From 16d7dee576614f2e48cc06f176a89a21a5c2121c Mon Sep 17 00:00:00 2001 From: VARUNSHIYAM <138989960+Varunshiyam@users.noreply.github.com> Date: Tue, 29 Oct 2024 14:32:32 +0530 Subject: [PATCH] Fixes Issue #94 Exoplanet Detection Added up my Project file as ipynb --- .../Exoplanet_Detection.ipynb | 1384 +++++++++++++++++ 1 file changed, 1384 insertions(+) create mode 100644 Deep_Learning_Models/Exoplanet_Detection/Exoplanet_Detection.ipynb diff --git a/Deep_Learning_Models/Exoplanet_Detection/Exoplanet_Detection.ipynb b/Deep_Learning_Models/Exoplanet_Detection/Exoplanet_Detection.ipynb new file mode 100644 index 0000000..183c382 --- /dev/null +++ b/Deep_Learning_Models/Exoplanet_Detection/Exoplanet_Detection.ipynb @@ -0,0 +1,1384 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 48, + "metadata": { + "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19", + "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5", + "execution": { + "iopub.execute_input": "2022-06-04T13:50:21.971488Z", + "iopub.status.busy": "2022-06-04T13:50:21.970587Z", + "iopub.status.idle": "2022-06-04T13:50:21.982584Z", + "shell.execute_reply": "2022-06-04T13:50:21.981531Z", + "shell.execute_reply.started": "2022-06-04T13:50:21.971443Z" + } + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "/kaggle/input/exoplanetsflux/exoTest.csv\n", + "/kaggle/input/exoplanetsflux/exoTrain.csv\n" + ] + } + ], + "source": [ + "# This Python 3 environment comes with many helpful analytics libraries installed\n", + "# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n", + "# For example, here's several helpful packages to load\n", + "\n", + "#import numpy as np # linear algebra\n", + "#import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n", + "\n", + "# Input data files are available in the read-only \"../input/\" directory\n", + "# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n", + "\n", + "#import os\n", + "for dirname, _, filenames in os.walk('/kaggle/input'):\n", + " for filename in filenames:\n", + " print(os.path.join(dirname, filename))\n", + "\n", + "# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n", + "# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# **Transit Exoplanets Detection with Deep Learning Models**\n", + "\n", + "Scientists use data collected by space telescopes to find new information that allows us to learn more about the universe. The NASA Kepler Space Telescope has been collecting light from thousands of stars for many years to detect the presence of exoplanets.\n", + "\n", + "![ Transit Method ](https://www.science-et-vie.com/wp-content/uploads/scienceetvie/2022/04/transit-planetaire-750x319.jpg) ![]()\n", + "\n", + "An ExoPlanet is a planet that orbits a star, just like the Earth; however these systems are hundreds or thousands of light years away from Earth, so it is essential to have tools that can assist scientists in understanding whether a given star is likely to have exoplanets. The data collected by space telescopes is huge and new artificial intelligence techniques enable advanced data analysis and powerful predictive models.\n", + "\n", + "In this project we used a dataset of exoplanets, coming from Mikulski Archive, a large archive of astronomical data for classifying the light curve of the stars to check the presence of the exoplanets. First of all, I'm going to apply different feature engineering techniques techniques to the dataset ant then we will present a Convolution Neural Network (CNN), which is a strong model in Deep Learning for time series classification (TSC). Sine the measure of brightness is a standard in this application, this predictive model can be useful for future works with other and new larger dataset." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# **About Dataset**\n", + "\n", + "The data describe the change in flux (light intensity) of several thousand stars. Each Star has a binary label of 2 or 1. 2 indicated that the star is confirmed to have at least one exoplanet in orbit; some observations are in fact multi-planet systems.\n", + "\n", + "As you can imagine, planets themeselves do not emit light, but the Stars that they orbit do. If said Star is watched over several months or years, there may be a regular \"dimming\"ofthe flux (the light intensity). This is evidence that there may be an orbiting body around the Star; such a star could be considered to be a \"candidate\" system. Further study of our candidate system, for example by a satellite that captures light at a different wavelenght, could solidity the belief that the candidate can in fact be \"confirmed\". \n", + "\n", + "# **Description**\n", + "\n", + "TrainSet:\n", + "* 5087 rows or observations\n", + "* 3198 columns or features\n", + "* Column 1 is the label vector. Column 2 - 3198 are the flux values over time\n", + "* 37 confirmed exoplanet-stars and 5050 non-exoplanet-stars\n", + "\n", + "TestSet :\n", + "* 570 rows or observations\n", + "* 3198 columns or features\n", + "* Column 1 is the label vector. Column 2 - 3198 are the flux values over time\n", + "* 5 confirmed exoplanet-stars and 565 non-exoplanet-stars\n", + "\n", + "# **Acknowledgements**\n", + "\n", + "The data presented here are cleaned and are derived from observations made by the NASA Kepler Space Telescope. The Mission is ongoing - for instance data from Campaign 12 was released on 8th Marth 2017. Over 99% of this dataset originates from Campaign 3. To boost the number of exoplanet-stars in the dataset, confirmed exoplanets from other campaigns were also included.