From 34f2dbbb9dab63ddf47235a717085f39f024a336 Mon Sep 17 00:00:00 2001
From: SrijanShovit <86592220+SrijanShovit@users.noreply.github.com>
Date: Fri, 21 Jan 2022 23:51:56 +0530
Subject: [PATCH] Add files via upload
---
cnntrain.ipynb | 2094 +++++++++++++++++++++++++++++++++++++++++++++++
seti_trained.h5 | Bin 0 -> 24707840 bytes
2 files changed, 2094 insertions(+)
create mode 100644 cnntrain.ipynb
create mode 100644 seti_trained.h5
diff --git a/cnntrain.ipynb b/cnntrain.ipynb
new file mode 100644
index 0000000..d99b5d8
--- /dev/null
+++ b/cnntrain.ipynb
@@ -0,0 +1,2094 @@
+{
+ "cells": [
+ {
+ "cell_type": "code",
+ "execution_count": 96,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "#Importing libraries\n",
+ "import pandas as pd\n",
+ "import numpy as np\n",
+ "import tensorflow as tf\n",
+ "from sklearn.model_selection import train_test_split\n",
+ "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
+ "from tensorflow import keras\n",
+ "from tensorflow.keras.layers import Conv2D,MaxPooling2D,Flatten,Dense\n",
+ "from tensorflow.keras.models import Sequential\n",
+ "from tensorflow.keras.callbacks import EarlyStopping, ModelCheckpoint\n",
+ "import matplotlib.pyplot as plt\n",
+ "from keras.layers import LeakyReLU\n",
+ "from sklearn import metrics\n",
+ "from sklearn.metrics import confusion_matrix\n",
+ "import seaborn as sns\n",
+ "from keras.models import load_model"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Load and Pre-process the dataset"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 21,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "img_train = pd.read_csv('./train/images.csv') \n",
+ "img_val = pd.read_csv('./valid/images.csv') \n",
+ "label_train = pd.read_csv('./train/labels.csv')\n",
+ "label_val = pd.read_csv('./valid/labels.csv')"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 22,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "3199\n",
+ "3199\n",
+ "799\n",
+ "799\n"
+ ]
+ }
+ ],
+ "source": [
+ "print(len(img_train))\n",
+ "print(len(label_train))\n",
+ "print(len(img_val))\n",
+ "print(len(label_val))"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 23,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "
\n",
+ "\n",
+ "
\n",
+ " \n",
+ " \n",
+ " | \n",
+ " 0.631373 | \n",
+ " 0.623529 | \n",
+ " 0.713726 | \n",
+ " 0.705882 | \n",
+ " 0.658824 | \n",
+ " 0.666667 | \n",
+ " 0.654902 | \n",
+ " 0.635294 | \n",
+ " 0.647059 | \n",
+ " 0.705882.1 | \n",
+ " ... | \n",
+ " 0.682353.349 | \n",
+ " 0.611765.91 | \n",
+ " 0.650980.397 | \n",
+ " 0.658824.443 | \n",
+ " 0.600000.48 | \n",
+ " 0.603922.77 | \n",
+ " 0.654902.426 | \n",
+ " 0.694118.222 | \n",
+ " 0.658824.444 | \n",
+ " 0.666667.441 | \n",
+ "
\n",
+ " \n",
+ " \n",
+ " \n",
+ " 0 | \n",
+ " 0.725490 | \n",
+ " 0.752941 | \n",
+ " 0.749020 | \n",
+ " 0.701961 | \n",
+ " 0.690196 | \n",
+ " 0.721569 | \n",
+ " 0.709804 | \n",
+ " 0.745098 | \n",
+ " 0.654902 | \n",
+ " 0.721569 | \n",
+ " ... | \n",
+ " 0.721569 | \n",
+ " 0.698039 | \n",
+ " 0.721569 | \n",
+ " 0.686275 | \n",
+ " 0.713726 | \n",
+ " 0.682353 | \n",
+ " 0.690196 | \n",
+ " 0.698039 | \n",
+ " 0.701961 | \n",
+ " 0.725490 | \n",
+ "
\n",
+ " \n",
+ " 1 | \n",
+ " 0.717647 | \n",
+ " 0.701961 | \n",
+ " 0.713726 | \n",
+ " 0.733333 | \n",
+ " 0.705882 | \n",
+ " 0.717647 | \n",
+ " 0.725490 | \n",
+ " 0.682353 | \n",
+ " 0.717647 | \n",
+ " 0.674510 | \n",
+ " ... | \n",
+ " 0.709804 | \n",
+ " 0.694118 | \n",
+ " 0.705882 | \n",
+ " 0.682353 | \n",
+ " 0.639216 | \n",
+ " 0.713726 | \n",
+ " 0.670588 | \n",
+ " 0.678431 | \n",
+ " 0.737255 | \n",
+ " 0.674510 | \n",
+ "
\n",
+ " \n",
+ " 2 | \n",
+ " 0.705882 | \n",
+ " 0.674510 | \n",
+ " 0.654902 | \n",
+ " 0.678431 | \n",
+ " 0.666667 | \n",
+ " 0.662745 | \n",
+ " 0.678431 | \n",
+ " 0.662745 | \n",
+ " 0.686275 | \n",
+ " 0.686275 | \n",
+ " ... | \n",
+ " 0.639216 | \n",
+ " 0.662745 | \n",
+ " 0.631373 | \n",
+ " 0.643137 | \n",
+ " 0.705882 | \n",
+ " 0.662745 | \n",
+ " 0.705882 | \n",
+ " 0.666667 | \n",
+ " 0.654902 | \n",
+ " 0.631373 | \n",
+ "
\n",
+ " \n",
+ " 3 | \n",
+ " 0.647059 | \n",
+ " 0.729412 | \n",
+ " 0.701961 | \n",
+ " 0.674510 | \n",
+ " 0.611765 | \n",
+ " 0.698039 | \n",
+ " 0.713726 | \n",
+ " 0.662745 | \n",
+ " 0.701961 | \n",
+ " 0.674510 | \n",
+ " ... | \n",
+ " 0.639216 | \n",
+ " 0.670588 | \n",
+ " 0.705882 | \n",
+ " 0.674510 | \n",
+ " 0.721569 | \n",
+ " 0.694118 | \n",
+ " 0.674510 | \n",
+ " 0.705882 | \n",
+ " 0.749020 | \n",
+ " 0.729412 | \n",
+ "
\n",
+ " \n",
+ " 4 | \n",
+ " 0.694118 | \n",
+ " 0.682353 | \n",
+ " 0.705882 | \n",
+ " 0.705882 | \n",
+ " 0.666667 | \n",
+ " 0.694118 | \n",
+ " 0.674510 | \n",
+ " 0.713726 | \n",
+ " 0.690196 | \n",
+ " 0.709804 | \n",
+ " ... | \n",
+ " 0.658824 | \n",
+ " 0.698039 | \n",
+ " 0.694118 | \n",
+ " 0.709804 | \n",
+ " 0.725490 | \n",
+ " 0.694118 | \n",
+ " 0.729412 | \n",
+ " 0.701961 | \n",
+ " 0.694118 | \n",
+ " 0.709804 | \n",
+ "
\n",
+ " \n",
+ "
\n",
+ "
5 rows × 8192 columns
\n",
+ "
"
+ ],
+ "text/plain": [
+ " 0.631373 0.623529 0.713726 0.705882 0.658824 0.666667 0.654902 \\\n",
+ "0 0.725490 0.752941 0.749020 0.701961 0.690196 0.721569 0.709804 \n",
+ "1 0.717647 0.701961 0.713726 0.733333 0.705882 0.717647 0.725490 \n",
+ "2 0.705882 0.674510 0.654902 0.678431 0.666667 0.662745 0.678431 \n",
+ "3 0.647059 0.729412 0.701961 0.674510 0.611765 0.698039 0.713726 \n",
+ "4 0.694118 0.682353 0.705882 0.705882 0.666667 0.694118 0.674510 \n",
+ "\n",
+ " 0.635294 0.647059 0.705882.1 ... 0.682353.349 0.611765.91 \\\n",
+ "0 0.745098 0.654902 0.721569 ... 0.721569 0.698039 \n",
+ "1 0.682353 0.717647 0.674510 ... 0.709804 0.694118 \n",
+ "2 0.662745 0.686275 0.686275 ... 0.639216 0.662745 \n",
+ "3 0.662745 0.701961 0.674510 ... 0.639216 0.670588 \n",
+ "4 0.713726 0.690196 0.709804 ... 0.658824 0.698039 \n",
+ "\n",
+ " 0.650980.397 0.658824.443 0.600000.48 0.603922.77 0.654902.426 \\\n",
+ "0 0.721569 0.686275 0.713726 0.682353 0.690196 \n",
+ "1 0.705882 0.682353 0.639216 0.713726 0.670588 \n",
+ "2 0.631373 0.643137 0.705882 0.662745 0.705882 \n",
+ "3 0.705882 0.674510 0.721569 0.694118 0.674510 \n",
+ "4 0.694118 0.709804 0.725490 0.694118 0.729412 \n",
+ "\n",
+ " 0.694118.222 0.658824.444 0.666667.441 \n",
+ "0 0.698039 0.701961 0.725490 \n",
+ "1 0.678431 0.737255 0.674510 \n",
+ "2 0.666667 0.654902 0.631373 \n",
+ "3 0.705882 0.749020 0.729412 \n",
+ "4 0.701961 0.694118 0.709804 \n",
+ "\n",
+ "[5 rows x 8192 columns]"
+ ]
+ },
+ "execution_count": 23,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "img_train.head()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 24,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "\n",
+ "\n",
+ "
\n",
+ " \n",
+ " \n",
+ " | \n",
+ " 1.000000 | \n",
+ " 0.000000 | \n",
+ " 0.000000.1 | \n",
+ " 0.000000.2 | \n",
+ "
\n",
+ " \n",
+ " \n",
+ " \n",
+ " 0 | \n",
+ " 1.0 | \n",
+ " 0.0 | \n",
+ " 0.0 | \n",
+ " 0.0 | \n",
+ "
\n",
+ " \n",
+ " 1 | \n",
+ " 1.0 | \n",
+ " 0.0 | \n",
+ " 0.0 | \n",
+ " 0.0 | \n",
+ "
\n",
+ " \n",
+ " 2 | \n",
+ " 1.0 | \n",
+ " 0.0 | \n",
+ " 0.0 | \n",
+ " 0.0 | \n",
+ "
\n",
+ " \n",
+ " 3 | \n",
+ " 1.0 | \n",
+ " 0.0 | \n",
+ " 0.0 | \n",
+ " 0.0 | \n",
+ "
\n",
+ " \n",
+ " 4 | \n",
+ " 1.0 | \n",
+ " 0.0 | \n",
+ " 0.0 | \n",
+ " 0.0 | \n",
+ "
\n",
+ " \n",
+ "
\n",
+ "
"
+ ],
+ "text/plain": [
+ " 1.000000 0.000000 0.000000.1 0.000000.2\n",
+ "0 1.0 0.0 0.0 0.0\n",
+ "1 1.0 0.0 0.0 0.0\n",
+ "2 1.0 0.0 0.0 0.0\n",
+ "3 1.0 0.0 0.0 0.0\n",
+ "4 1.0 0.0 0.0 0.0"
+ ]
+ },
+ "execution_count": 24,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "label_train.head()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 25,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "img_category_train = []\n",
+ "img_category_val = []\n",
+ "\n",
+ "for index,row in label_train.iterrows():\n",
+ " category = 0\n",
+ " if (row[0] == 1):\n",
+ " category = 0 \n",
+ " elif (row[1] == 1):\n",
+ " category = 1 \n",
+ " elif (row[2] == 1):\n",
+ " category = 2 \n",
+ " else:\n",
+ " category = 3\n",
+ " \n",
+ " img_category_train.append(category)\n",
+ " \n",
+ "for index,row in label_val.iterrows():\n",
+ " category = 0\n",
+ " if (row[0] == 1):\n",
+ " category = 0 \n",
+ " elif (row[1] == 1):\n",
+ " category = 1 \n",
+ " elif (row[2] == 1):\n",
+ " category = 2 \n",
+ " else:\n",
+ " category = 3\n",
+ " \n",
+ " img_category_val.append(category)\n",
+ " "
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 26,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "(3199, 8192)\n",
+ "(3199, 4)\n",
+ "(799, 8192)\n",
+ "(799, 4)\n"
+ ]
+ }
+ ],
+ "source": [
+ "print(np.shape(img_train))\n",
+ "print(np.shape(label_train))\n",
+ "print(np.shape(img_val))\n",
+ "print(np.shape(label_val))"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 27,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "(3199, 8192)\n",
+ "(799, 8192)\n",
+ "(3199, 4)\n",
+ "(799, 4)\n",
+ "(3199, 64, 128, 1)\n",
+ "(799, 64, 128, 1)\n"
+ ]
+ }
+ ],
+ "source": [
+ "\n",
+ "img_train_data = img_train.to_numpy()\n",
+ "img_data_val = img_val.to_numpy()\n",
+ "label_train_data = label_train.to_numpy()\n",
+ "label_val_data = label_val.to_numpy()\n",
+ "print(img_train_data.shape)\n",
+ "print(img_data_val.shape)\n",
+ "print(label_train_data.shape)\n",
+ "print(label_val_data.shape)\n",
+ "raw_img_data_train = img_train_data.reshape(-1,64,128)\n",
+ "raw_img_data_val = img_data_val.reshape(-1,64,128)\n",
+ "\n",
+ "reshaped_img_data_train = np.reshape(raw_img_data_train,(-1, 64, 128, 1))\n",
+ "reshaped_img_data_val = np.reshape(raw_img_data_val,(-1, 64, 128,1))\n",
+ "\n",
+ "print(reshaped_img_data_train.shape)\n",
+ "print(reshaped_img_data_val.shape)\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 28,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "X_train, X_test, y_train, y_test = train_test_split(reshaped_img_data_train,label_train_data, test_size=0.2, random_state=69)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 29,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "2559 640 2559 640\n"
+ ]
+ }
+ ],
+ "source": [
+ "print(len(X_train) , len(X_test) , len(y_train) , len(y_test))"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 30,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "(2559, 64, 128, 1)"
+ ]
+ },
+ "execution_count": 30,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "X_train.shape\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 31,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "spectrograms = [ \"squiggle\", \"narrowband\", \"narrowbanddrd\", \"noise\"]"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 32,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# interpolation='nearest' simply displays an image without trying to interpolate between pixels if the display resolution is not the same as the image resolution"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Visualize the dataset"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 35,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "[1. 0. 0. 0.]\n"
+ ]
+ },
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ "