
" inflating: dataset/test/269.jpg \n",
" inflating: dataset/test/429.jpg \n",
" inflating: dataset/test/291.jpg \n",
" inflating: dataset/test/461.jpg \n",
" inflating: dataset/test/256.jpg \n",
" inflating: dataset/test/48.jpg \n",
" inflating: dataset/test/290.jpg \n",
" inflating: dataset/test/437.jpg \n",
" inflating: dataset/test/498.jpg \n",
" inflating: dataset/test/173.jpg \n",
" inflating: dataset/test/280.jpg \n",
" inflating: dataset/test/119.jpg \n",
" inflating: dataset/test/363.jpg \n",
" inflating: dataset/test/97.jpg \n",
" inflating: dataset/test/414.jpg \n",
" inflating: dataset/test/386.jpg \n",
" inflating: dataset/test/424.jpg \n",
" inflating: dataset/test/157.jpg \n",
" inflating: dataset/test/33.jpg \n",
" inflating: dataset/test/354.jpg \n",
" inflating: dataset/test/57.jpg \n",
" inflating: dataset/test/108.jpg \n",
" inflating: dataset/test/239.jpg \n",
" inflating: dataset/test/441.jpg \n",
" inflating: dataset/test/215.jpg \n",
" inflating: dataset/test/373.jpg \n",
" inflating: dataset/test/292.jpg \n",
" inflating: dataset/test/110.jpg \n",
" inflating: dataset/test/510.jpg \n",
" inflating: dataset/test/275.jpg \n",
" inflating: dataset/test/366.jpg \n",
" inflating: dataset/test/226.jpg \n",
" inflating: dataset/test/35.jpg \n",
" inflating: dataset/test/458.jpg \n",
" inflating: dataset/test/358.jpg \n",
" creating: dataset/train/\n",
" inflating: dataset/train/96.jpg \n",
" inflating: dataset/train/163.jpg \n",
" inflating: dataset/train/450.jpg \n",
" inflating: dataset/train/219.jpg \n",
" inflating: dataset/train/455.jpg \n",
" inflating: dataset/train/46.jpg \n",
" inflating: dataset/train/326.jpg \n",
" inflating: dataset/train/408.jpg \n",
" inflating: dataset/train/365.jpg \n",
" inflating: dataset/train/422.jpg \n",
" inflating: dataset/train/397.jpg \n",
" inflating: dataset/train/400.jpg \n",
" inflating: dataset/train/32.jpg \n",
" inflating: dataset/train/350.jpg \n",
" inflating: dataset/train/234.jpg \n",
" inflating: dataset/train/492.jpg \n",
" inflating: dataset/train/333.jpg \n",
" inflating: dataset/train/376.jpg \n",
" inflating: dataset/train/103.jpg \n",
" inflating: dataset/train/39.jpg \n",
" inflating: dataset/train/87.jpg \n",
" inflating: dataset/train/155.jpg \n",
" inflating: dataset/train/388.jpg \n",
" inflating: dataset/train/384.jpg \n",
" inflating: dataset/train/248.jpg \n",
" inflating: dataset/train/494.jpg \n",
" inflating: dataset/train/47.jpg \n",
" inflating: dataset/train/488.jpg \n",
" inflating: dataset/train/462.jpg \n",
" inflating: dataset/train/347.jpg \n",
" inflating: dataset/train/242.jpg \n",
" inflating: dataset/train/16.jpg \n",
" inflating: dataset/train/178.jpg \n",
" inflating: dataset/train/348.jpg \n",
" inflating: dataset/train/164.jpg \n",
" inflating: dataset/train/2.jpg \n",
" inflating: dataset/train/141.jpg \n",
" inflating: dataset/train/187.jpg \n",
" inflating: dataset/train/464.jpg \n",
" inflating: dataset/train/472.jpg \n",
" inflating: dataset/train/489.jpg \n",
" inflating: dataset/train/387.jpg \n",
" inflating: dataset/train/109.jpg \n",
" inflating: dataset/train/255.jpg \n",
" inflating: dataset/train/274.jpg \n",
" inflating: dataset/train/509.jpg \n",
" inflating: dataset/train/19.jpg \n",
" inflating: dataset/train/102.jpg \n",
" inflating: dataset/train/24.jpg \n",
" inflating: dataset/train/294.jpg \n",
" inflating: dataset/train/501.jpg \n",
" inflating: dataset/train/335.jpg \n",
" inflating: dataset/train/321.jpg \n",
" inflating: dataset/train/480.jpg \n",
" inflating: dataset/train/393.jpg \n",
" inflating: dataset/train/344.jpg \n",
" inflating: dataset/train/259.jpg \n",
" inflating: dataset/train/76.jpg \n",
" inflating: dataset/train/343.jpg \n",
" inflating: dataset/train/4.jpg \n",
" inflating: dataset/train/486.jpg \n",
" inflating: dataset/train/361.jpg \n",
" inflating: dataset/train/90.jpg \n",
" inflating: dataset/train/231.jpg \n",
" inflating: dataset/train/331.jpg \n",
" inflating: dataset/train/243.jpg \n",
" inflating: dataset/train/104.jpg \n",
" inflating: dataset/train/113.jpg \n",
" inflating: dataset/train/21.jpg \n",
" inflating: dataset/train/15.jpg \n",
" inflating: dataset/train/370.jpg \n",
" inflating: dataset/train/124.jpg \n",
" inflating: dataset/train/241.jpg \n",
" inflating: dataset/train/56.jpg \n",
" inflating: dataset/train/296.jpg \n",
" inflating: dataset/train/81.jpg \n",
" inflating: dataset/train/29.jpg \n",
" inflating: dataset/train/423.jpg \n",
" inflating: dataset/train/445.jpg \n",
" inflating: dataset/train/406.jpg \n",
" inflating: dataset/train/116.jpg \n",
" inflating: dataset/train/407.jpg \n",
" inflating: dataset/train/118.jpg \n",
" inflating: dataset/train/367.jpg \n",
" inflating: dataset/train/30.jpg \n",
" inflating: dataset/train/395.jpg \n",
" inflating: dataset/train/265.jpg \n",
" inflating: dataset/train/73.jpg \n",
" inflating: dataset/train/277.jpg \n",
" inflating: dataset/train/122.jpg \n",
" inflating: dataset/train/516.jpg \n",
" inflating: dataset/train/377.jpg \n",
" inflating: dataset/train/513.jpg \n",
" inflating: dataset/train/405.jpg \n",
" inflating: dataset/train/172.jpg \n",
" inflating: dataset/train/299.jpg \n",
" inflating: dataset/train/149.jpg \n",
" inflating: dataset/train/165.jpg \n",
" inflating: dataset/train/371.jpg \n",
" inflating: dataset/train/18.jpg \n",
" inflating: dataset/train/213.jpg \n",
" inflating: dataset/train/303.jpg \n",
" inflating: dataset/train/353.jpg \n",
" inflating: dataset/train/135.jpg \n",
" inflating: dataset/train/481.jpg \n",
" inflating: dataset/train/3.jpg \n",
" inflating: dataset/train/329.jpg \n",
" inflating: dataset/train/345.jpg \n",
" inflating: dataset/train/443.jpg \n",
" inflating: dataset/train/504.jpg \n",
" inflating: dataset/train/12.jpg \n",
" inflating: dataset/train/74.jpg \n",
" inflating: dataset/train/185.jpg \n",
" inflating: dataset/train/161.jpg \n",
" inflating: dataset/train/309.jpg \n",
" inflating: dataset/train/205.jpg \n",
" inflating: dataset/train/379.jpg \n",
" inflating: dataset/train/82.jpg \n",
" inflating: dataset/train/276.jpg \n",
" inflating: dataset/train/154.jpg \n",
" inflating: dataset/train/254.jpg \n",
" inflating: dataset/train/26.jpg \n",
" inflating: dataset/train/181.jpg \n",
" inflating: dataset/train/425.jpg \n",
" inflating: dataset/train/131.jpg \n",
" inflating: dataset/train/272.jpg \n",
" inflating: dataset/train/117.jpg \n",
" inflating: dataset/train/252.jpg \n",
" inflating: dataset/train/180.jpg \n",
" inflating: dataset/train/142.jpg \n",
" inflating: dataset/train/261.jpg \n",
" inflating: dataset/train/136.jpg \n",
" inflating: dataset/train/85.jpg \n",
" inflating: dataset/train/88.jpg \n",
" inflating: dataset/train/372.jpg \n",
" inflating: dataset/train/279.jpg \n",
" inflating: dataset/train/313.jpg \n",
" inflating: dataset/train/159.jpg \n",
" inflating: dataset/train/483.jpg \n",
" inflating: dataset/train/262.jpg \n",
" inflating: dataset/train/519.jpg \n",
" inflating: dataset/train/308.jpg \n",
" inflating: dataset/train/211.jpg \n",
" inflating: dataset/train/497.jpg \n",
" inflating: dataset/train/325.jpg \n",
" inflating: dataset/train/169.jpg \n",
" inflating: dataset/train/244.jpg \n",
" inflating: dataset/train/474.jpg \n",
" inflating: dataset/train/322.jpg \n",
" inflating: dataset/train/94.jpg \n",
" inflating: dataset/train/314.jpg \n",
" inflating: dataset/train/356.jpg \n",
" inflating: dataset/train/22.jpg \n",
" inflating: dataset/train/512.jpg \n",
" inflating: dataset/train/182.jpg \n",
" inflating: dataset/train/156.jpg \n",
" inflating: dataset/train/289.jpg \n",
" inflating: dataset/train/25.jpg \n",
" inflating: dataset/train/436.jpg \n",
" inflating: dataset/train/93.jpg \n",
" inflating: dataset/train/336.jpg \n",
" inflating: dataset/train/225.jpg \n",
" inflating: dataset/train/298.jpg \n",
" inflating: dataset/train/214.jpg \n",
" inflating: dataset/train/27.jpg \n",
" inflating: dataset/train/342.jpg \n",
" inflating: dataset/train/273.jpg \n",
" inflating: dataset/train/133.jpg \n",
" inflating: dataset/train/168.jpg \n",
" inflating: dataset/train/409.jpg \n",
" inflating: dataset/train/459.jpg \n",
" inflating: dataset/train/31.jpg \n",
" inflating: dataset/train/468.jpg \n",
" inflating: dataset/train/434.jpg \n",
" inflating: dataset/train/312.jpg \n",
" inflating: dataset/train/383.jpg \n",
" inflating: dataset/train/476.jpg \n",
" inflating: dataset/train/334.jpg \n",
" inflating: dataset/train/151.jpg \n",
" inflating: dataset/train/166.jpg \n",
" inflating: dataset/train/227.jpg \n",
" inflating: dataset/train/471.jpg \n",
" inflating: dataset/train/267.jpg \n",
" inflating: dataset/train/100.jpg \n",
" inflating: dataset/train/446.jpg \n",
" inflating: dataset/train/208.jpg \n",
" inflating: dataset/train/293.jpg \n",
" inflating: dataset/train/297.jpg \n",
" inflating: dataset/train/412.jpg \n",
" inflating: dataset/train/58.jpg \n",
" inflating: dataset/train/288.jpg \n",
" inflating: dataset/train/105.jpg \n",
" inflating: dataset/train/477.jpg \n",
" inflating: dataset/train/502.jpg \n",
" inflating: dataset/train/418.jpg \n",
" inflating: dataset/train/233.jpg \n",
" inflating: dataset/train/134.jpg \n",
" inflating: dataset/train/197.jpg \n",
" inflating: dataset/train/245.jpg \n",
" inflating: dataset/train/176.jpg \n",
" inflating: dataset/train/203.jpg \n",
" inflating: dataset/train/247.jpg \n",
" inflating: dataset/train/493.jpg \n",
" inflating: dataset/train/68.jpg \n",
" inflating: dataset/train/442.jpg \n",
" inflating: dataset/train/115.jpg \n",
" inflating: dataset/train/357.jpg \n",
" inflating: dataset/train/389.jpg \n",
" inflating: dataset/train/218.jpg \n",
" inflating: dataset/train/263.jpg \n",
" inflating: dataset/train/184.jpg \n",
" inflating: dataset/train/315.jpg \n",
" inflating: dataset/train/295.jpg \n",
" inflating: dataset/train/42.jpg \n",
" inflating: dataset/train/427.jpg \n",
" inflating: dataset/train/287.jpg \n",
" inflating: dataset/train/430.jpg \n",
" inflating: dataset/train/359.jpg \n",
" inflating: dataset/train/249.jpg \n",
" inflating: dataset/train/54.jpg \n",
" inflating: dataset/train/305.jpg \n",
" inflating: dataset/train/319.jpg \n",
" inflating: dataset/train/496.jpg \n",
" inflating: dataset/train/60.jpg \n",
" inflating: dataset/train/37.jpg \n",
" inflating: dataset/train/192.jpg \n",
" inflating: dataset/train/304.jpg \n",
" inflating: dataset/train/258.jpg \n",
" inflating: dataset/train/364.jpg \n",
" inflating: dataset/train/49.jpg \n",
" inflating: dataset/train/144.jpg \n",
" inflating: dataset/train/7.jpg \n",
" inflating: dataset/train/98.jpg \n",
" inflating: dataset/train/223.jpg \n",
" inflating: dataset/train/320.jpg \n",
" inflating: dataset/train/70.jpg \n",
" inflating: dataset/train/426.jpg \n",
" inflating: dataset/train/84.jpg \n",
" inflating: dataset/train/44.jpg \n",
" inflating: dataset/train/177.jpg \n",
" inflating: dataset/train/50.jpg \n",