\n", + "\n", + "To be clear, all observations from Campaign 3 are included. And in addition to this, confirmed exoplanet-stars from other campaigns are also included.\n", + "\n", + "The datasets were prepared late-summer 2016. \n", + "\n", + "Campaign 3 was used because \"it was felt\" that this Campaign is unlikely to contain any undiscovered (i.e. wrongly labelled) exoplanets.\n", + "\n", + "NASA open-sources the original Kepler Mission data and it is hosted at the Mikulsi Archive. After being beamed down to Earth, NASA applies de-noising algorithms to remove artefacts generated by the telescope. The data - in the astrophysicist, anyone with an ineternet connection can embark on a search to find and retrieve the datafiles from the Archive.\n", + "\n", + "The Transit Method Descriptive image is copyright @ 2021 by [Science & Vie](https://www.science-et-vie.com)" + ] + }, + { + "cell_type": "code", + "execution_count": 76, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T14:42:19.567963Z", + "iopub.status.busy": "2022-06-04T14:42:19.567585Z", + "iopub.status.idle": "2022-06-04T14:42:19.576724Z", + "shell.execute_reply": "2022-06-04T14:42:19.575967Z", + "shell.execute_reply.started": "2022-06-04T14:42:19.567933Z" + } + }, + "outputs": [], + "source": [ + "# Import data analysis Packages\n", + "import numpy as np # linear algebra\n", + "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n", + "import matplotlib.pyplot as plt # data visualization\n", + "from scipy import signal #\n", + "from scipy.ndimage.filters import gaussian_filter\n", + "from scipy.fftpack import fft\n", + "import scipy\n", + "import seaborn as sns # data visualization\n", + "# import models as m # model creation package\n", + "\n", + "# Import Machine Learning and Deep Learninig packages\n", + "#import sklearn.linear_model as lm\n", + "#import tensorflow as tf\n", + "#import sklearn.svm as svm\n", + "\n", + "#from tensorflow.keras import models\n", + "#from tensorflow.keras import layers\n", + "#from tensorflow.keras.preprocessing import sequence\n", + "\n", + "# Model evaluation methods\n", + "import sklearn.preprocessing as pproc\n", + "from sklearn.model_selection import train_test_split\n", + "#from sklearn.metrics import accuracy_score\n", + "#from sklearn.metrics import confusion_matrix, classification_report\n", + "from sklearn.metrics import plot_confusion_matrix\n", + "from sklearn.preprocessing import normalize\n", + "\n", + "from imblearn.over_sampling import RandomOverSampler" + ] + }, + { + "cell_type": "code", + "execution_count": 50, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T13:50:36.015223Z", + "iopub.status.busy": "2022-06-04T13:50:36.014661Z", + "iopub.status.idle": "2022-06-04T13:50:36.019548Z", + "shell.execute_reply": "2022-06-04T13:50:36.018671Z", + "shell.execute_reply.started": "2022-06-04T13:50:36.015190Z" + } + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "2.6.4\n" + ] + } + ], + "source": [ + "# Display Tensorflow version\n", + "print(tf.__version__)" + ] + }, + { + "cell_type": "code", + "execution_count": 51, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T13:50:41.858434Z", + "iopub.status.busy": "2022-06-04T13:50:41.857997Z", + "iopub.status.idle": "2022-06-04T13:50:48.350833Z", + "shell.execute_reply": "2022-06-04T13:50:48.349979Z", + "shell.execute_reply.started": "2022-06-04T13:50:41.858397Z" + } + }, + "outputs": [], + "source": [ + "# Import the dataset\n", + "data_train = pd.read_csv('/kaggle/input/exoplanetsflux/exoTrain.csv') # TrainSet\n", + "data_test = pd.read_csv('/kaggle/input/exoplanetsflux/exoTest.csv')" + ] + }, + { + "cell_type": "code", + "execution_count": 53, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T13:51:13.237512Z", + "iopub.status.busy": "2022-06-04T13:51:13.236761Z", + "iopub.status.idle": "2022-06-04T13:51:13.276734Z", + "shell.execute_reply": "2022-06-04T13:51:13.275987Z", + "shell.execute_reply.started": "2022-06-04T13:51:13.237474Z" + } + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
LABELFLUX.1FLUX.2FLUX.3FLUX.4FLUX.5FLUX.6FLUX.7FLUX.8FLUX.9...FLUX.3188FLUX.3189FLUX.3190FLUX.3191FLUX.3192FLUX.3193FLUX.3194FLUX.3195FLUX.3196FLUX.3197
0293.8583.8120.10-26.98-39.56-124.71-135.18-96.27-79.89...-78.07-102.15-102.1525.1348.5792.5439.3261.425.08-39.54
12-38.88-33.83-58.54-40.09-79.31-72.81-86.55-85.33-83.97...-3.28-32.21-32.21-24.89-4.860.76-11.706.4616.0019.93
22532.64535.92513.73496.92456.45466.00464.50486.39436.56...-71.6913.3113.31-29.89-20.885.06-11.80-28.91-70.02-96.67
32326.52347.39302.35298.13317.74312.70322.33311.31312.42...5.71-3.73-3.7330.0520.03-12.67-8.77-17.31-17.3513.98
42-1107.21-1112.59-1118.95-1095.10-1057.55-1034.48-998.34-1022.71-989.57...-594.37-401.66-401.66-357.24-443.76-438.54-399.71-384.65-411.79-510.54
..................................................................
50771125.5778.6998.2991.1678.4245.8261.6922.7339.09...32.3563.2357.9890.43115.12210.093.8016.3327.3521.30
507817.4510.026.87-2.82-1.56-4.30-7.01-6.97-2.54...-5.25-8.560.53-4.29-6.608.75-10.69-9.54-2.48-8.69
50791475.61395.50423.61376.36338.94321.26326.34342.84251.23...543.25453.87344.35266.16242.18163.0286.2913.06161.22213.60
50801-46.63-55.39-64.88-88.75-75.40-64.06-66.37-41.95-68.07...29.646.9032.9456.6328.7128.82-20.12-14.41-43.35-30.04
50811299.41302.77278.68263.48236.89186.93145.45151.20123.38...-126.36-133.82-134.02-98.76-106.60-74.95-46.29-3.08-28.43-48.68
\n", + "

5082 rows × 3198 columns

\n", + "
" + ], + "text/plain": [ + " LABEL FLUX.1 FLUX.2 FLUX.3 FLUX.4 FLUX.5 FLUX.6 FLUX.7 \\\n", + "0 2 93.85 83.81 20.10 -26.98 -39.56 -124.71 -135.18 \n", + "1 2 -38.88 -33.83 -58.54 -40.09 -79.31 -72.81 -86.55 \n", + "2 2 532.64 535.92 513.73 496.92 456.45 466.00 464.50 \n", + "3 2 326.52 347.39 302.35 298.13 317.74 312.70 322.33 \n", + "4 2 -1107.21 -1112.59 -1118.95 -1095.10 -1057.55 -1034.48 -998.34 \n", + "... ... ... ... ... ... ... ... ... \n", + "5077 1 125.57 78.69 98.29 91.16 78.42 45.82 61.69 \n", + "5078 1 7.45 10.02 6.87 -2.82 -1.56 -4.30 -7.01 \n", + "5079 1 475.61 395.50 423.61 376.36 338.94 321.26 326.34 \n", + "5080 1 -46.63 -55.39 -64.88 -88.75 -75.40 -64.06 -66.37 \n", + "5081 1 299.41 302.77 278.68 263.48 236.89 186.93 145.45 \n", + "\n", + " FLUX.8 FLUX.9 ... FLUX.3188 FLUX.3189 FLUX.3190 FLUX.3191 \\\n", + "0 -96.27 -79.89 ... -78.07 -102.15 -102.15 25.13 \n", + "1 -85.33 -83.97 ... -3.28 -32.21 -32.21 -24.89 \n", + "2 486.39 436.56 ... -71.69 13.31 13.31 -29.89 \n", + "3 311.31 312.42 ... 5.71 -3.73 -3.73 30.05 \n", + "4 -1022.71 -989.57 ... -594.37 -401.66 -401.66 -357.24 \n", + "... ... ... ... ... ... ... ... \n", + "5077 22.73 39.09 ... 32.35 63.23 57.98 90.43 \n", + "5078 -6.97 -2.54 ... -5.25 -8.56 0.53 -4.29 \n", + "5079 342.84 251.23 ... 543.25 453.87 344.35 266.16 \n", + "5080 -41.95 -68.07 ... 29.64 6.90 32.94 56.63 \n", + "5081 151.20 123.38 ... -126.36 -133.82 -134.02 -98.76 \n", + "\n", + " FLUX.3192 FLUX.3193 FLUX.3194 FLUX.3195 FLUX.3196 FLUX.3197 \n", + "0 48.57 92.54 39.32 61.42 5.08 -39.54 \n", + "1 -4.86 0.76 -11.70 6.46 16.00 19.93 \n", + "2 -20.88 5.06 -11.80 -28.91 -70.02 -96.67 \n", + "3 20.03 -12.67 -8.77 -17.31 -17.35 13.98 \n", + "4 -443.76 -438.54 -399.71 -384.65 -411.79 -510.54 \n", + "... ... ... ... ... ... ... \n", + "5077 115.12 210.09 3.80 16.33 27.35 21.30 \n", + "5078 -6.60 8.75 -10.69 -9.54 -2.48 -8.69 \n", + "5079 242.18 163.02 86.29 13.06 161.22 213.60 \n", + "5080 28.71 28.82 -20.12 -14.41 -43.35 -30.04 \n", + "5081 -106.60 -74.95 -46.29 -3.08 -28.43 -48.68 \n", + "\n", + "[5082 rows x 3198 columns]" + ] + }, + "execution_count": 53, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# Data Analysis\n", + "data_train.head(-5)" + ] + }, + { + "cell_type": "code", + "execution_count": 54, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T13:51:17.802311Z", + "iopub.status.busy": "2022-06-04T13:51:17.801967Z", + "iopub.status.idle": "2022-06-04T13:51:18.041207Z", + "shell.execute_reply": "2022-06-04T13:51:18.040369Z", + "shell.execute_reply.started": "2022-06-04T13:51:17.802283Z" + } + }, + "outputs": [], + "source": [ + "# Permute the dataset\n", + "data_train = np.random.permutation(np.asarray(data_train))\n", + "data_test = np.random.permutation(np.asarray(data_test))" + ] + }, + { + "cell_type": "code", + "execution_count": 55, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T13:51:20.572931Z", + "iopub.status.busy": "2022-06-04T13:51:20.572393Z", + "iopub.status.idle": "2022-06-04T13:51:20.636874Z", + "shell.execute_reply": "2022-06-04T13:51:20.635991Z", + "shell.execute_reply.started": "2022-06-04T13:51:20.572890Z" + } + }, + "outputs": [], + "source": [ + "# Get the label column and delete the class column and rescale\n", + "y1 = data_train[:,0]\n", + "y2 = data_test[:,0]\n", + "\n", + "y_train = (y1 - min(y1)) / (max(y1) - min(y1))\n", + "y_test = (y2- min(y2)) / (max(y2) - min(y2))\n", + "\n", + "data_train = np.delete(data_train,1,1)\n", + "data_test = np.delete(data_test,1,1)" + ] + }, + { + "cell_type": "code", + "execution_count": 56, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T13:51:26.148675Z", + "iopub.status.busy": "2022-06-04T13:51:26.148293Z", + "iopub.status.idle": "2022-06-04T13:51:26.360230Z", + "shell.execute_reply": "2022-06-04T13:51:26.359504Z", + "shell.execute_reply.started": "2022-06-04T13:51:26.148619Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "[]" + ] + }, + "execution_count": 56, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "# Print the light curve\n", + "time = np.arange(len(data_train[0])) * (36/60) # time in hours\n", + "\n", + "plt.figure(figsize=(20,5))\n", + "plt.title('Flux of Star 10 with confirmed planet')\n", + "plt.ylabel('Flux')\n", + "plt.xlabel('Hours')\n", + "plt.plot(time, data_train[10]) # Change the number to plot what you want" + ] + }, + { + "cell_type": "code", + "execution_count": 57, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T13:51:46.850494Z", + "iopub.status.busy": "2022-06-04T13:51:46.849671Z", + "iopub.status.idle": "2022-06-04T13:51:46.955455Z", + "shell.execute_reply": "2022-06-04T13:51:46.954600Z", + "shell.execute_reply.started": "2022-06-04T13:51:46.850458Z" + } + }, + "outputs": [], + "source": [ + "# Normalized data\n", + "data_train_norm = normalize(data_train)\n", + "data_test_norm = normalize(data_test)" + ] + }, + { + "cell_type": "code", + "execution_count": 58, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T13:51:49.728777Z", + "iopub.status.busy": "2022-06-04T13:51:49.727835Z", + "iopub.status.idle": "2022-06-04T13:51:49.734208Z", + "shell.execute_reply": "2022-06-04T13:51:49.733318Z", + "shell.execute_reply.started": "2022-06-04T13:51:49.728729Z" + } + }, + "outputs": [], + "source": [ + "# Function to apply gaussian filter to all data\n", + "def gauss_filter(dataset, sigma):\n", + " dts = []\n", + " for x in range(dataset.shape[0]):\n", + " dts.append(gaussian_filter(dataset[x], sigma))\n", + " \n", + " return np.asarray(dts)" + ] + }, + { + "cell_type": "code", + "execution_count": 59, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T13:52:02.258151Z", + "iopub.status.busy": "2022-06-04T13:52:02.257426Z", + "iopub.status.idle": "2022-06-04T13:52:03.057856Z", + "shell.execute_reply": "2022-06-04T13:52:03.056885Z", + "shell.execute_reply.started": "2022-06-04T13:52:02.258113Z" + } + }, + "outputs": [], + "source": [ + "# Apply the gaussian filter to all rows data\n", + "data_train_gaussian = gauss_filter(data_train_norm,7.0)\n", + "data_test_gaussian = gauss_filter(data_test_norm,7.0)" + ] + }, + { + "cell_type": "code", + "execution_count": 63, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T13:56:52.431648Z", + "iopub.status.busy": "2022-06-04T13:56:52.431269Z", + "iopub.status.idle": "2022-06-04T13:56:52.635192Z", + "shell.execute_reply": "2022-06-04T13:56:52.634466Z", + "shell.execute_reply.started": "2022-06-04T13:56:52.431598Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "[]" + ] + }, + "execution_count": 63, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "# Print the light curves smoothed\n", + "plt.figure(figsize=(20,5))\n", + "plt.title('Flux of Start 10 with confirmed planet, smoothed')\n", + "plt.ylabel('Flux')\n", + "plt.xlabel(\"Hours\")\n", + "plt.plot(time, data_train_gaussian[10])" + ] + }, + { + "cell_type": "code", + "execution_count": 65, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T14:05:24.177695Z", + "iopub.status.busy": "2022-06-04T14:05:24.177259Z", + "iopub.status.idle": "2022-06-04T14:05:24.736718Z", + "shell.execute_reply": "2022-06-04T14:05:24.735825Z", + "shell.execute_reply.started": "2022-06-04T14:05:24.177654Z" + } + }, + "outputs": [], + "source": [ + "# Apply FFT to the data smoothed\n", + "frequency = np.arange(len(data_train[0])) * (1/(36.0*60.0))\n", + "\n", + "data_train_fft1 = scipy.fft.fft2(data_train_norm, axes=1)\n", + "data_test_fft1 = scipy.fft.fft2(data_test_norm, axes=1)\n", + "\n", + "data_train_fft = np.abs(data_train_fft1) # calculate the abs value\n", + "data_test_fft = np.abs(data_test_fft1)" + ] + }, + { + "cell_type": "code", + "execution_count": 66, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T14:07:06.303680Z", + "iopub.status.busy": "2022-06-04T14:07:06.303282Z", + "iopub.status.idle": "2022-06-04T14:07:06.308097Z", + "shell.execute_reply": "2022-06-04T14:07:06.307007Z", + "shell.execute_reply.started": "2022-06-04T14:07:06.303642Z" + } + }, + "outputs": [], + "source": [ + "# Get the lenght of the FFT data, make something here below in order to make the sequences of the same size\n", + "# only if they have differet dimensions\n", + "len_seq = len(data_train_fft[0])" + ] + }, + { + "cell_type": "code", + "execution_count": 67, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T14:09:27.527084Z", + "iopub.status.busy": "2022-06-04T14:09:27.526704Z", + "iopub.status.idle": "2022-06-04T14:09:27.743746Z", + "shell.execute_reply": "2022-06-04T14:09:27.743038Z", + "shell.execute_reply.started": "2022-06-04T14:09:27.527053Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "[]" + ] + }, + "execution_count": 67, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "# Plot the FFT of the signals\n", + "plt.figure(figsize=(20,5))\n", + "plt.title('Flux of Star 1 (with confirmed planet) in domain of frequencies')\n", + "plt.ylabel('Abs value of FFT result')\n", + "plt.xlabel('Frequency')\n", + "plt.plot(frequency, data_train_fft[1])" + ] + }, + { + "cell_type": "code", + "execution_count": 72, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T14:28:13.243479Z", + "iopub.status.busy": "2022-06-04T14:28:13.243133Z", + "iopub.status.idle": "2022-06-04T14:28:13.386604Z", + "shell.execute_reply": "2022-06-04T14:28:13.385684Z", + "shell.execute_reply.started": "2022-06-04T14:28:13.243451Z" + } + }, + "outputs": [], + "source": [ + "# Oversampling technique to the data\n", + "rm = RandomOverSampler(sampling_strategy=0.5)\n", + "data_train_ovs, y_train_ovs = rm.fit_resample(data_train_fft, y_train)" + ] + }, + { + "cell_type": "code", + "execution_count": 73, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T14:30:32.958216Z", + "iopub.status.busy": "2022-06-04T14:30:32.957371Z", + "iopub.status.idle": "2022-06-04T14:30:33.011389Z", + "shell.execute_reply": "2022-06-04T14:30:33.009905Z", + "shell.execute_reply.started": "2022-06-04T14:30:32.958177Z" + } + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "After oversampling, counts of label '1': 2525\n", + "After oversampling, counts of label '0': 5050\n" + ] + } + ], + "source": [ + "# Recap dataset after ovesampling\n", + "print(\"After oversampling, counts of label '1': {}\".format(sum(y_train_ovs==1)))\n", + "print(\"After oversampling, counts of label '0': {}\".format(sum(y_train_ovs==0)))" + ] + }, + { + "cell_type": "code", + "execution_count": 74, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T14:35:45.197387Z", + "iopub.status.busy": "2022-06-04T14:35:45.196801Z", + "iopub.status.idle": "2022-06-04T14:35:45.201620Z", + "shell.execute_reply": "2022-06-04T14:35:45.200752Z", + "shell.execute_reply.started": "2022-06-04T14:35:45.197350Z" + } + }, + "outputs": [], + "source": [ + "# Reshape the data for the neural network model\n", + "data_train_ovs = np.asarray(data_train_ovs)\n", + "data_test_fft = np.asarray(data_test_fft)" + ] + }, + { + "cell_type": "code", + "execution_count": 75, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T14:36:04.601441Z", + "iopub.status.busy": "2022-06-04T14:36:04.600833Z", + "iopub.status.idle": "2022-06-04T14:36:04.606586Z", + "shell.execute_reply": "2022-06-04T14:36:04.605535Z", + "shell.execute_reply.started": "2022-06-04T14:36:04.601400Z" + } + }, + "outputs": [], + "source": [ + "data_train_ovs_nn = data_train_ovs.reshape((data_train_ovs.shape[0], data_train_ovs.shape[1], 1))\n", + "data_test_fft_nn = data_test_fft.reshape((data_test_fft.shape[0], data_test_fft.shape[1], 1))" + ] + }, + { + "cell_type": "code", + "execution_count": 77, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T14:44:04.035858Z", + "iopub.status.busy": "2022-06-04T14:44:04.035175Z", + "iopub.status.idle": "2022-06-04T14:44:04.048805Z", + "shell.execute_reply": "2022-06-04T14:44:04.047923Z", + "shell.execute_reply.started": "2022-06-04T14:44:04.035817Z" + } + }, + "outputs": [], + "source": [ + "# Import all modules for models creation\n", + "import tensorflow as tf\n", + "import pydot\n", + "\n", + "from sklearn.metrics import accuracy_score\n", + "from sklearn.metrics import confusion_matrix, classification_report\n", + "from keras.utils.vis_utils import plot_model\n", + "\n", + "from tensorflow.keras import models\n", + "from tensorflow.keras import layers\n", + "from tensorflow.keras.preprocessing import sequence\n", + "\n", + "\n", + "from sklearn.model_selection import GridSearchCV\n", + "from sklearn.svm import SVC\n", + "\n", + "\n", + "#create the neural network\n", + "def FCN_model(len_seq):\n", + " \n", + " # len_seq = the size of the input sequences\n", + " \n", + " model = tf.keras.Sequential()\n", + " \n", + " #change the input shape if you have sequences less long\n", + " model.add(layers.Conv1D(filters=256, kernel_size=8, activation='relu', input_shape=(len_seq,1)))\n", + " model.add(layers.MaxPool1D(strides=5))\n", + " model.add(layers.BatchNormalization())\n", + " \n", + " \n", + " model.add(layers.Conv1D(filters=340, kernel_size=6, activation='relu'))\n", + " model.add(layers.MaxPool1D(strides=5))\n", + " model.add(layers.BatchNormalization())\n", + " \n", + " \n", + " model.add(layers.Conv1D(filters=256, kernel_size=4, activation='relu'))\n", + " model.add(layers.MaxPool1D(strides=5))\n", + " model.add(layers.BatchNormalization())\n", + " \n", + " \n", + " model.add(layers.Flatten())\n", + " model.add(layers.Dropout(0.3))\n", + " \n", + " \n", + " model.add(layers.Dense(24, activation='relu'))\n", + " model.add(layers.Dropout(0.3))\n", + " \n", + " model.add(layers.Dense(12, activation='relu'))\n", + " \n", + " \n", + " model.add(layers.Dense(8, activation = 'relu'))\n", + " \n", + " \n", + " model.add(layers.Dense(1, activation='sigmoid'))\n", + " \n", + " return model\n", + "\n", + "\n", + "\n", + "#create the SVC model\n", + "def SVC_model():\n", + " \n", + " tuned_parameters = [{'kernel': ['rbf'], 'gamma': [1e-3, 1e-4],\n", + " 'C': [1, 10, 100, 1000]},\n", + " {'kernel': ['linear'], 'C': [1, 10, 100, 1000]}]\n", + "\n", + " \n", + " clf = GridSearchCV( SVC(), param_grid = tuned_parameters,scoring = 'recall')\n", + " \n", + " \n", + " return clf" + ] + }, + { + "cell_type": "code", + "execution_count": 78, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T14:45:30.300906Z", + "iopub.status.busy": "2022-06-04T14:45:30.300132Z", + "iopub.status.idle": "2022-06-04T14:47:56.747937Z", + "shell.execute_reply": "2022-06-04T14:47:56.746964Z", + "shell.execute_reply.started": "2022-06-04T14:45:30.300855Z" + } + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2022-06-04 14:45:30.395920: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:937] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n", + "2022-06-04 14:45:30.539557: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:937] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n", + "2022-06-04 14:45:30.540308: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:937] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n", + "2022-06-04 14:45:30.541502: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 AVX512F FMA\n", + "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n", + "2022-06-04 14:45:30.541824: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:937] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n", + "2022-06-04 14:45:30.542534: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:937] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n", + "2022-06-04 14:45:30.543200: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:937] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n", + "2022-06-04 14:45:32.761058: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:937] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n", + "2022-06-04 14:45:32.762439: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:937] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n", + "2022-06-04 14:45:32.763487: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:937] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n", + "2022-06-04 14:45:32.764420: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1510] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 15403 MB memory: -> device: 0, name: Tesla P100-PCIE-16GB, pci bus id: 0000:00:04.0, compute capability: 6.0\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Model: \"sequential\"\n", + "_________________________________________________________________\n", + "Layer (type) Output Shape Param # \n", + "=================================================================\n", + "conv1d (Conv1D) (None, 3190, 256) 2304 \n", + "_________________________________________________________________\n", + "max_pooling1d (MaxPooling1D) (None, 638, 256) 0 \n", + "_________________________________________________________________\n", + "batch_normalization (BatchNo (None, 638, 256) 1024 \n", + "_________________________________________________________________\n", + "conv1d_1 (Conv1D) (None, 633, 340) 522580 \n", + "_________________________________________________________________\n", + "max_pooling1d_1 (MaxPooling1 (None, 127, 340) 0 \n", + "_________________________________________________________________\n", + "batch_normalization_1 (Batch (None, 127, 340) 1360 \n", + "_________________________________________________________________\n", + "conv1d_2 (Conv1D) (None, 124, 256) 348416 \n", + "_________________________________________________________________\n", + "max_pooling1d_2 (MaxPooling1 (None, 25, 256) 0 \n", + "_________________________________________________________________\n", + "batch_normalization_2 (Batch (None, 25, 256) 1024 \n", + "_________________________________________________________________\n", + "flatten (Flatten) (None, 6400) 0 \n", + "_________________________________________________________________\n", + "dropout (Dropout) (None, 6400) 0 \n", + "_________________________________________________________________\n", + "dense (Dense) (None, 24) 153624 \n", + "_________________________________________________________________\n", + "dropout_1 (Dropout) (None, 24) 0 \n", + "_________________________________________________________________\n", + "dense_1 (Dense) (None, 12) 300 \n", + "_________________________________________________________________\n", + "dense_2 (Dense) (None, 8) 104 \n", + "_________________________________________________________________\n", + "dense_3 (Dense) (None, 1) 9 \n", + "=================================================================\n", + "Total params: 1,030,745\n", + "Trainable params: 1,029,041\n", + "Non-trainable params: 1,704\n", + "_________________________________________________________________\n", + "None\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2022-06-04 14:45:33.704953: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:185] None of the MLIR Optimization Passes are enabled (registered 2)\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Epoch 1/15\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2022-06-04 14:45:35.725775: I tensorflow/stream_executor/cuda/cuda_dnn.cc:369] Loaded cuDNN version 8005\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "758/758 [==============================] - 17s 12ms/step - loss: 0.1499 - accuracy: 0.9345 - val_loss: 0.0324 - val_accuracy: 0.9912\n", + "Epoch 2/15\n", + "758/758 [==============================] - 9s 12ms/step - loss: 0.0259 - accuracy: 0.9919 - val_loss: 0.0235 - val_accuracy: 0.9930\n", + "Epoch 3/15\n", + "758/758 [==============================] - 9s 12ms/step - loss: 0.0189 - accuracy: 0.9942 - val_loss: 0.0288 - val_accuracy: 0.9895\n", + "Epoch 4/15\n", + "758/758 [==============================] - 9s 12ms/step - loss: 0.0128 - accuracy: 0.9963 - val_loss: 0.0199 - val_accuracy: 0.9912\n", + "Epoch 5/15\n", + "758/758 [==============================] - 9s 11ms/step - loss: 0.0127 - accuracy: 0.9975 - val_loss: 0.0154 - val_accuracy: 0.9965\n", + "Epoch 6/15\n", + "758/758 [==============================] - 8s 11ms/step - loss: 0.0098 - accuracy: 0.9974 - val_loss: 0.0048 - val_accuracy: 0.9965\n", + "Epoch 7/15\n", + "758/758 [==============================] - 9s 12ms/step - loss: 0.0165 - accuracy: 0.9933 - val_loss: 0.0175 - val_accuracy: 0.9947\n", + "Epoch 8/15\n", + "758/758 [==============================] - 9s 11ms/step - loss: 0.0058 - accuracy: 0.9988 - val_loss: 0.0026 - val_accuracy: 1.0000\n", + "Epoch 9/15\n", + "758/758 [==============================] - 9s 11ms/step - loss: 0.0123 - accuracy: 0.9962 - val_loss: 0.0254 - val_accuracy: 0.9947\n", + "Epoch 10/15\n", + "758/758 [==============================] - 9s 11ms/step - loss: 0.0043 - accuracy: 0.9989 - val_loss: 0.0660 - val_accuracy: 0.9895\n", + "Epoch 11/15\n", + "758/758 [==============================] - 9s 12ms/step - loss: 0.0097 - accuracy: 0.9976 - val_loss: 0.0532 - val_accuracy: 0.9895\n", + "Epoch 12/15\n", + "758/758 [==============================] - 9s 11ms/step - loss: 0.0039 - accuracy: 0.9991 - val_loss: 0.0185 - val_accuracy: 0.9947\n", + "Epoch 13/15\n", + "758/758 [==============================] - 9s 11ms/step - loss: 0.0017 - accuracy: 0.9995 - val_loss: 0.0302 - val_accuracy: 0.9965\n", + "Epoch 14/15\n", + "758/758 [==============================] - 9s 12ms/step - loss: 0.0037 - accuracy: 0.9997 - val_loss: 0.0878 - val_accuracy: 0.9895\n", + "Epoch 15/15\n", + "758/758 [==============================] - 9s 12ms/step - loss: 0.0074 - accuracy: 0.9980 - val_loss: 0.0390 - val_accuracy: 0.9965\n" + ] + } + ], + "source": [ + "# Create F.C.N model and run it\n", + "model = FCN_model(len_seq)\n", + "\n", + "model.compile(loss='binary_crossentropy', optimizer=tf.keras.optimizers.Adam(learning_rate=0.001),metrics=['accuracy'])\n", + "\n", + "print(model.summary())\n", + "\n", + "history = model.fit(data_train_ovs_nn, y_train_ovs , epochs=15, batch_size = 10, validation_data=(data_test_fft_nn, y_test))" + ] + }, + { + "cell_type": "code", + "execution_count": 81, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T15:01:01.003365Z", + "iopub.status.busy": "2022-06-04T15:01:01.002944Z", + "iopub.status.idle": "2022-06-04T15:01:01.007603Z", + "shell.execute_reply": "2022-06-04T15:01:01.006506Z", + "shell.execute_reply.started": "2022-06-04T15:01:01.003330Z" + } + }, + "outputs": [], + "source": [ + "# Save the model\n", + "# model.save(\"/kaggle/working/exoplanetflux_model_1\")\n", + "# Save -format using the HDF5 standard. \n", + "#model.save(\"/kaggle/working/exoplanetflux_model_1.h5\")\n", + "#load the model if already exsist\n", + "#model = tf.keras.models.load_model(\"/kaggle/working/exoplanetflux_model_1.h5\")" + ] + }, + { + "cell_type": "code", + "execution_count": 84, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T15:03:08.815736Z", + "iopub.status.busy": "2022-06-04T15:03:08.815349Z", + "iopub.status.idle": "2022-06-04T15:03:09.020271Z", + "shell.execute_reply": "2022-06-04T15:03:09.019523Z", + "shell.execute_reply.started": "2022-06-04T15:03:08.815705Z" + } + }, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "# Plot Accuracy\n", + "acc = history.history['accuracy']\n", + "#acc_val = history.history['val_accuracy']\n", + "epochs = range(1, len(acc)+1)\n", + "plt.plot(epochs, acc, 'b', label='accuracy_train')\n", + "#plt.plot(epochs, acc_val, 'g', label='accuracy_val')\n", + "plt.title('accuracy')\n", + "plt.xlabel('epochs')\n", + "plt.ylabel('value of accuracy')\n", + "plt.legend()\n", + "plt.grid()\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 85, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T15:03:55.244560Z", + "iopub.status.busy": "2022-06-04T15:03:55.244201Z", + "iopub.status.idle": "2022-06-04T15:03:55.437292Z", + "shell.execute_reply": "2022-06-04T15:03:55.436578Z", + "shell.execute_reply.started": "2022-06-04T15:03:55.244527Z" + } + }, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "# Plot Loss\n", + "loss = history.history['loss']\n", + "#loss_val = history.history['val_loss']\n", + "epochs = range(1, len(acc)+1)\n", + "plt.plot(epochs, loss, 'b', label='loss_train')\n", + "#plt.plot(epochs, loss_val, 'g', label='loss_val')\n", + "plt.title('loss')\n", + "plt.xlabel('epochs')\n", + "plt.ylabel('value of loss')\n", + "plt.legend()\n", + "plt.grid()\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 86, + "metadata": { + "execution": { + "iopub.execute_input": "2022-06-04T15:05:44.197762Z", + "iopub.status.busy": "2022-06-04T15:05:44.196895Z", + "iopub.status.idle": "2022-06-04T15:05:44.768248Z", + "shell.execute_reply": "2022-06-04T15:05:44.767490Z", + "shell.execute_reply.started": "2022-06-04T15:05:44.197727Z" + } + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "accuracy : 0.9964912280701754\n", + " precision recall f1-score support\n", + "\n", + " NO exoplanet confirmed 1.00 1.00 1.00 565\n", + "YES exoplanet confirmed 1.00 0.60 0.75 5\n", + "\n", + " accuracy 1.00 570\n", + " macro avg 1.00 0.80 0.87 570\n", + " weighted avg 1.00 1.00 1.00 570\n", + "\n" + ] + }, + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 86, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "#predict the test set and plot results\n", + "y_test_pred = model.predict(data_test_fft_nn)\n", + "y_test_pred = (y_test_pred > 0.5)\n", + "\n", + "\n", + "accuracy = accuracy_score(y_test, y_test_pred)\n", + "print(\"accuracy : \", accuracy)\n", + "\n", + "print(classification_report(y_test, y_test_pred, target_names=[\"NO exoplanet confirmed\",\"YES exoplanet confirmed\"]))\n", + "\n", + "conf_matrix = confusion_matrix([int(x) for x in y_test ], [int(y) for y in y_test_pred ])\n", + "sns.heatmap(conf_matrix, annot=True, cmap='Blues')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Result \n", + "The CNN has achieved good results. With all the problems written above with a sufficient number of epochs (according to our tests they must be greater than 15) the model is able to detect 5 exoplanet in the test set ( 100 % accuracy ).\n", + "\n", + "# Future work\n", + "In the field of applied astrophysics the transit detection of exoplanets is an important technique for the discovery of new planets. It would be very interesting to test our models ( in particular CNN ) in new larger datasets. Recall that in the MAST archive ( Mikulski Archive for Space Telescopes ) it is possible to obtain the light curves of thousands of other stars collected by the kepler telescope. However these data are not in a form directly usable by a machine learning model and therefore would require to be downloaded and processed to extract the light curves. In this project due to a limited time it was not possible to test the models with new datasets. In case someone would like to do this the code has been constructed to be able to change a number of reduced variables to adapt the models to new datasets." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# **THANKS YOU**" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.11.7" + } + }, + "nbformat": 4, + "nbformat_minor": 4 +}