COVID-19-Prediction.py

# -*- coding: utf-8 -*-
"""COVID-19 Prediction.ipynb

Automatically generated by Colaboratory.

Original file is located at
    https://colab.research.google.com/drive/1oeA2Rp5B_8VWtOXEoY7R4dQMaO4pZe_y

# 1) IMPORTING LIBRARIES
"""

from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Conv2D, MaxPool2D,Dropout, Flatten, Dense
from tensorflow.keras.optimizers import Adam
from tensorflow.keras.preprocessing.image import ImageDataGenerator
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

"""# 2) CLONING DATASET"""

# CLONING THE DATASET FROM THE GITHUB REPOSITORY
!git clone https://github.com/RishitToteja/Chext-X-ray-Images-Data-Set.git

import os

main_dir = "/content/Chext-X-ray-Images-Data-Set/DataSet/Data"

# SETTING TRAIN AND TEST DIRECTORY
train_dir = os.path.join(main_dir, "train")
test_dir = os.path.join(main_dir, "test")

#SETING DIRECTORY FOR COVID AND NORMAL IMAGES DIRECTORY
train_covid_dir = os.path.join(train_dir, "COVID19")
train_normal_dir = os.path.join(train_dir, "NORMAL")

test_covid_dir = os.path.join(test_dir, "COVID19")
test_normal_dir = os.path.join(test_dir, "NORMAL")

# MAKING SEPERATE FILES : 
train_covid_names = os.listdir(train_covid_dir)
train_normal_names = os.listdir(train_normal_dir)

test_covid_names = os.listdir(test_covid_dir)
test_normal_names = os.listdir(test_normal_dir)

"""# 3) PERFORMING DATA VISUALIZATION


"""

import matplotlib.image as mpimg

rows = 4
columns = 4

fig = plt.gcf()
fig.set_size_inches(12,12)

covid_img = [os.path.join(train_covid_dir, filename) for filename in train_covid_names[0:8]]
normal_img = [os.path.join(train_normal_dir, filename) for filename in train_normal_names[0:8]]

print(covid_img)
print(normal_img)

merged_img = covid_img + normal_img

for i, img_path in enumerate(merged_img):
  title = img_path.split("/", 6)[6]
  plot = plt.subplot(rows, columns, i+1)
  plot.axis("Off")
  img = mpimg.imread(img_path)
  plot.set_title(title, fontsize = 11)
  plt.imshow(img, cmap= "gray")

plt.show()

"""# 4) DATA PREPROCESSING AND AUGMENTATION"""

# CREATING TRAINING, TESTING AND VALIDATION BATCHES

dgen_train = ImageDataGenerator(rescale = 1./255,
                                validation_split = 0.2,
                                zoom_range = 0.2,
                                horizontal_flip = True)

dgen_validation = ImageDataGenerator(rescale = 1./255,
                                     )

dgen_test = ImageDataGenerator(rescale = 1./255,
                              )

train_generator = dgen_train.flow_from_directory(train_dir,
                                                 target_size = (150, 150), 
                                                 subset = 'training',
                                                 batch_size = 32,
                                                 class_mode = 'binary')
validation_generator = dgen_train.flow_from_directory(train_dir,
                                                      target_size = (150, 150), 
                                                      subset = "validation", 
                                                      batch_size = 32, 
                                                      class_mode = "binary")
test_generator = dgen_test.flow_from_directory(test_dir,
                                               target_size = (150, 150), 
                                               batch_size = 32, 
                                               class_mode = "binary")

print("Class Labels are: ", train_generator.class_indices)
print("Image shape is : ", train_generator.image_shape)

"""# 5) BUILDING CONVOLUTIONAL NEURAL NETWORK MODEL"""

from tensorflow.keras.layers import Conv2D, MaxPooling2D

model = Sequential()

# 1) CONVOLUTIONAL LAYER - 1
model.add(Conv2D(32, (5,5), padding = "same", activation = "relu", input_shape = train_generator.image_shape))

# 2) POOLING LAYER - 1
model.add(MaxPooling2D(pool_size=(2,2)))

# 3) DROPOUT LAYER -2
model.add(Dropout(0.5))

# 4) CONVOLUTIONAL LAYER - 2
model.add(Conv2D(64, (5,5), padding = "same", activation = "relu"))

# 5) POOLING LAYER - 2
model.add(MaxPooling2D(pool_size=(2,2)))

# 6) DROPOUT LAYER - 2
model.add(Dropout(0.5))

# 7) FLATTENING LAYER TO 2D SHAPE
model.add(Flatten())

# 8) ADDING A DENSE LAYER
model.add(Dense(256, activation = 'relu'))

# 9 DROPOUT LAYER - 3
model.add(Dropout(0.5))

# 10) FINAL OUTPUT LAYER
model.add(Dense(1, activation = 'sigmoid'))

### PRINTING MODEL SUMMARY
model.summary()

"""# 6) COMPILING AND TRAINING THE NEURAL NETWORK MODEL"""

# COMPILING THE MODEL

model.compile(Adam(learning_rate = 0.001), loss = 'binary_crossentropy', metrics = ['accuracy'])

# TRAINING THE MODEL
history = model.fit(train_generator, 
                    epochs = 35, 
                    validation_data = validation_generator)

"""# 7) PERFORMING EVALUATION

"""

# KEYS OF HISTORY OBJECT
history.history.keys()

# PLOT GRAPH BETWEEN TRAINING AND VALIDATION LOSS
plt.plot(history.history['loss'])
plt.plot(history.history['val_loss'])
plt.legend(['Training', 'Validation'])
plt.title("Training and validation losses")
plt.xlabel('epoch')

# PLOT GRAPH BETWEEN TRAINING AND VALIDATION ACCURACY
plt.plot(history.history['accuracy'])
plt.plot(history.history['val_accuracy'])
plt.legend(['Training', 'Validation'])
plt.title("Training and validation accuracy")
plt.xlabel('epoch')

# GETTING TEST ACCURACY AND LOSS

test_loss, test_acc = model.evaluate(test_generator)
print("Test Set Loss : ", test_loss)
print("Test Set Accuracy : ", test_acc)

"""# 9) PREDICTION ON NEW DATA (UPLOAD FILES)"""

from google.colab import files
from keras.preprocessing import image

uploaded = files.upload()

for filename in uploaded.keys():
  img_path = '/content/' + filename
  img = image.load_img(img_path, target_size = (150,150))
  images = image.img_to_array(img)
  images = np.expand_dims(images, axis = 0)
  prediction = model.predict(images)
  
  if prediction == 0:
    print("The report is COVID-19 Positive")
  else:
    print("The report is COVID-19 Negative")

model.save("my_model.h5")