" inflating: dataset/train/417.jpg \n",
" inflating: dataset/train/191.jpg \n",
" inflating: dataset/train/221.jpg \n",
" inflating: dataset/train/500.jpg \n",
" inflating: dataset/train/428.jpg \n",
" inflating: dataset/train/137.jpg \n",
" inflating: dataset/train/195.jpg \n",
" inflating: dataset/train/413.jpg \n",
" inflating: dataset/train/228.jpg \n",
" inflating: dataset/train/478.jpg \n",
" inflating: dataset/train/123.jpg \n",
" inflating: dataset/train/369.jpg \n",
" inflating: dataset/train/328.jpg \n",
" inflating: dataset/train/1.jpg \n",
" inflating: dataset/train/126.jpg \n",
" inflating: dataset/train/503.jpg \n",
" inflating: dataset/train/352.jpg \n",
" inflating: dataset/train/120.jpg \n",
" inflating: dataset/train/452.jpg \n",
" inflating: dataset/train/339.jpg \n",
" inflating: dataset/train/266.jpg \n",
" inflating: dataset/train/201.jpg \n",
" inflating: dataset/train/229.jpg \n",
" inflating: dataset/train/402.jpg \n",
" inflating: dataset/train/381.jpg \n",
" inflating: dataset/train/449.jpg \n",
" inflating: dataset/train/107.jpg \n",
" inflating: dataset/train/444.jpg \n",
" inflating: dataset/train/327.jpg \n",
" inflating: dataset/train/8.jpg \n",
" inflating: dataset/train/451.jpg \n",
" inflating: dataset/train/77.jpg \n",
" inflating: dataset/train/148.jpg \n",
" inflating: dataset/train/399.jpg \n",
" inflating: dataset/train/130.jpg \n",
" inflating: dataset/train/362.jpg \n",
" inflating: dataset/train/310.jpg \n",
" inflating: dataset/train/79.jpg \n",
" inflating: dataset/train/127.jpg \n",
" inflating: dataset/train/237.jpg \n",
" inflating: dataset/train/257.jpg \n",
" inflating: dataset/train/160.jpg \n",
" inflating: dataset/train/52.jpg \n",
" inflating: dataset/train/190.jpg \n",
" inflating: dataset/train/382.jpg \n",
" inflating: dataset/train/143.jpg \n",
" inflating: dataset/train/199.jpg \n",
" inflating: dataset/train/282.jpg \n",
" inflating: dataset/train/207.jpg \n",
" inflating: dataset/train/403.jpg \n",
" inflating: dataset/train/337.jpg \n",
" inflating: dataset/train/355.jpg \n",
" inflating: dataset/train/251.jpg \n",
" inflating: dataset/train/188.jpg \n",
" inflating: dataset/train/338.jpg \n",
" inflating: dataset/train/341.jpg \n",
" inflating: dataset/train/433.jpg \n",
" inflating: dataset/train/69.jpg \n",
" inflating: dataset/train/236.jpg \n",
" inflating: dataset/train/106.jpg \n",
" inflating: dataset/train/99.jpg \n",
" inflating: dataset/train/179.jpg \n",
" inflating: dataset/train/204.jpg \n",
" inflating: dataset/train/300.jpg \n",
" inflating: dataset/train/432.jpg \n",
" inflating: dataset/train/490.jpg \n",
" inflating: dataset/train/368.jpg \n",
" inflating: dataset/train/153.jpg \n",
" inflating: dataset/train/174.jpg \n",
" inflating: dataset/train/51.jpg \n",
" inflating: dataset/train/62.jpg \n",
" inflating: dataset/train/420.jpg \n",
" inflating: dataset/train/75.jpg \n",
" inflating: dataset/train/230.jpg \n",
" inflating: dataset/train/307.jpg \n",
" inflating: dataset/train/212.jpg \n",
" inflating: dataset/train/209.jpg \n",
" inflating: dataset/train/415.jpg \n",
" inflating: dataset/train/491.jpg \n",
" inflating: dataset/train/453.jpg \n",
" inflating: dataset/train/351.jpg \n",
" inflating: dataset/train/111.jpg \n",
" inflating: dataset/train/391.jpg \n",
" inflating: dataset/train/515.jpg \n",
" inflating: dataset/train/65.jpg \n",
" inflating: dataset/train/240.jpg \n",
" inflating: dataset/train/91.jpg \n",
" inflating: dataset/train/222.jpg \n",
" inflating: dataset/train/53.jpg \n",
" inflating: dataset/train/158.jpg \n",
" inflating: dataset/train/475.jpg \n",
" inflating: dataset/train/121.jpg \n",
" inflating: dataset/train/396.jpg \n",
" inflating: dataset/train/63.jpg \n",
" inflating: dataset/train/268.jpg \n",
" inflating: dataset/train/78.jpg \n",
" inflating: dataset/train/167.jpg \n",
" inflating: dataset/train/250.jpg \n",
" inflating: dataset/train/454.jpg \n",
" inflating: dataset/train/186.jpg \n",
" inflating: dataset/train/499.jpg \n",
" inflating: dataset/train/101.jpg \n",
" inflating: dataset/train/435.jpg \n",
" inflating: dataset/train/469.jpg \n",
" inflating: dataset/train/139.jpg \n",
" inflating: dataset/train/83.jpg \n",
" inflating: dataset/train/43.jpg \n",
" inflating: dataset/train/281.jpg \n",
" inflating: dataset/train/36.jpg \n",
" inflating: dataset/train/419.jpg \n",
" inflating: dataset/train/66.jpg \n",
" inflating: dataset/train/260.jpg \n",
" inflating: dataset/train/466.jpg \n",
" inflating: dataset/train/140.jpg \n",
" inflating: dataset/train/5.jpg \n",
" inflating: dataset/train/10.jpg \n",
" inflating: dataset/train/235.jpg \n",
" inflating: dataset/train/511.jpg \n",
" inflating: dataset/train/72.jpg \n",
" inflating: dataset/train/253.jpg \n",
" inflating: dataset/train/404.jpg \n",
" inflating: dataset/train/89.jpg \n",
" inflating: dataset/train/311.jpg \n",
" inflating: dataset/test.csv \n",
" inflating: dataset/train.csv \n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "1-KvvWgkkGJO",
"colab_type": "text"
},
"source": [
"In this file we are using Transfer Learning concept to classify Indian dance form. Transfer Learning used when we have very less training data. In image processing, training with less data does not give good results. So we are using Transfer Learning to get weights. This notebook use *tensorflow* *VGG16* and the purpose of this notbook to show how to use Transfer Learning for image classification "
]
},
{
"cell_type": "code",
"metadata": {
"_uuid": "4242ebab-da29-4ac3-ba12-1c2217255c29",
"_cell_guid": "8705da2e-9d7b-4eb0-891e-9eab3aa1106b",
"trusted": true,
"id": "UjmD6p24kGJU",
"colab_type": "code",
"colab": {}
},
"source": [
"# import required libraries \n",
"import os\n",
"import pickle\n",
"import pandas as pd\n",
"import numpy as np\n",
"import cv2\n",
"import matplotlib.pyplot as plt\n",
"import tensorflow as tf\n",
"from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
"from tensorflow.keras.applications.vgg16 import VGG16\n",
"from tensorflow.keras.applications.vgg16 import preprocess_input\n",
"from tensorflow.keras.applications import *\n",
"from tensorflow.keras.layers import Flatten,Dense,Dropout,BatchNormalization\n",
"from tensorflow.keras.models import Model,Sequential\n",
"from tensorflow.keras.utils import to_categorical\n",
"from tensorflow.keras.layers import Conv2D, MaxPooling2D, BatchNormalization\n",
"\n",
"from tensorflow.keras.callbacks import ReduceLROnPlateau\n",
"\n",
"from tensorflow.keras.optimizers import Adam,SGD,Adagrad,Adadelta,RMSprop\n",
"\n",
"from sklearn.preprocessing import LabelEncoder"
],
"execution_count": 5,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"_uuid": "9471514d-4526-43d1-b568-d451ce141df6",
"_cell_guid": "f6931f33-f202-4a40-8419-c94dea6ba2a2",
"trusted": true,
"id": "uvN7ZEPVkGJr",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 374
},
"outputId": "b688b822-af63-4e0c-853a-b20c6f2d810f"
},
"source": [
"# Load train and test csv file for image class\n",
"train = pd.read_csv(ROOT+'train.csv')\n",
"test = pd.read_csv(ROOT+'test.csv')\n",
"\n",
"print(train.head())\n",
"print(test.head())\n",
"print(train['target'].value_counts())"
],
"execution_count": 6,
"outputs": [
{
"output_type": "stream",
"text": [
" Image target\n",
"0 96.jpg manipuri\n",
"1 163.jpg bharatanatyam\n",
"2 450.jpg odissi\n",
"3 219.jpg kathakali\n",
"4 455.jpg odissi\n",
" Image\n",
"0 508.jpg\n",
"1 246.jpg\n",
"2 473.jpg\n",
"3 485.jpg\n",
"4 128.jpg\n",
"mohiniyattam 50\n",
"odissi 49\n",
"bharatanatyam 47\n",
"kathakali 47\n",
"kuchipudi 46\n",
"sattriya 45\n",
"kathak 44\n",
"manipuri 36\n",
"Name: target, dtype: int64\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"trusted": true,
"id": "VYYguqNOkGJ9",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 204
},
"outputId": "cb01acd6-ac3c-450a-a3ef-b5ecfecf5c2a"
},
"source": [
"train.head()"
],
"execution_count": 7,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Image</th>\n",
" <th>target</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>96.jpg</td>\n",
" <td>manipuri</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>163.jpg</td>\n",
" <td>bharatanatyam</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>450.jpg</td>\n",
" <td>odissi</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>219.jpg</td>\n",
" <td>kathakali</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>455.jpg</td>\n",
" <td>odissi</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Image target\n",
"0 96.jpg manipuri\n",
"1 163.jpg bharatanatyam\n",
"2 450.jpg odissi\n",
"3 219.jpg kathakali\n",
"4 455.jpg odissi"
]
},
"metadata": {
"tags": []
},
"execution_count": 7
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "ZSa3VDPYkGKQ",
"colab_type": "text"
},
"source": [
"Basic Histrogram plot to check number of training data for each dance form. "
]
},
{
"cell_type": "code",
"metadata": {
"_uuid": "f84443f3-c7fa-437b-af7d-3107bd6a9b96",
"_cell_guid": "f32bfd68-2e4a-4474-8577-b293c885b01a",
"trusted": true,
"id": "gO4tZatikGKT",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 349
},
"outputId": "542a20af-1027-44ca-c6a4-4e99a0283013"
},
"source": [
"#Histogram chart for target\n",
"train['target'].value_counts().plot(kind='bar')"
],
"execution_count": 8,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"<matplotlib.axes._subplots.AxesSubplot at 0x7f5b65dc4dd8>"
]
},
"metadata": {
"tags": []
},
"execution_count": 8
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
}
]
},
{
"cell_type": "code",
"metadata": {
"_uuid": "d1970312-392f-4956-98e4-5745098b257b",
"_cell_guid": "2c290850-1f68-407f-b2a8-5bc6813a761a",
"trusted": true,
"id": "lWLhsmK0kGKg",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 51
},
"outputId": "7c8017c0-e08a-4130-899f-f5a51cdf3f68"
},
"source": [
"\n",
"train_dir = os.path.join(ROOT+ 'train/')\n",
"test_dir = os.path.join(ROOT+'test/')\n",
"\n",
"train_fnames = os.listdir(train_dir)\n",
"test_fnames = os.listdir(test_dir)\n",
"\n",
"print(train_fnames[:9])\n",
"print(test_fnames[:9])"
],
"execution_count": 9,
"outputs": [
{
"output_type": "stream",
"text": [
"['494.jpg', '298.jpg', '266.jpg', '197.jpg', '24.jpg', '294.jpg', '229.jpg', '335.jpg', '250.jpg']\n",
"['332.jpg', '38.jpg', '375.jpg', '346.jpg', '23.jpg', '57.jpg', '196.jpg', '80.jpg', '114.jpg']\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"_uuid": "184d83c9-44b9-4ce5-a043-3e00293fcb0f",
"_cell_guid": "485ab552-ea07-4b29-8c7d-6fc6cccae2e4",
"trusted": true,
"id": "NpUWH_XdkGKp",
"colab_type": "code",
"colab": {}
},
"source": [
"# Images migh be in different size. In this section I assigning all image at same size of 224*224\n",
"img_width = 224\n",
"img_height = 224"
],
"execution_count": 11,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "6lqzemLnkGKw",
"colab_type": "text"
},
"source": [
"Below two section used for data preprocessing. We are reading image data using OpenCV and converting into numeric formate."
]
},
{
"cell_type": "code",
"metadata": {
"_uuid": "9eac3e8e-36fb-40a8-beaa-ab8666284798",
"_cell_guid": "6b931fb3-4df1-445f-8d28-193626c426a7",
"trusted": true,
"id": "8RSFyaLQkGKx",
"colab_type": "code",
"colab": {}
},
"source": [
"# this function reads image from the disk,train file for image and class maping and returning output in numpy array formate\n",
"# for input and target data\n",
"def train_data_preparation(list_of_images, train, train_dir):\n",
" \"\"\"\n",
" Returns two arrays: \n",
" x is an array of resized images\n",
" y is an array of labels\n",
" \"\"\"\n",
" x = [] # images as arrays\n",
" y = [] # labels\n",
" for image in list_of_images:\n",
" x.append((cv2.resize(cv2.imread(train_dir+image), (img_width,img_height), interpolation=cv2.INTER_CUBIC)).astype('float32'))\n",
"\n",
" if image in list(train['Image']):\n",
" y.append(train.loc[train['Image'] == image, 'target'].values[0])\n",
" \n",
" \n",
" return x, y"
],
"execution_count": 12,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"_uuid": "6d8da7a1-acce-4b9a-8743-cb861feab693",
"_cell_guid": "966a3be7-65e3-4589-bf39-aa7c859afbfb",
"trusted": true,
"id": "Lk5876abkGK6",
"colab_type": "code",
"colab": {}
},
"source": [
"def test_data_prepare(list_of_images, test_dir):\n",
" \"\"\"\n",
" Returns two arrays: \n",
" x is an array of resized images\n",
" y is an array of labels\n",
" \"\"\"\n",
" x = [] # images as arrays\n",
" \n",
" for image in list_of_images:\n",
" x.append((cv2.resize(cv2.imread(test_dir+image), (img_width,img_height), interpolation=cv2.INTER_CUBIC)).astype('float32')) \n",
" \n",
" return x"
],
"execution_count": 13,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"_uuid": "8e86a433-00d7-416b-82e3-83ee594c234b",
"_cell_guid": "17ed4be2-28b2-46eb-ac17-2d265bdafbe0",
"trusted": true,
"id": "d5hZnWdZkGK_",
"colab_type": "code",
"colab": {}
},
"source": [
"training_data, training_labels = train_data_preparation(train_fnames, train, train_dir)"
],
"execution_count": 14,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"trusted": true,
"id": "p41Gb8jJkGLE",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 357
},
"outputId": "fc9deca2-57f1-45bc-d2bd-81c0768fd995"
},
"source": [
"training_labels[:20]"
],
"execution_count": 15,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"['odissi',\n",
" 'mohiniyattam',\n",
" 'mohiniyattam',\n",
" 'kathakali',\n",
" 'kathak',\n",
" 'mohiniyattam',\n",
" 'kathakali',\n",
" 'sattriya',\n",
" 'mohiniyattam',\n",
" 'sattriya',\n",
" 'sattriya',\n",
" 'kathakali',\n",
" 'bharatanatyam',\n",
" 'manipuri',\n",
" 'kathak',\n",
" 'kathak',\n",
" 'kuchipudi',\n",
" 'manipuri',\n",
" 'kuchipudi',\n",
" 'bharatanatyam']"
]
},
"metadata": {
"tags": []
},
"execution_count": 15
}
]
},
{
"cell_type": "code",
"metadata": {
"_uuid": "1d837321-2bee-4081-8e1d-69cffa39e971",
"_cell_guid": "be7ae0f9-912f-4c27-a489-0ada13fedc1d",
"trusted": true,
"id": "Z0T5y_JdkGLK",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 850
},
"outputId": "c32c93f6-e0c3-44ca-9901-0687f1679fc2"
},
"source": [
"training_data[0]"
],
"execution_count": 16,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"array([[[13., 6., 3.],\n",
" [14., 7., 4.],\n",
" [13., 6., 3.],\n",
" ...,\n",
" [20., 1., 22.],\n",
" [21., 1., 24.],\n",
" [21., 1., 24.]],\n",
"\n",
" [[13., 6., 3.],\n",
" [14., 7., 4.],\n",
" [13., 6., 3.],\n",
" ...,\n",
" [20., 1., 22.],\n",
" [21., 1., 24.],\n",
" [21., 1., 24.]],\n",
"\n",
" [[13., 6., 3.],\n",
" [13., 6., 3.],\n",
" [13., 6., 3.],\n",
" ...,\n",
" [20., 1., 22.],\n",
" [21., 1., 24.],\n",
" [21., 1., 24.]],\n",
"\n",
" ...,\n",
"\n",
" [[19., 13., 19.],\n",
" [18., 21., 35.],\n",
" [20., 29., 44.],\n",
" ...,\n",
" [15., 10., 7.],\n",
" [13., 8., 5.],\n",
" [12., 7., 5.]],\n",
"\n",
" [[21., 15., 20.],\n",
" [19., 22., 36.],\n",
" [20., 29., 43.],\n",
" ...,\n",
" [15., 10., 7.],\n",
" [13., 8., 5.],\n",
" [13., 8., 5.]],\n",
"\n",
" [[22., 16., 21.],\n",
" [17., 22., 37.],\n",
" [20., 29., 43.],\n",
" ...,\n",
" [13., 8., 5.],\n",
" [16., 11., 8.],\n",
" [15., 10., 7.]]], dtype=float32)"
]
},
"metadata": {
"tags": []
},
"execution_count": 16
}
]
},
{
"cell_type": "code",
"metadata": {
"_uuid": "b8b3520d-52d6-48e5-9740-1abf86faf1a3",
"_cell_guid": "b5f813a6-b9d9-4d8b-bb84-9c7dc25be251",
"trusted": true,
"id": "a4XKQh4pkGLQ",
"colab_type": "code",
"colab": {}
},
"source": [
"\n",
"def show_batch(image_batch, label_batch):\n",
" plt.figure(figsize=(12,12))\n",
" for n in range(25):\n",
" ax = plt.subplot(5,5,n+1)\n",
" plt.imshow(image_batch[n])\n",
" plt.title(label_batch[n].title())\n",
" plt.axis('off')"
],
"execution_count": 63,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "GbffSu2skGLW",
"colab_type": "text"
},
"source": [
"Just showing loaded data for first 25 image"
]
},
{
"cell_type": "code",
"metadata": {
"_uuid": "c4790fe6-4c7c-4dcf-831f-036bc3be2d9e",
"_cell_guid": "abf254fb-1c87-453a-bbd9-a9147b1eb87a",
"trusted": true,
"id": "iX4l9RA_kGLX",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 429
},
"outputId": "253d9c7e-f3a7-4e0a-a199-8b84997bf332"
},
"source": [
"show_batch(training_data, training_labels)"
],
"execution_count": 65,
"outputs": [
{
"output_type": "stream",
"text": [
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
],
"name": "stderr"
},
{
"output_type": "error",
"ename": "AttributeError",
"evalue": "ignored",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mAttributeError\u001b[0m Traceback (most recent call last)",
"\u001b[0;32m<ipython-input-65-abbd9efd2258>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mshow_batch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtraining_data\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtraining_labels\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
"\u001b[0;32m<ipython-input-63-8c691c60f04c>\u001b[0m in \u001b[0;36mshow_batch\u001b[0;34m(image_batch, label_batch)\u001b[0m\n\u001b[1;32m 5\u001b[0m \u001b[0max\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msubplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m5\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m5\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mn\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 6\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mimshow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mimage_batch\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mn\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 7\u001b[0;31m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtitle\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlabel_batch\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mn\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtitle\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 8\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0maxis\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'off'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;31mAttributeError\u001b[0m: 'numpy.int64' object has no attribute 'title'"
]
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 864x864 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
}
]
},
{
"cell_type": "code",
"metadata": {
"_uuid": "a66fd65f-bb62-4b09-8401-3e72f8888c25",
"_cell_guid": "696d2f31-1096-4bde-97a4-94d50be6474d",
"trusted": true,
"id": "VPSNJNVTkGLc",
"colab_type": "code",
"colab": {}
},
"source": [
"testing_data = test_data_prepare(test_fnames, test_dir)"
],
"execution_count": 19,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "YEj3eq6KkGLj",
"colab_type": "text"
},
"source": [
"Using label incoder converting target class to numeric format"
]
},
{
"cell_type": "code",
"metadata": {
"_uuid": "1281b7f1-018b-4566-a711-26099e6b7867",
"_cell_guid": "14622e31-731d-4fd8-ad46-1d6be47d1efc",
"trusted": true,
"id": "bjUHaLG7kGLl",
"colab_type": "code",
"colab": {}
},
"source": [
"le =LabelEncoder()\n",
"training_labels=le.fit_transform(training_labels)"
],
"execution_count": 20,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "IXqCj255kGLt",
"colab_type": "text"
},
"source": [
"In this section I am using ougumentation techniques to generate more data for given input"
]
},
{
"cell_type": "code",
"metadata": {
"_uuid": "8937cb05-661c-4580-983d-c7001218717d",
"_cell_guid": "a76645f7-9d2d-43f3-8bfc-c6fef4ce4135",
"trusted": true,
"id": "ykDffFmnkGLu",
"colab_type": "code",
"colab": {}
},
"source": [
"datagenerator = ImageDataGenerator(\n",
" rescale=1. / 255,\n",
" featurewise_center=False, \n",
" samplewise_center=False, \n",
" featurewise_std_normalization=False, \n",
" samplewise_std_normalization=False, \n",
" rotation_range=10, \n",
" zoom_range = 0.10, \n",
" width_shift_range=0.10, \n",
" height_shift_range=0.10, \n",
" horizontal_flip=True, \n",
" vertical_flip=False) \n",
"\n",
"\n",
"test_datagenerator=ImageDataGenerator(\n",
" rescale=1. / 255\n",
")\n",
"datagenerator.fit(training_data)\n",
"test_datagenerator.fit(training_data)\n",
"training_data=np.array(training_data)\n",
"testing_data=np.array(testing_data)"
],
"execution_count": 21,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"trusted": true,
"id": "EAELTupfkGLy",
"colab_type": "code",
"colab": {}
},
"source": [
"# subsetting validation data\n",
"validation_data = training_data[300:]\n",
"validation_label =training_labels[300:]"
],
"execution_count": 22,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"trusted": true,
"id": "mMSgnEOqkGL7",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 850
},
"outputId": "e07d6923-bdcd-4328-dd7a-0b1bbe639747"
},
"source": [
"training_data = training_data[:300]\n",
"training_labels = training_labels[:300]\n",
"\n",
"training_data[0]"
],
"execution_count": 23,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"array([[[13., 6., 3.],\n",
" [14., 7., 4.],\n",
" [13., 6., 3.],\n",
" ...,\n",
" [20., 1., 22.],\n",
" [21., 1., 24.],\n",
" [21., 1., 24.]],\n",
"\n",
" [[13., 6., 3.],\n",
" [14., 7., 4.],\n",
" [13., 6., 3.],\n",
" ...,\n",
" [20., 1., 22.],\n",
" [21., 1., 24.],\n",
" [21., 1., 24.]],\n",
"\n",
" [[13., 6., 3.],\n",
" [13., 6., 3.],\n",
" [13., 6., 3.],\n",
" ...,\n",
" [20., 1., 22.],\n",
" [21., 1., 24.],\n",
" [21., 1., 24.]],\n",
"\n",
" ...,\n",
"\n",
" [[19., 13., 19.],\n",
" [18., 21., 35.],\n",
" [20., 29., 44.],\n",
" ...,\n",
" [15., 10., 7.],\n",
" [13., 8., 5.],\n",
" [12., 7., 5.]],\n",
"\n",
" [[21., 15., 20.],\n",
" [19., 22., 36.],\n",
" [20., 29., 43.],\n",
" ...,\n",
" [15., 10., 7.],\n",
" [13., 8., 5.],\n",
" [13., 8., 5.]],\n",
"\n",
" [[22., 16., 21.],\n",
" [17., 22., 37.],\n",
" [20., 29., 43.],\n",
" ...,\n",
" [13., 8., 5.],\n",
" [16., 11., 8.],\n",
" [15., 10., 7.]]], dtype=float32)"
]
},
"metadata": {
"tags": []
},
"execution_count": 23
}
]
},
{
"cell_type": "code",
"metadata": {
"_uuid": "fb56689b-87ea-4735-8086-5c7843774a14",
"_cell_guid": "b826e3e8-4bf5-4a37-96b0-30d074fa8a78",
"trusted": true,
"id": "E_mgYbUVkGMB",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 85
},
"outputId": "78dc9ac6-96d4-4cf8-cdca-37ce2f156cff"
},
"source": [
"print(training_data.shape)\n",
"print(training_labels.shape)\n",
"print(validation_data.shape)\n",
"print(validation_label.shape)"
],
"execution_count": 24,
"outputs": [
{
"output_type": "stream",
"text": [
"(300, 224, 224, 3)\n",
"(300,)\n",
"(64, 224, 224, 3)\n",
"(64,)\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "ef5dI5yFkGMG",
"colab_type": "text"
},
"source": [
"In below code we are loading *VGG16* weights for image classifier using transfer learning"
]
},
{
"cell_type": "code",
"metadata": {
"_uuid": "ae045ec6-a8fe-4f88-9205-3d2e96232ecc",
"_cell_guid": "987fff4c-6a3a-4181-b6e0-3abd2f870b84",
"trusted": true,
"id": "znkWy9JmkGMH",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 887
},
"outputId": "c2ceaf00-0dca-4560-fbc4-dcd3542fcc5b"
},
"source": [
"# traning using transfer learning\n",
"\n",
"vggmodel =VGG16(weights='imagenet', include_top=False, input_shape = (224, 224, 3),pooling='avg')\n",
"\n",
" # Print the model summary\n",
"vggmodel.summary()"
],
"execution_count": 25,
"outputs": [
{
"output_type": "stream",
"text": [
"Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/vgg16/vgg16_weights_tf_dim_ordering_tf_kernels_notop.h5\n",
"58892288/58889256 [==============================] - 1s 0us/step\n",
"Model: \"vgg16\"\n",
"_________________________________________________________________\n",
"Layer (type) Output Shape Param # \n",
"=================================================================\n",
"input_1 (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",
"global_average_pooling2d (Gl (None, 512) 0 \n",
"=================================================================\n",
"Total params: 14,714,688\n",
"Trainable params: 14,714,688\n",
"Non-trainable params: 0\n",
"_________________________________________________________________\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "NqqyP56fkGMM",
"colab_type": "text"
},
"source": [
"Using already trained model for our task and bulding 2 fully connected layer with *softmax* activation function"
]
},
{
"cell_type": "code",
"metadata": {
"_uuid": "cc563137-b352-42e2-b779-9af692ab6d54",
"_cell_guid": "7afc81b7-32e9-4c40-ad47-ac028f929b73",
"trusted": true,
"id": "kEveCsURkGMN",
"colab_type": "code",
"colab": {}
},
"source": [
"vggmodel.trainable = False\n",
"model = Sequential([\n",
" vggmodel, \n",
" Dense(1024, activation='relu'),\n",
" Dropout(0.4),\n",
" Dense(256, activation='relu'),\n",
" Dense(8, activation='softmax'),\n",
"])"
],
"execution_count": 26,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"_uuid": "86b07c2b-5c5d-493c-87b1-e455196605de",
"_cell_guid": "b8be0c2a-3417-4440-87ad-918c07db001e",
"trusted": true,
"id": "HROwFaNKkGMR",
"colab_type": "code",
"colab": {}
},
"source": [
"\n",
"reduce_learning_rate = ReduceLROnPlateau(monitor='loss',\n",
" factor=0.1,\n",
" patience=2,\n",
" cooldown=2,\n",
" min_lr=0.00001,\n",
" verbose=1)\n",
"\n",
"callbacks = [reduce_learning_rate]"
],
"execution_count": 27,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "Zexvx9mPkGMX",
"colab_type": "text"
},
"source": [
"In the below code we are compiling and traing our image data"
]
},
{
"cell_type": "code",
"metadata": {
"_uuid": "a0d7c6e3-f815-4653-b07e-f10e56ecc068",
"_cell_guid": "fdd24057-3683-4ebf-9f1a-6bb0c9cdff18",
"trusted": true,
"id": "2zrqJfB_kGMZ",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 1000
},
"outputId": "cee9a3d4-9f24-455f-960b-d6bd6a352aa7"
},
"source": [
"model.compile( optimizer='adam',loss='categorical_crossentropy',metrics=['accuracy'])\n",
"history =model.fit_generator(\n",
" datagenerator.flow(training_data, to_categorical(training_labels,8), batch_size=32),\n",
" validation_data=datagenerator.flow(validation_data, to_categorical(validation_label,8), batch_size=32),\n",
" verbose=2,\n",
" epochs=30,\n",
" callbacks=callbacks\n",
")"
],
"execution_count": 28,
"outputs": [
{
"output_type": "stream",
"text": [
"WARNING:tensorflow:From <ipython-input-28-53a71270ff0c>:7: Model.fit_generator (from tensorflow.python.keras.engine.training) is deprecated and will be removed in a future version.\n",
"Instructions for updating:\n",
"Please use Model.fit, which supports generators.\n",
"Epoch 1/30\n",
"10/10 - 7s - loss: 2.1711 - accuracy: 0.1333 - val_loss: 2.0442 - val_accuracy: 0.2188 - lr: 0.0010\n",
"Epoch 2/30\n",
"10/10 - 5s - loss: 1.9933 - accuracy: 0.2467 - val_loss: 1.8793 - val_accuracy: 0.3906 - lr: 0.0010\n",
"Epoch 3/30\n",
"10/10 - 5s - loss: 1.8018 - accuracy: 0.3800 - val_loss: 1.6916 - val_accuracy: 0.4375 - lr: 0.0010\n",
"Epoch 4/30\n",
"10/10 - 5s - loss: 1.6582 - accuracy: 0.3967 - val_loss: 1.5616 - val_accuracy: 0.4375 - lr: 0.0010\n",
"Epoch 5/30\n",
"10/10 - 5s - loss: 1.5129 - accuracy: 0.4733 - val_loss: 1.4451 - val_accuracy: 0.4531 - lr: 0.0010\n",
"Epoch 6/30\n",
"10/10 - 5s - loss: 1.4268 - accuracy: 0.4600 - val_loss: 1.3445 - val_accuracy: 0.5312 - lr: 0.0010\n",
"Epoch 7/30\n",
"10/10 - 5s - loss: 1.2517 - accuracy: 0.5867 - val_loss: 1.2431 - val_accuracy: 0.5312 - lr: 0.0010\n",
"Epoch 8/30\n",
"10/10 - 5s - loss: 1.1550 - accuracy: 0.6100 - val_loss: 1.1261 - val_accuracy: 0.6406 - lr: 0.0010\n",
"Epoch 9/30\n",
"10/10 - 5s - loss: 1.0371 - accuracy: 0.6900 - val_loss: 1.2041 - val_accuracy: 0.5781 - lr: 0.0010\n",
"Epoch 10/30\n",
"10/10 - 5s - loss: 1.0084 - accuracy: 0.6600 - val_loss: 1.0855 - val_accuracy: 0.6094 - lr: 0.0010\n",
"Epoch 11/30\n",
"10/10 - 5s - loss: 0.8848 - accuracy: 0.6900 - val_loss: 1.0847 - val_accuracy: 0.6562 - lr: 0.0010\n",
"Epoch 12/30\n",
"10/10 - 5s - loss: 0.9019 - accuracy: 0.6900 - val_loss: 0.9575 - val_accuracy: 0.6875 - lr: 0.0010\n",
"Epoch 13/30\n",
"10/10 - 5s - loss: 0.7895 - accuracy: 0.7033 - val_loss: 1.2095 - val_accuracy: 0.5469 - lr: 0.0010\n",
"Epoch 14/30\n",
"10/10 - 5s - loss: 0.7651 - accuracy: 0.7533 - val_loss: 1.0083 - val_accuracy: 0.6562 - lr: 0.0010\n",
"Epoch 15/30\n",
"10/10 - 5s - loss: 0.7295 - accuracy: 0.7467 - val_loss: 0.9432 - val_accuracy: 0.7031 - lr: 0.0010\n",
"Epoch 16/30\n",
"10/10 - 5s - loss: 0.6489 - accuracy: 0.7900 - val_loss: 0.9389 - val_accuracy: 0.7188 - lr: 0.0010\n",
"Epoch 17/30\n",
"10/10 - 5s - loss: 0.6018 - accuracy: 0.8100 - val_loss: 0.8987 - val_accuracy: 0.7188 - lr: 0.0010\n",
"Epoch 18/30\n",
"10/10 - 5s - loss: 0.6923 - accuracy: 0.7433 - val_loss: 1.0198 - val_accuracy: 0.6562 - lr: 0.0010\n",
"Epoch 19/30\n",
"\n",
"Epoch 00019: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.\n",
"10/10 - 5s - loss: 0.6062 - accuracy: 0.7867 - val_loss: 1.0228 - val_accuracy: 0.6406 - lr: 0.0010\n",
"Epoch 20/30\n",
"10/10 - 5s - loss: 0.5061 - accuracy: 0.8167 - val_loss: 0.9986 - val_accuracy: 0.6719 - lr: 1.0000e-04\n",
"Epoch 21/30\n",
"10/10 - 5s - loss: 0.4134 - accuracy: 0.8900 - val_loss: 0.8821 - val_accuracy: 0.7188 - lr: 1.0000e-04\n",
"Epoch 22/30\n",
"10/10 - 5s - loss: 0.4116 - accuracy: 0.8933 - val_loss: 0.8590 - val_accuracy: 0.7188 - lr: 1.0000e-04\n",
"Epoch 23/30\n",
"10/10 - 5s - loss: 0.4079 - accuracy: 0.9000 - val_loss: 0.9847 - val_accuracy: 0.6406 - lr: 1.0000e-04\n",
"Epoch 24/30\n",
"10/10 - 5s - loss: 0.4036 - accuracy: 0.8833 - val_loss: 0.9113 - val_accuracy: 0.7031 - lr: 1.0000e-04\n",
"Epoch 25/30\n",
"10/10 - 5s - loss: 0.4155 - accuracy: 0.8600 - val_loss: 0.9453 - val_accuracy: 0.7031 - lr: 1.0000e-04\n",
"Epoch 26/30\n",
"\n",
"Epoch 00026: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.\n",
"10/10 - 5s - loss: 0.4085 - accuracy: 0.8733 - val_loss: 0.8971 - val_accuracy: 0.7031 - lr: 1.0000e-04\n",
"Epoch 27/30\n",
"10/10 - 5s - loss: 0.3838 - accuracy: 0.9033 - val_loss: 0.9657 - val_accuracy: 0.6719 - lr: 1.0000e-05\n",
"Epoch 28/30\n",
"10/10 - 5s - loss: 0.3954 - accuracy: 0.8867 - val_loss: 0.9213 - val_accuracy: 0.6875 - lr: 1.0000e-05\n",
"Epoch 29/30\n",
"\n",
"Epoch 00029: ReduceLROnPlateau reducing learning rate to 1e-05.\n",
"10/10 - 5s - loss: 0.3997 - accuracy: 0.8833 - val_loss: 0.9971 - val_accuracy: 0.6562 - lr: 1.0000e-05\n",
"Epoch 30/30\n",
"10/10 - 5s - loss: 0.4069 - accuracy: 0.8933 - val_loss: 0.8941 - val_accuracy: 0.6875 - lr: 1.0000e-05\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"trusted": true,
"id": "TcyFK9PpkGMf",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 562
},
"outputId": "86824a4e-42e5-4651-c49c-7b9df6feeb98"
},
"source": [
"import matplotlib.image as mpimg\n",
"import matplotlib.pyplot as plt\n",
"\n",
"acc = history.history[ 'accuracy' ]\n",
"val_acc = history.history[ 'val_accuracy' ]\n",
"loss = history.history[ 'loss' ]\n",
"val_loss = history.history['val_loss' ]\n",
"\n",
"epochs = range(len(acc)) # Get number of epochs\n",
"\n",
"#------------------------------------------------\n",
"# Plot training and validation accuracy per epoch\n",
"#------------------------------------------------\n",
"plt.plot ( epochs, acc )\n",
"plt.plot ( epochs, val_acc )\n",
"plt.title ('Training and validation accuracy')\n",
"plt.figure()\n",
"\n",
"#------------------------------------------------\n",
"# Plot training and validation loss per epoch\n",
"#------------------------------------------------\n",
"plt.plot ( epochs, loss )\n",
"plt.plot ( epochs, val_loss )\n",
"plt.title ('Training and validation loss' )"
],
"execution_count": 29,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"Text(0.5, 1.0, 'Training and validation loss')"
]
},
"metadata": {
"tags": []
},
"execution_count": 29
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
}
]
},
{
"cell_type": "code",
"metadata": {
"_uuid": "127ba8a4-8987-4664-a6dc-0da7528f5dc6",
"_cell_guid": "346838f7-8f72-4a42-8eb4-a21b777d3fc6",
"trusted": true,
"id": "65HQHA3OkGMj",
"colab_type": "code",
"colab": {}
},
"source": [
"labels = model.predict(testing_data)\n",
"label = [np.argmax(i) for i in labels]\n"
],
"execution_count": 30,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"trusted": true,
"id": "9HVRHFaTkGMs",
"colab_type": "code",
"colab": {}
},
"source": [
"target=le.inverse_transform(label)"
],
"execution_count": 31,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"trusted": true,
"id": "cPjZMSULkGMy",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 119
},
"outputId": "0684f45a-d1b9-4722-fc2d-7bea4bead92d"
},
"source": [
"target[:25]"
],
"execution_count": 32,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"array(['kathakali', 'bharatanatyam', 'kuchipudi', 'kathakali',\n",
" 'kathakali', 'kathakali', 'kathakali', 'kathakali', 'kathakali',\n",
" 'kathakali', 'kathakali', 'kathakali', 'bharatanatyam',\n",
" 'kathakali', 'kathakali', 'bharatanatyam', 'bharatanatyam',\n",
" 'kathakali', 'kathakali', 'kathakali', 'mohiniyattam', 'kathakali',\n",
" 'bharatanatyam', 'sattriya', 'kathakali'], dtype='<U13')"
]
},
"metadata": {
"tags": []
},
"execution_count": 32
}
]
},
{
"cell_type": "code",
"metadata": {
"_uuid": "990412ff-d38c-4a9f-ad54-8a0c2f209d4e",
"_cell_guid": "cafb6a9f-b517-4fa2-b239-4d46fed5f3e8",
"trusted": true,
"id": "RnpD4yEqkGM2",
"colab_type": "code",
"colab": {}
},
"source": [
"\n",
"submission = pd.DataFrame({ 'Image': test.Image, 'target': target })\n",
"\n",
"with open('output.csv','w+') as f:\n",
" submission.to_csv(f)"
],
"execution_count": 33,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"_uuid": "ab1ad8f4-be2e-4a11-8318-ac5628b885ec",
"_cell_guid": "b5adf29c-75ef-46b9-8eb0-b2462b2cdc89",
"trusted": true,
"id": "q_mY193okGM8",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 221
},
"outputId": "1b63c904-38b3-4f8d-bce5-57d89d5ad6a4"
},
"source": [
"submission['Image']"
],
"execution_count": 34,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"0 508.jpg\n",
"1 246.jpg\n",
"2 473.jpg\n",
"3 485.jpg\n",
"4 128.jpg\n",
" ... \n",
"151 366.jpg\n",
"152 226.jpg\n",
"153 35.jpg\n",
"154 458.jpg\n",
"155 358.jpg\n",
"Name: Image, Length: 156, dtype: object"
]
},
"metadata": {
"tags": []
},
"execution_count": 34
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "X4iUZKUT0Pr1",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 221
},
"outputId": "b1caaf4a-ea07-4b8d-d8cb-110bf5f99184"
},
"source": [
"submission['target']"
],
"execution_count": null,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"0 kathakali\n",
"1 bharatanatyam\n",
"2 kuchipudi\n",
"3 kathakali\n",
"4 kathakali\n",
" ... \n",
"151 kathakali\n",
"152 bharatanatyam\n",
"153 kathakali\n",
"154 kathakali\n",
"155 kuchipudi\n",
"Name: target, Length: 156, dtype: object"
]
},
"metadata": {
"tags": []
},
"execution_count": 45
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "LjPs9u0h0ULq",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 357
},
"outputId": "47dcd75b-f41c-486f-823f-7214b2df161f"
},
"source": [
"plt.imshow(testing_data[2]/255)\n",
"model.save(ROOT)\n"
],
"execution_count": 35,
"outputs": [
{
"output_type": "stream",
"text": [
"WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/resource_variable_ops.py:1817: calling BaseResourceVariable.__init__ (from tensorflow.python.ops.resource_variable_ops) with constraint is deprecated and will be removed in a future version.\n",
"Instructions for updating:\n",
"If using Keras pass *_constraint arguments to layers.\n",
"INFO:tensorflow:Assets written to: /content/dataset/assets\n"
],
"name": "stdout"
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "nB1AsETP2gqw",
"colab_type": "code",
"colab": {}
},
"source": [
"#create model\n",
"model = Sequential()\n",
"#add model layers\n",
"model.add(Conv2D(64, kernel_size=3, activation='relu', input_shape=(224,224,3)))\n",
"model.add(Conv2D(32, kernel_size=3, activation='relu'))\n",
"model.add(Flatten())\n",
"model.add(Dense(8, activation='softmax'))\n",
"model.compile( optimizer='adam',loss='categorical_crossentropy',metrics=['accuracy'])"
],
"execution_count": 36,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "amaC3ICt4XAv",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 289
},
"outputId": "71e1e3ad-1bb4-469c-c758-edc5f1ed34bd"
},
"source": [
"model.summary()"
],
"execution_count": 37,
"outputs": [
{
"output_type": "stream",
"text": [
"Model: \"sequential_1\"\n",
"_________________________________________________________________\n",
"Layer (type) Output Shape Param # \n",
"=================================================================\n",
"conv2d (Conv2D) (None, 222, 222, 64) 1792 \n",
"_________________________________________________________________\n",
"conv2d_1 (Conv2D) (None, 220, 220, 32) 18464 \n",
"_________________________________________________________________\n",
"flatten (Flatten) (None, 1548800) 0 \n",
"_________________________________________________________________\n",
"dense_3 (Dense) (None, 8) 12390408 \n",
"=================================================================\n",
"Total params: 12,410,664\n",
"Trainable params: 12,410,664\n",
"Non-trainable params: 0\n",
"_________________________________________________________________\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "H4_yNvuP4e2c",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 1000
},
"outputId": "ae9a4863-a4b1-403b-8243-9efc373a59f6"
},
"source": [
"history =model.fit_generator(\n",
" datagenerator.flow(training_data, to_categorical(training_labels,8), batch_size=32),\n",
" validation_data=datagenerator.flow(validation_data, to_categorical(validation_label,8), batch_size=32),\n",
" verbose=2,\n",
" epochs=30,\n",
" callbacks=callbacks\n",
")"
],
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"text": [
"Epoch 1/30\n",
"10/10 - 6s - loss: 12.0320 - accuracy: 0.1267 - val_loss: 2.1038 - val_accuracy: 0.1094 - lr: 0.0010\n",
"Epoch 2/30\n",
"10/10 - 5s - loss: 2.1636 - accuracy: 0.2300 - val_loss: 2.0700 - val_accuracy: 0.2969 - lr: 0.0010\n",
"Epoch 3/30\n",
"10/10 - 5s - loss: 2.0504 - accuracy: 0.1700 - val_loss: 1.9631 - val_accuracy: 0.2500 - lr: 0.0010\n",
"Epoch 4/30\n",
"10/10 - 5s - loss: 1.9965 - accuracy: 0.1600 - val_loss: 2.0221 - val_accuracy: 0.1562 - lr: 0.0010\n",
"Epoch 5/30\n",
"10/10 - 5s - loss: 2.0006 - accuracy: 0.2533 - val_loss: 1.9226 - val_accuracy: 0.2969 - lr: 0.0010\n",
"Epoch 6/30\n",
"10/10 - 5s - loss: 1.9014 - accuracy: 0.2833 - val_loss: 1.8029 - val_accuracy: 0.3281 - lr: 0.0010\n",
"Epoch 7/30\n",
"10/10 - 5s - loss: 1.8067 - accuracy: 0.3067 - val_loss: 1.7943 - val_accuracy: 0.3906 - lr: 0.0010\n",
"Epoch 8/30\n",
"10/10 - 5s - loss: 1.8249 - accuracy: 0.3333 - val_loss: 1.8745 - val_accuracy: 0.2500 - lr: 0.0010\n",
"Epoch 9/30\n",
"10/10 - 5s - loss: 1.7950 - accuracy: 0.3033 - val_loss: 1.7712 - val_accuracy: 0.2969 - lr: 0.0010\n",
"Epoch 10/30\n",
"10/10 - 5s - loss: 1.6610 - accuracy: 0.4133 - val_loss: 1.8666 - val_accuracy: 0.3906 - lr: 0.0010\n",
"Epoch 11/30\n",
"10/10 - 5s - loss: 1.6565 - accuracy: 0.3900 - val_loss: 1.8949 - val_accuracy: 0.3281 - lr: 0.0010\n",
"Epoch 12/30\n",
"10/10 - 5s - loss: 1.5739 - accuracy: 0.4133 - val_loss: 1.9954 - val_accuracy: 0.2969 - lr: 0.0010\n",
"Epoch 13/30\n",
"10/10 - 5s - loss: 1.6497 - accuracy: 0.4000 - val_loss: 2.1188 - val_accuracy: 0.2969 - lr: 0.0010\n",
"Epoch 14/30\n",
"\n",
"Epoch 00014: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.\n",
"10/10 - 5s - loss: 1.5774 - accuracy: 0.4633 - val_loss: 2.1318 - val_accuracy: 0.4219 - lr: 0.0010\n",
"Epoch 15/30\n",
"10/10 - 5s - loss: 1.5384 - accuracy: 0.4233 - val_loss: 1.9483 - val_accuracy: 0.4062 - lr: 1.0000e-04\n",
"Epoch 16/30\n",
"10/10 - 5s - loss: 1.4229 - accuracy: 0.4833 - val_loss: 2.1808 - val_accuracy: 0.3125 - lr: 1.0000e-04\n",
"Epoch 17/30\n",
"10/10 - 5s - loss: 1.4122 - accuracy: 0.4667 - val_loss: 2.0510 - val_accuracy: 0.3438 - lr: 1.0000e-04\n",
"Epoch 18/30\n",
"10/10 - 5s - loss: 1.4169 - accuracy: 0.4833 - val_loss: 2.0156 - val_accuracy: 0.4062 - lr: 1.0000e-04\n",
"Epoch 19/30\n",
"\n",
"Epoch 00019: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.\n",
"10/10 - 5s - loss: 1.5091 - accuracy: 0.4733 - val_loss: 2.0512 - val_accuracy: 0.3281 - lr: 1.0000e-04\n",
"Epoch 20/30\n",
"10/10 - 5s - loss: 1.4688 - accuracy: 0.4733 - val_loss: 2.1007 - val_accuracy: 0.3594 - lr: 1.0000e-05\n",
"Epoch 21/30\n",
"10/10 - 5s - loss: 1.4454 - accuracy: 0.4767 - val_loss: 1.9039 - val_accuracy: 0.3125 - lr: 1.0000e-05\n",
"Epoch 22/30\n",
"\n",
"Epoch 00022: ReduceLROnPlateau reducing learning rate to 1e-05.\n",
"10/10 - 5s - loss: 1.4204 - accuracy: 0.4833 - val_loss: 2.0230 - val_accuracy: 0.3125 - lr: 1.0000e-05\n",
"Epoch 23/30\n",
"10/10 - 5s - loss: 1.4356 - accuracy: 0.4933 - val_loss: 1.9726 - val_accuracy: 0.2812 - lr: 1.0000e-05\n",
"Epoch 24/30\n",
"10/10 - 5s - loss: 1.2924 - accuracy: 0.5233 - val_loss: 1.9144 - val_accuracy: 0.3125 - lr: 1.0000e-05\n",
"Epoch 25/30\n",
"10/10 - 5s - loss: 1.3754 - accuracy: 0.4933 - val_loss: 1.8894 - val_accuracy: 0.3750 - lr: 1.0000e-05\n",
"Epoch 26/30\n",
"10/10 - 5s - loss: 1.3555 - accuracy: 0.5167 - val_loss: 1.9851 - val_accuracy: 0.3281 - lr: 1.0000e-05\n",
"Epoch 27/30\n",
"10/10 - 5s - loss: 1.4318 - accuracy: 0.4900 - val_loss: 1.9248 - val_accuracy: 0.4219 - lr: 1.0000e-05\n",
"Epoch 28/30\n",
"10/10 - 5s - loss: 1.4064 - accuracy: 0.5200 - val_loss: 1.9244 - val_accuracy: 0.3438 - lr: 1.0000e-05\n",
"Epoch 29/30\n",
"10/10 - 5s - loss: 1.3949 - accuracy: 0.4867 - val_loss: 1.8455 - val_accuracy: 0.3438 - lr: 1.0000e-05\n",
"Epoch 30/30\n",
"10/10 - 5s - loss: 1.4236 - accuracy: 0.4433 - val_loss: 1.8588 - val_accuracy: 0.2969 - lr: 1.0000e-05\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "dikQ0Cwe43JE",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 562
},
"outputId": "a338eb0b-69ea-41c9-f4c6-af7892d645eb"
},
"source": [
"acc = history.history['accuracy' ]\n",
"val_acc = history.history['val_accuracy' ]\n",
"loss = history.history['loss']\n",
"val_loss = history.history['val_loss' ]\n",
"\n",
"epochs = range(len(acc)) # Get number of epochs\n",
"\n",
"#------------------------------------------------\n",
"# Plot training and validation accuracy per epoch\n",
"#------------------------------------------------\n",
"plt.plot ( epochs, acc )\n",
"plt.plot ( epochs, val_acc )\n",
"plt.title ('Training and validation accuracy')\n",
"plt.figure()\n",
"\n",
"#------------------------------------------------\n",
"# Plot training and validation loss per epoch\n",
"#------------------------------------------------\n",
"plt.plot ( epochs, loss )\n",
"plt.plot ( epochs, val_loss )\n",
"plt.title ('Training and validation loss')"
],
"execution_count": null,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"Text(0.5, 1.0, 'Training and validation loss')"
]
},
"metadata": {
"tags": []
},
"execution_count": 86
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "JjG44bca70MY",
"colab_type": "code",
"colab": {}
},
"source": [
"img_height = 299\n",
"img_width = 299\n",
"\n",
"training_data, training_labels = train_data_preparation(train_fnames, train, train_dir)\n",
"testing_data = test_data_prepare(test_fnames, test_dir)\n",
"xcmod = Xception(include_top=False,weights='imagenet',input_shape=(299,299,3),pooling='avg')"
],
"execution_count": 39,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "YZyQvHDooZlW",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 1000
},
"outputId": "13259079-c212-414c-c3ee-909aac34e73f"
},
"source": [
"xcmod.summary()"
],
"execution_count": 40,
"outputs": [
{
"output_type": "stream",
"text": [
"Model: \"xception\"\n",
"__________________________________________________________________________________________________\n",
"Layer (type) Output Shape Param # Connected to \n",
"==================================================================================================\n",
"input_3 (InputLayer) [(None, 299, 299, 3) 0 \n",
"__________________________________________________________________________________________________\n",
"block1_conv1 (Conv2D) (None, 149, 149, 32) 864 input_3[0][0] \n",
"__________________________________________________________________________________________________\n",
"block1_conv1_bn (BatchNormaliza (None, 149, 149, 32) 128 block1_conv1[0][0] \n",
"__________________________________________________________________________________________________\n",
"block1_conv1_act (Activation) (None, 149, 149, 32) 0 block1_conv1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block1_conv2 (Conv2D) (None, 147, 147, 64) 18432 block1_conv1_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block1_conv2_bn (BatchNormaliza (None, 147, 147, 64) 256 block1_conv2[0][0] \n",
"__________________________________________________________________________________________________\n",
"block1_conv2_act (Activation) (None, 147, 147, 64) 0 block1_conv2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block2_sepconv1 (SeparableConv2 (None, 147, 147, 128 8768 block1_conv2_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block2_sepconv1_bn (BatchNormal (None, 147, 147, 128 512 block2_sepconv1[0][0] \n",
"__________________________________________________________________________________________________\n",
"block2_sepconv2_act (Activation (None, 147, 147, 128 0 block2_sepconv1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block2_sepconv2 (SeparableConv2 (None, 147, 147, 128 17536 block2_sepconv2_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block2_sepconv2_bn (BatchNormal (None, 147, 147, 128 512 block2_sepconv2[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_6 (Conv2D) (None, 74, 74, 128) 8192 block1_conv2_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block2_pool (MaxPooling2D) (None, 74, 74, 128) 0 block2_sepconv2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_4 (BatchNor (None, 74, 74, 128) 512 conv2d_6[0][0] \n",
"__________________________________________________________________________________________________\n",
"add_12 (Add) (None, 74, 74, 128) 0 block2_pool[0][0] \n",
" batch_normalization_4[0][0] \n",
"__________________________________________________________________________________________________\n",
"block3_sepconv1_act (Activation (None, 74, 74, 128) 0 add_12[0][0] \n",
"__________________________________________________________________________________________________\n",
"block3_sepconv1 (SeparableConv2 (None, 74, 74, 256) 33920 block3_sepconv1_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block3_sepconv1_bn (BatchNormal (None, 74, 74, 256) 1024 block3_sepconv1[0][0] \n",
"__________________________________________________________________________________________________\n",
"block3_sepconv2_act (Activation (None, 74, 74, 256) 0 block3_sepconv1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block3_sepconv2 (SeparableConv2 (None, 74, 74, 256) 67840 block3_sepconv2_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block3_sepconv2_bn (BatchNormal (None, 74, 74, 256) 1024 block3_sepconv2[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_7 (Conv2D) (None, 37, 37, 256) 32768 add_12[0][0] \n",
"__________________________________________________________________________________________________\n",
"block3_pool (MaxPooling2D) (None, 37, 37, 256) 0 block3_sepconv2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_5 (BatchNor (None, 37, 37, 256) 1024 conv2d_7[0][0] \n",
"__________________________________________________________________________________________________\n",
"add_13 (Add) (None, 37, 37, 256) 0 block3_pool[0][0] \n",
" batch_normalization_5[0][0] \n",
"__________________________________________________________________________________________________\n",
"block4_sepconv1_act (Activation (None, 37, 37, 256) 0 add_13[0][0] \n",
"__________________________________________________________________________________________________\n",
"block4_sepconv1 (SeparableConv2 (None, 37, 37, 728) 188672 block4_sepconv1_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block4_sepconv1_bn (BatchNormal (None, 37, 37, 728) 2912 block4_sepconv1[0][0] \n",
"__________________________________________________________________________________________________\n",
"block4_sepconv2_act (Activation (None, 37, 37, 728) 0 block4_sepconv1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block4_sepconv2 (SeparableConv2 (None, 37, 37, 728) 536536 block4_sepconv2_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block4_sepconv2_bn (BatchNormal (None, 37, 37, 728) 2912 block4_sepconv2[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_8 (Conv2D) (None, 19, 19, 728) 186368 add_13[0][0] \n",
"__________________________________________________________________________________________________\n",
"block4_pool (MaxPooling2D) (None, 19, 19, 728) 0 block4_sepconv2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_6 (BatchNor (None, 19, 19, 728) 2912 conv2d_8[0][0] \n",
"__________________________________________________________________________________________________\n",
"add_14 (Add) (None, 19, 19, 728) 0 block4_pool[0][0] \n",
" batch_normalization_6[0][0] \n",
"__________________________________________________________________________________________________\n",
"block5_sepconv1_act (Activation (None, 19, 19, 728) 0 add_14[0][0] \n",
"__________________________________________________________________________________________________\n",
"block5_sepconv1 (SeparableConv2 (None, 19, 19, 728) 536536 block5_sepconv1_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block5_sepconv1_bn (BatchNormal (None, 19, 19, 728) 2912 block5_sepconv1[0][0] \n",
"__________________________________________________________________________________________________\n",
"block5_sepconv2_act (Activation (None, 19, 19, 728) 0 block5_sepconv1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block5_sepconv2 (SeparableConv2 (None, 19, 19, 728) 536536 block5_sepconv2_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block5_sepconv2_bn (BatchNormal (None, 19, 19, 728) 2912 block5_sepconv2[0][0] \n",
"__________________________________________________________________________________________________\n",
"block5_sepconv3_act (Activation (None, 19, 19, 728) 0 block5_sepconv2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block5_sepconv3 (SeparableConv2 (None, 19, 19, 728) 536536 block5_sepconv3_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block5_sepconv3_bn (BatchNormal (None, 19, 19, 728) 2912 block5_sepconv3[0][0] \n",
"__________________________________________________________________________________________________\n",
"add_15 (Add) (None, 19, 19, 728) 0 block5_sepconv3_bn[0][0] \n",
" add_14[0][0] \n",
"__________________________________________________________________________________________________\n",
"block6_sepconv1_act (Activation (None, 19, 19, 728) 0 add_15[0][0] \n",
"__________________________________________________________________________________________________\n",
"block6_sepconv1 (SeparableConv2 (None, 19, 19, 728) 536536 block6_sepconv1_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block6_sepconv1_bn (BatchNormal (None, 19, 19, 728) 2912 block6_sepconv1[0][0] \n",
"__________________________________________________________________________________________________\n",
"block6_sepconv2_act (Activation (None, 19, 19, 728) 0 block6_sepconv1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block6_sepconv2 (SeparableConv2 (None, 19, 19, 728) 536536 block6_sepconv2_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block6_sepconv2_bn (BatchNormal (None, 19, 19, 728) 2912 block6_sepconv2[0][0] \n",
"__________________________________________________________________________________________________\n",
"block6_sepconv3_act (Activation (None, 19, 19, 728) 0 block6_sepconv2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block6_sepconv3 (SeparableConv2 (None, 19, 19, 728) 536536 block6_sepconv3_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block6_sepconv3_bn (BatchNormal (None, 19, 19, 728) 2912 block6_sepconv3[0][0] \n",
"__________________________________________________________________________________________________\n",
"add_16 (Add) (None, 19, 19, 728) 0 block6_sepconv3_bn[0][0] \n",
" add_15[0][0] \n",
"__________________________________________________________________________________________________\n",
"block7_sepconv1_act (Activation (None, 19, 19, 728) 0 add_16[0][0] \n",
"__________________________________________________________________________________________________\n",
"block7_sepconv1 (SeparableConv2 (None, 19, 19, 728) 536536 block7_sepconv1_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block7_sepconv1_bn (BatchNormal (None, 19, 19, 728) 2912 block7_sepconv1[0][0] \n",
"__________________________________________________________________________________________________\n",
"block7_sepconv2_act (Activation (None, 19, 19, 728) 0 block7_sepconv1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block7_sepconv2 (SeparableConv2 (None, 19, 19, 728) 536536 block7_sepconv2_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block7_sepconv2_bn (BatchNormal (None, 19, 19, 728) 2912 block7_sepconv2[0][0] \n",
"__________________________________________________________________________________________________\n",
"block7_sepconv3_act (Activation (None, 19, 19, 728) 0 block7_sepconv2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block7_sepconv3 (SeparableConv2 (None, 19, 19, 728) 536536 block7_sepconv3_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block7_sepconv3_bn (BatchNormal (None, 19, 19, 728) 2912 block7_sepconv3[0][0] \n",
"__________________________________________________________________________________________________\n",
"add_17 (Add) (None, 19, 19, 728) 0 block7_sepconv3_bn[0][0] \n",
" add_16[0][0] \n",
"__________________________________________________________________________________________________\n",
"block8_sepconv1_act (Activation (None, 19, 19, 728) 0 add_17[0][0] \n",
"__________________________________________________________________________________________________\n",
"block8_sepconv1 (SeparableConv2 (None, 19, 19, 728) 536536 block8_sepconv1_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block8_sepconv1_bn (BatchNormal (None, 19, 19, 728) 2912 block8_sepconv1[0][0] \n",
"__________________________________________________________________________________________________\n",
"block8_sepconv2_act (Activation (None, 19, 19, 728) 0 block8_sepconv1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block8_sepconv2 (SeparableConv2 (None, 19, 19, 728) 536536 block8_sepconv2_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block8_sepconv2_bn (BatchNormal (None, 19, 19, 728) 2912 block8_sepconv2[0][0] \n",
"__________________________________________________________________________________________________\n",
"block8_sepconv3_act (Activation (None, 19, 19, 728) 0 block8_sepconv2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block8_sepconv3 (SeparableConv2 (None, 19, 19, 728) 536536 block8_sepconv3_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block8_sepconv3_bn (BatchNormal (None, 19, 19, 728) 2912 block8_sepconv3[0][0] \n",
"__________________________________________________________________________________________________\n",
"add_18 (Add) (None, 19, 19, 728) 0 block8_sepconv3_bn[0][0] \n",
" add_17[0][0] \n",
"__________________________________________________________________________________________________\n",
"block9_sepconv1_act (Activation (None, 19, 19, 728) 0 add_18[0][0] \n",
"__________________________________________________________________________________________________\n",
"block9_sepconv1 (SeparableConv2 (None, 19, 19, 728) 536536 block9_sepconv1_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block9_sepconv1_bn (BatchNormal (None, 19, 19, 728) 2912 block9_sepconv1[0][0] \n",
"__________________________________________________________________________________________________\n",
"block9_sepconv2_act (Activation (None, 19, 19, 728) 0 block9_sepconv1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block9_sepconv2 (SeparableConv2 (None, 19, 19, 728) 536536 block9_sepconv2_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block9_sepconv2_bn (BatchNormal (None, 19, 19, 728) 2912 block9_sepconv2[0][0] \n",
"__________________________________________________________________________________________________\n",
"block9_sepconv3_act (Activation (None, 19, 19, 728) 0 block9_sepconv2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block9_sepconv3 (SeparableConv2 (None, 19, 19, 728) 536536 block9_sepconv3_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block9_sepconv3_bn (BatchNormal (None, 19, 19, 728) 2912 block9_sepconv3[0][0] \n",
"__________________________________________________________________________________________________\n",
"add_19 (Add) (None, 19, 19, 728) 0 block9_sepconv3_bn[0][0] \n",
" add_18[0][0] \n",
"__________________________________________________________________________________________________\n",
"block10_sepconv1_act (Activatio (None, 19, 19, 728) 0 add_19[0][0] \n",
"__________________________________________________________________________________________________\n",
"block10_sepconv1 (SeparableConv (None, 19, 19, 728) 536536 block10_sepconv1_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block10_sepconv1_bn (BatchNorma (None, 19, 19, 728) 2912 block10_sepconv1[0][0] \n",
"__________________________________________________________________________________________________\n",
"block10_sepconv2_act (Activatio (None, 19, 19, 728) 0 block10_sepconv1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block10_sepconv2 (SeparableConv (None, 19, 19, 728) 536536 block10_sepconv2_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block10_sepconv2_bn (BatchNorma (None, 19, 19, 728) 2912 block10_sepconv2[0][0] \n",
"__________________________________________________________________________________________________\n",
"block10_sepconv3_act (Activatio (None, 19, 19, 728) 0 block10_sepconv2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block10_sepconv3 (SeparableConv (None, 19, 19, 728) 536536 block10_sepconv3_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block10_sepconv3_bn (BatchNorma (None, 19, 19, 728) 2912 block10_sepconv3[0][0] \n",
"__________________________________________________________________________________________________\n",
"add_20 (Add) (None, 19, 19, 728) 0 block10_sepconv3_bn[0][0] \n",
" add_19[0][0] \n",
"__________________________________________________________________________________________________\n",
"block11_sepconv1_act (Activatio (None, 19, 19, 728) 0 add_20[0][0] \n",
"__________________________________________________________________________________________________\n",
"block11_sepconv1 (SeparableConv (None, 19, 19, 728) 536536 block11_sepconv1_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block11_sepconv1_bn (BatchNorma (None, 19, 19, 728) 2912 block11_sepconv1[0][0] \n",
"__________________________________________________________________________________________________\n",
"block11_sepconv2_act (Activatio (None, 19, 19, 728) 0 block11_sepconv1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block11_sepconv2 (SeparableConv (None, 19, 19, 728) 536536 block11_sepconv2_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block11_sepconv2_bn (BatchNorma (None, 19, 19, 728) 2912 block11_sepconv2[0][0] \n",
"__________________________________________________________________________________________________\n",
"block11_sepconv3_act (Activatio (None, 19, 19, 728) 0 block11_sepconv2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block11_sepconv3 (SeparableConv (None, 19, 19, 728) 536536 block11_sepconv3_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block11_sepconv3_bn (BatchNorma (None, 19, 19, 728) 2912 block11_sepconv3[0][0] \n",
"__________________________________________________________________________________________________\n",
"add_21 (Add) (None, 19, 19, 728) 0 block11_sepconv3_bn[0][0] \n",
" add_20[0][0] \n",
"__________________________________________________________________________________________________\n",
"block12_sepconv1_act (Activatio (None, 19, 19, 728) 0 add_21[0][0] \n",
"__________________________________________________________________________________________________\n",
"block12_sepconv1 (SeparableConv (None, 19, 19, 728) 536536 block12_sepconv1_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block12_sepconv1_bn (BatchNorma (None, 19, 19, 728) 2912 block12_sepconv1[0][0] \n",
"__________________________________________________________________________________________________\n",
"block12_sepconv2_act (Activatio (None, 19, 19, 728) 0 block12_sepconv1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block12_sepconv2 (SeparableConv (None, 19, 19, 728) 536536 block12_sepconv2_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block12_sepconv2_bn (BatchNorma (None, 19, 19, 728) 2912 block12_sepconv2[0][0] \n",
"__________________________________________________________________________________________________\n",
"block12_sepconv3_act (Activatio (None, 19, 19, 728) 0 block12_sepconv2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block12_sepconv3 (SeparableConv (None, 19, 19, 728) 536536 block12_sepconv3_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block12_sepconv3_bn (BatchNorma (None, 19, 19, 728) 2912 block12_sepconv3[0][0] \n",
"__________________________________________________________________________________________________\n",
"add_22 (Add) (None, 19, 19, 728) 0 block12_sepconv3_bn[0][0] \n",
" add_21[0][0] \n",
"__________________________________________________________________________________________________\n",
"block13_sepconv1_act (Activatio (None, 19, 19, 728) 0 add_22[0][0] \n",
"__________________________________________________________________________________________________\n",
"block13_sepconv1 (SeparableConv (None, 19, 19, 728) 536536 block13_sepconv1_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block13_sepconv1_bn (BatchNorma (None, 19, 19, 728) 2912 block13_sepconv1[0][0] \n",
"__________________________________________________________________________________________________\n",
"block13_sepconv2_act (Activatio (None, 19, 19, 728) 0 block13_sepconv1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block13_sepconv2 (SeparableConv (None, 19, 19, 1024) 752024 block13_sepconv2_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block13_sepconv2_bn (BatchNorma (None, 19, 19, 1024) 4096 block13_sepconv2[0][0] \n",
"__________________________________________________________________________________________________\n",
"conv2d_9 (Conv2D) (None, 10, 10, 1024) 745472 add_22[0][0] \n",
"__________________________________________________________________________________________________\n",
"block13_pool (MaxPooling2D) (None, 10, 10, 1024) 0 block13_sepconv2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"batch_normalization_7 (BatchNor (None, 10, 10, 1024) 4096 conv2d_9[0][0] \n",
"__________________________________________________________________________________________________\n",
"add_23 (Add) (None, 10, 10, 1024) 0 block13_pool[0][0] \n",
" batch_normalization_7[0][0] \n",
"__________________________________________________________________________________________________\n",
"block14_sepconv1 (SeparableConv (None, 10, 10, 1536) 1582080 add_23[0][0] \n",
"__________________________________________________________________________________________________\n",
"block14_sepconv1_bn (BatchNorma (None, 10, 10, 1536) 6144 block14_sepconv1[0][0] \n",
"__________________________________________________________________________________________________\n",
"block14_sepconv1_act (Activatio (None, 10, 10, 1536) 0 block14_sepconv1_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"block14_sepconv2 (SeparableConv (None, 10, 10, 2048) 3159552 block14_sepconv1_act[0][0] \n",
"__________________________________________________________________________________________________\n",
"block14_sepconv2_bn (BatchNorma (None, 10, 10, 2048) 8192 block14_sepconv2[0][0] \n",
"__________________________________________________________________________________________________\n",
"block14_sepconv2_act (Activatio (None, 10, 10, 2048) 0 block14_sepconv2_bn[0][0] \n",
"__________________________________________________________________________________________________\n",
"global_average_pooling2d_2 (Glo (None, 2048) 0 block14_sepconv2_act[0][0] \n",
"==================================================================================================\n",
"Total params: 20,861,480\n",
"Trainable params: 20,806,952\n",
"Non-trainable params: 54,528\n",
"__________________________________________________________________________________________________\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "pSpn5nKPomUV",
"colab_type": "code",
"colab": {}
},
"source": [
"xcmod.trainable = False\n",
"model = Sequential([\n",
" xcmod, \n",
" Dense(1024, activation='relu'),\n",
" Dropout(0.4),\n",
" Dense(512,activation='relu'),\n",
" Dropout(0.5),\n",
" Dense(256, activation='relu'),\n",
" Dense(8, activation='softmax'),\n",
"])"
],
"execution_count": 42,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "MGrObQINo-Ym",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 391
},
"outputId": "52f31c9c-425b-478d-c793-953bd919c1dd"
},
"source": [
"model.summary()"
],
"execution_count": 43,
"outputs": [
{
"output_type": "stream",
"text": [
"Model: \"sequential_2\"\n",
"_________________________________________________________________\n",
"Layer (type) Output Shape Param # \n",
"=================================================================\n",
"xception (Model) (None, 2048) 20861480 \n",
"_________________________________________________________________\n",
"dense_4 (Dense) (None, 1024) 2098176 \n",
"_________________________________________________________________\n",
"dropout_1 (Dropout) (None, 1024) 0 \n",
"_________________________________________________________________\n",
"dense_5 (Dense) (None, 512) 524800 \n",
"_________________________________________________________________\n",
"dropout_2 (Dropout) (None, 512) 0 \n",
"_________________________________________________________________\n",
"dense_6 (Dense) (None, 256) 131328 \n",
"_________________________________________________________________\n",
"dense_7 (Dense) (None, 8) 2056 \n",
"=================================================================\n",
"Total params: 23,617,840\n",
"Trainable params: 2,756,360\n",
"Non-trainable params: 20,861,480\n",
"_________________________________________________________________\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "_p03ZLZdpzec",
"colab_type": "code",
"colab": {}
},
"source": [
"datagenerator = ImageDataGenerator(\n",
" rescale=1. / 255,\n",
" featurewise_center=False, \n",
" samplewise_center=False, \n",
" featurewise_std_normalization=False, \n",
" samplewise_std_normalization=False, \n",
" rotation_range=10, \n",
" zoom_range = 0.10, \n",
" width_shift_range=0.10, \n",
" height_shift_range=0.10, \n",
" horizontal_flip=True, \n",
" vertical_flip=False) \n",
"\n",
"\n",
"test_datagenerator=ImageDataGenerator(\n",
" rescale=1. / 255\n",
")\n",
"datagenerator.fit(training_data)\n",
"test_datagenerator.fit(training_data)\n",
"training_data=np.array(training_data)\n",
"testing_data=np.array(testing_data)\n",
"\n",
"le =LabelEncoder()\n",
"training_labels=le.fit_transform(training_labels)"
],
"execution_count": 46,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "Vql9A5JQpB9h",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 1000
},
"outputId": "6716e14c-7a8a-4adf-c504-e533c77736f2"
},
"source": [
"# subsetting validation data\n",
"validation_data = training_data[300:]\n",
"validation_label =training_labels[300:]\n",
"model.compile( optimizer='adam',loss='categorical_crossentropy',metrics=['accuracy'])\n",
"history =model.fit_generator(\n",
" datagenerator.flow(training_data, to_categorical(training_labels,8), batch_size=32),\n",
" validation_data=datagenerator.flow(validation_data, to_categorical(validation_label,8), batch_size=32),\n",
" verbose=2,\n",
" epochs=30,\n",
" callbacks=callbacks\n",
")"
],
"execution_count": 47,
"outputs": [
{
"output_type": "stream",
"text": [
"Epoch 1/30\n",
"12/12 - 12s - loss: 2.0560 - accuracy: 0.2060 - val_loss: 1.6165 - val_accuracy: 0.3906 - lr: 0.0010\n",
"Epoch 2/30\n",
"12/12 - 11s - loss: 1.7635 - accuracy: 0.3187 - val_loss: 1.2267 - val_accuracy: 0.6406 - lr: 0.0010\n",
"Epoch 3/30\n",
"12/12 - 11s - loss: 1.4761 - accuracy: 0.4423 - val_loss: 1.0161 - val_accuracy: 0.6562 - lr: 0.0010\n",
"Epoch 4/30\n",
"12/12 - 11s - loss: 1.3347 - accuracy: 0.4753 - val_loss: 0.8744 - val_accuracy: 0.7812 - lr: 0.0010\n",
"Epoch 5/30\n",
"12/12 - 11s - loss: 1.1286 - accuracy: 0.6071 - val_loss: 0.7694 - val_accuracy: 0.7656 - lr: 0.0010\n",
"Epoch 6/30\n",
"12/12 - 11s - loss: 0.9766 - accuracy: 0.6566 - val_loss: 0.6724 - val_accuracy: 0.7656 - lr: 0.0010\n",
"Epoch 7/30\n",
"12/12 - 11s - loss: 0.9507 - accuracy: 0.6291 - val_loss: 0.5133 - val_accuracy: 0.8281 - lr: 0.0010\n",
"Epoch 8/30\n",
"12/12 - 11s - loss: 0.7980 - accuracy: 0.7088 - val_loss: 0.4421 - val_accuracy: 0.8906 - lr: 0.0010\n",
"Epoch 9/30\n",
"12/12 - 11s - loss: 0.7269 - accuracy: 0.7363 - val_loss: 0.4933 - val_accuracy: 0.8438 - lr: 0.0010\n",
"Epoch 10/30\n",
"12/12 - 11s - loss: 0.6605 - accuracy: 0.7582 - val_loss: 0.4024 - val_accuracy: 0.8750 - lr: 0.0010\n",
"Epoch 11/30\n",
"12/12 - 11s - loss: 0.6109 - accuracy: 0.7885 - val_loss: 0.3098 - val_accuracy: 0.9062 - lr: 0.0010\n",
"Epoch 12/30\n",
"12/12 - 11s - loss: 0.5595 - accuracy: 0.7830 - val_loss: 0.3240 - val_accuracy: 0.8750 - lr: 0.0010\n",
"Epoch 13/30\n",
"12/12 - 11s - loss: 0.4885 - accuracy: 0.8297 - val_loss: 0.3765 - val_accuracy: 0.8750 - lr: 0.0010\n",
"Epoch 14/30\n",
"12/12 - 11s - loss: 0.4539 - accuracy: 0.8269 - val_loss: 0.1817 - val_accuracy: 0.9531 - lr: 0.0010\n",
"Epoch 15/30\n",
"12/12 - 11s - loss: 0.4273 - accuracy: 0.8242 - val_loss: 0.2014 - val_accuracy: 0.9531 - lr: 0.0010\n",
"Epoch 16/30\n",
"12/12 - 11s - loss: 0.3790 - accuracy: 0.8489 - val_loss: 0.2087 - val_accuracy: 0.9219 - lr: 0.0010\n",
"Epoch 17/30\n",
"12/12 - 11s - loss: 0.3521 - accuracy: 0.8901 - val_loss: 0.3155 - val_accuracy: 0.8906 - lr: 0.0010\n",
"Epoch 18/30\n",
"12/12 - 11s - loss: 0.3265 - accuracy: 0.8846 - val_loss: 0.1752 - val_accuracy: 0.9531 - lr: 0.0010\n",
"Epoch 19/30\n",
"12/12 - 11s - loss: 0.3079 - accuracy: 0.8984 - val_loss: 0.1545 - val_accuracy: 0.9375 - lr: 0.0010\n",
"Epoch 20/30\n",
"12/12 - 11s - loss: 0.3749 - accuracy: 0.8599 - val_loss: 0.2385 - val_accuracy: 0.9531 - lr: 0.0010\n",
"Epoch 21/30\n",
"12/12 - 11s - loss: 0.2517 - accuracy: 0.9148 - val_loss: 0.1358 - val_accuracy: 0.9531 - lr: 0.0010\n",
"Epoch 22/30\n",
"12/12 - 11s - loss: 0.2291 - accuracy: 0.9286 - val_loss: 0.2002 - val_accuracy: 0.9219 - lr: 0.0010\n",
"Epoch 23/30\n",
"12/12 - 12s - loss: 0.3213 - accuracy: 0.8791 - val_loss: 0.1503 - val_accuracy: 0.9688 - lr: 0.0010\n",
"Epoch 24/30\n",
"\n",
"Epoch 00024: ReduceLROnPlateau reducing learning rate to 0.00010000000474974513.\n",
"12/12 - 12s - loss: 0.3810 - accuracy: 0.8626 - val_loss: 0.1167 - val_accuracy: 1.0000 - lr: 0.0010\n",
"Epoch 25/30\n",
"12/12 - 11s - loss: 0.1940 - accuracy: 0.9313 - val_loss: 0.0496 - val_accuracy: 1.0000 - lr: 1.0000e-04\n",
"Epoch 26/30\n",
"12/12 - 11s - loss: 0.1437 - accuracy: 0.9560 - val_loss: 0.0597 - val_accuracy: 1.0000 - lr: 1.0000e-04\n",
"Epoch 27/30\n",
"12/12 - 11s - loss: 0.1854 - accuracy: 0.9313 - val_loss: 0.0578 - val_accuracy: 1.0000 - lr: 1.0000e-04\n",
"Epoch 28/30\n",
"\n",
"Epoch 00028: ReduceLROnPlateau reducing learning rate to 1.0000000474974514e-05.\n",
"12/12 - 11s - loss: 0.1664 - accuracy: 0.9478 - val_loss: 0.0860 - val_accuracy: 0.9688 - lr: 1.0000e-04\n",
"Epoch 29/30\n",
"12/12 - 11s - loss: 0.1568 - accuracy: 0.9505 - val_loss: 0.0592 - val_accuracy: 1.0000 - lr: 1.0000e-05\n",
"Epoch 30/30\n",
"12/12 - 11s - loss: 0.1691 - accuracy: 0.9423 - val_loss: 0.0374 - val_accuracy: 1.0000 - lr: 1.0000e-05\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "6za5-whYregG",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 562
},
"outputId": "ed325d09-656c-45d5-c8ac-2a7c3727f1a6"
},
"source": [
"acc = history.history[ 'accuracy' ]\n",
"val_acc = history.history[ 'val_accuracy' ]\n",
"loss = history.history[ 'loss' ]\n",
"val_loss = history.history['val_loss' ]\n",
"\n",
"epochs = range(len(acc)) # Get number of epochs\n",
"\n",
"#------------------------------------------------\n",
"# Plot training and validation accuracy per epoch\n",
"#------------------------------------------------\n",
"plt.plot ( epochs, acc )\n",
"plt.plot ( epochs, val_acc )\n",
"plt.title ('Training and validation accuracy')\n",
"plt.figure()\n",
"\n",
"#------------------------------------------------\n",
"# Plot training and validation loss per epoch\n",
"#------------------------------------------------\n",
"plt.plot ( epochs, loss )\n",
"plt.plot ( epochs, val_loss )\n",
"plt.title ('Training and validation loss' )"
],
"execution_count": 48,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"Text(0.5, 1.0, 'Training and validation loss')"
]
},
"metadata": {
"tags": []
},
"execution_count": 48
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "McInHDborlnA",
"colab_type": "code",
"colab": {}
},
"source": [
"labels = model.predict(testing_data)\n",
"label = [np.argmax(i) for i in labels]\n",
"target=le.inverse_transform(label)\n",
"submission = pd.DataFrame({ 'Image': test.Image, 'target': target })\n",
"\n",
"with open('output.csv','w+') as f:\n",
" submission.to_csv(f)"
],
"execution_count": 49,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "QG8spDbNsKpR",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 1000
},
"outputId": "90178d56-ba64-400e-adfc-f0242185e9c1"
},
"source": [
"show_batch(testing_data,target)"
],
"execution_count": 51,
"outputs": [
{
"output_type": "stream",
"text": [
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
"Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
],
"name": "stderr"
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 864x864 with 25 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "o4Bc31lMtCPM",
"colab_type": "code",
"colab": {}
},
"source": [
"\n",
"submission = pd.DataFrame({ 'Image': test.Image, 'target': target })\n",
"\n",
"submission.to_csv('/content/out.csv',index=False)"
],
"execution_count": 54,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "bML78WS9tYo2",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 204
},
"outputId": "d46a68b4-8c8b-4925-f417-5d3817803f29"
},
"source": [
"df=pd.read_csv('/content/out.csv')\n",
"df.head()"
],
"execution_count": 55,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Image</th>\n",
" <th>target</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>508.jpg</td>\n",
" <td>kathakali</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>246.jpg</td>\n",
" <td>kathakali</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>473.jpg</td>\n",
" <td>kathakali</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>485.jpg</td>\n",
" <td>mohiniyattam</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>128.jpg</td>\n",
" <td>sattriya</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Image target\n",
"0 508.jpg kathakali\n",
"1 246.jpg kathakali\n",
"2 473.jpg kathakali\n",
"3 485.jpg mohiniyattam\n",
"4 128.jpg sattriya"
]
},
"metadata": {
"tags": []
},
"execution_count": 55
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "cm4Q6FPFzEcn",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 34
},
"outputId": "c4f4c061-063a-4e2d-d096-0483c49cc53b"
},
"source": [
"model.save('/content/model')"
],
"execution_count": 56,
"outputs": [
{
"output_type": "stream",
"text": [
"INFO:tensorflow:Assets written to: /content/model/assets\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "6PPUhyEpzQ6W",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 136
},
"outputId": "58049f7a-80e1-4ad6-969f-fc9e05659b65"
},
"source": [
"!zip -r model.zip /content/model"
],
"execution_count": 58,
"outputs": [
{
"output_type": "stream",
"text": [
" adding: content/model/ (stored 0%)\n",
" adding: content/model/variables/ (stored 0%)\n",
" adding: content/model/variables/variables.index (deflated 77%)\n",
" adding: content/model/variables/variables.data-00000-of-00002 (deflated 85%)\n",
" adding: content/model/variables/variables.data-00001-of-00002 (deflated 7%)\n",
" adding: content/model/assets/ (stored 0%)\n",
" adding: content/model/saved_model.pb (deflated 93%)\n"
],
"name": "stdout"
}
]
}
]
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment