From 629fd8aac17c23314521672263e6605fe5b30151 Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Thu, 22 Apr 2021 20:13:05 +0530
Subject: [PATCH 01/32] cnn classification file

---
 machine_learning/cnn_classification.py | 99 ++++++++++++++++++++++++++
 1 file changed, 99 insertions(+)
 create mode 100644 machine_learning/cnn_classification.py

diff --git a/machine_learning/cnn_classification.py b/machine_learning/cnn_classification.py
new file mode 100644
index 000000000000..f5edc137d1e0
--- /dev/null
+++ b/machine_learning/cnn_classification.py
@@ -0,0 +1,99 @@
+# Convolutional Neural Network
+
+ # Installing Tensorflow
+ # pip install tensorflow
+
+ # Installing Keras
+ # pip install --upgrade keras
+
+ # Part 1 - Building the CNN
+
+ # Importing the Keras libraries and packages
+ from keras.models import Sequential
+ from keras.layers import Conv2D
+ from keras.layers import MaxPooling2D
+ from keras.layers import Flatten
+ from keras.layers import Dense
+
+ # Initialising the CNN
+ classifier = Sequential()
+
+ # Step 1 - Convolution
+ classifier.add(Conv2D(32, (3, 3), input_shape = (64, 64, 3), activation = 'relu'))
+
+ # Step 2 - Pooling
+ classifier.add(MaxPooling2D(pool_size = (2, 2)))
+
+ # Adding a second convolutional layer
+ classifier.add(Conv2D(32, (3, 3), activation = 'relu'))
+ classifier.add(MaxPooling2D(pool_size = (2, 2)))
+
+ # Step 3 - Flattening
+ classifier.add(Flatten())
+
+ # Step 4 - Full connection
+ classifier.add(Dense(units = 128, activation = 'relu'))
+ classifier.add(Dense(units = 1, activation = 'sigmoid'))
+
+ # Compiling the CNN
+ classifier.compile(optimizer = 'adam', loss = 'binary_crossentropy', metrics = ['accuracy'])
+
+ # Part 2 - Fitting the CNN to the images
+
+
+# Load Trained model weights
+
+ # from keras.models import load_model
+ # regressor=load_model('cnn.h5')
+
+
+ from keras.preprocessing.image import ImageDataGenerator
+ train_datagen = ImageDataGenerator(rescale = 1./255, shear_range = 0.2, zoom_range = 0.2,
+ horizontal_flip = True)
+
+ train_datagen = ImageDataGenerator(rescale = 1./255,
+                                    shear_range = 0.2,
+                                    zoom_range = 0.2,
+                                    horizontal_flip = True)
+
+ test_datagen = ImageDataGenerator(rescale = 1./255)
+ training_set = train_datagen.flow_from_directory('dataset/training_set',target_size = (64, 64),
+ batch_size = 32, class_mode = 'binary')
+ test_set = test_datagen.flow_from_directory('dataset/test_set', target_size = (64, 64), batch_size =
+ 32, class_mode = 'binary')
+ classifier.fit_generator(training_set, _per_epoch = 23, = 25, validation_data = test_set,
+ validation_steps = 2)
+
+ training_set = train_datagen.flow_from_directory('dataset/training_set',
+                                                  target_size = (64, 64),
+                                                  batch_size = 32,
+                                                  class_mode = 'binary')
+
+ test_set = test_datagen.flow_from_directory('dataset/test_set',
+                                             target_size = (64, 64),
+                                             batch_size = 32,
+                                             class_mode = 'binary')
+
+ classifier.fit_generator(training_set,
+                          steps_per_epoch = 5,
+                          epochs = 30,
+                          validation_data = test_set)
+ #classifier.save("cnn.h5")
+
+ # Part 3 - Making new predictions
+
+ import numpy as np
+ from keras.preprocessing import image
+ test_image = image.load_img('dataset/single_prediction/image.png', target_size = (64, 64))
+ test_image = image.load_img(image, target_size = (64, 64))
+ test_image = image.img_to_array(test_image)
+ test_image = np.expand_dims(test_image, axis = 0)
+ result = classifier.predict(test_image)
+ training_set.class_indices
+ if result[0][0] == 0:
+ 	prediction = 'TB'
+ if result[0][0] == 1:
+     prediction = 'Abnormality detected'
+ else:
+ 	prediction = ' NO TB' 
+     prediction = 'Normal'
\ No newline at end of file

From c104a738b846cd510d9f38f8fe6b1e1943675d63 Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Thu, 22 Apr 2021 20:23:07 +0530
Subject: [PATCH 02/32] Update cnn_classification.py

---
 machine_learning/cnn_classification.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/machine_learning/cnn_classification.py b/machine_learning/cnn_classification.py
index f5edc137d1e0..b1d5e2bb777f 100644
--- a/machine_learning/cnn_classification.py
+++ b/machine_learning/cnn_classification.py
@@ -1,5 +1,7 @@
 # Convolutional Neural Network
 
+# Resoucres : https://en.wikipedia.org/wiki/Convolutional_neural_network
+
  # Installing Tensorflow
  # pip install tensorflow
 
@@ -96,4 +98,4 @@
      prediction = 'Abnormality detected'
  else:
  	prediction = ' NO TB' 
-     prediction = 'Normal'
\ No newline at end of file
+     prediction = 'Normal'

From 054fdc7f4ee648611aeef9e09c2cb7e8ccbf9bc1 Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Thu, 22 Apr 2021 20:25:43 +0530
Subject: [PATCH 03/32] Delete cnn_classification.py

---
 machine_learning/cnn_classification.py | 101 -------------------------
 1 file changed, 101 deletions(-)
 delete mode 100644 machine_learning/cnn_classification.py

diff --git a/machine_learning/cnn_classification.py b/machine_learning/cnn_classification.py
deleted file mode 100644
index b1d5e2bb777f..000000000000
--- a/machine_learning/cnn_classification.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# Convolutional Neural Network
-
-# Resoucres : https://en.wikipedia.org/wiki/Convolutional_neural_network
-
- # Installing Tensorflow
- # pip install tensorflow
-
- # Installing Keras
- # pip install --upgrade keras
-
- # Part 1 - Building the CNN
-
- # Importing the Keras libraries and packages
- from keras.models import Sequential
- from keras.layers import Conv2D
- from keras.layers import MaxPooling2D
- from keras.layers import Flatten
- from keras.layers import Dense
-
- # Initialising the CNN
- classifier = Sequential()
-
- # Step 1 - Convolution
- classifier.add(Conv2D(32, (3, 3), input_shape = (64, 64, 3), activation = 'relu'))
-
- # Step 2 - Pooling
- classifier.add(MaxPooling2D(pool_size = (2, 2)))
-
- # Adding a second convolutional layer
- classifier.add(Conv2D(32, (3, 3), activation = 'relu'))
- classifier.add(MaxPooling2D(pool_size = (2, 2)))
-
- # Step 3 - Flattening
- classifier.add(Flatten())
-
- # Step 4 - Full connection
- classifier.add(Dense(units = 128, activation = 'relu'))
- classifier.add(Dense(units = 1, activation = 'sigmoid'))
-
- # Compiling the CNN
- classifier.compile(optimizer = 'adam', loss = 'binary_crossentropy', metrics = ['accuracy'])
-
- # Part 2 - Fitting the CNN to the images
-
-
-# Load Trained model weights
-
- # from keras.models import load_model
- # regressor=load_model('cnn.h5')
-
-
- from keras.preprocessing.image import ImageDataGenerator
- train_datagen = ImageDataGenerator(rescale = 1./255, shear_range = 0.2, zoom_range = 0.2,
- horizontal_flip = True)
-
- train_datagen = ImageDataGenerator(rescale = 1./255,
-                                    shear_range = 0.2,
-                                    zoom_range = 0.2,
-                                    horizontal_flip = True)
-
- test_datagen = ImageDataGenerator(rescale = 1./255)
- training_set = train_datagen.flow_from_directory('dataset/training_set',target_size = (64, 64),
- batch_size = 32, class_mode = 'binary')
- test_set = test_datagen.flow_from_directory('dataset/test_set', target_size = (64, 64), batch_size =
- 32, class_mode = 'binary')
- classifier.fit_generator(training_set, _per_epoch = 23, = 25, validation_data = test_set,
- validation_steps = 2)
-
- training_set = train_datagen.flow_from_directory('dataset/training_set',
-                                                  target_size = (64, 64),
-                                                  batch_size = 32,
-                                                  class_mode = 'binary')
-
- test_set = test_datagen.flow_from_directory('dataset/test_set',
-                                             target_size = (64, 64),
-                                             batch_size = 32,
-                                             class_mode = 'binary')
-
- classifier.fit_generator(training_set,
-                          steps_per_epoch = 5,
-                          epochs = 30,
-                          validation_data = test_set)
- #classifier.save("cnn.h5")
-
- # Part 3 - Making new predictions
-
- import numpy as np
- from keras.preprocessing import image
- test_image = image.load_img('dataset/single_prediction/image.png', target_size = (64, 64))
- test_image = image.load_img(image, target_size = (64, 64))
- test_image = image.img_to_array(test_image)
- test_image = np.expand_dims(test_image, axis = 0)
- result = classifier.predict(test_image)
- training_set.class_indices
- if result[0][0] == 0:
- 	prediction = 'TB'
- if result[0][0] == 1:
-     prediction = 'Abnormality detected'
- else:
- 	prediction = ' NO TB' 
-     prediction = 'Normal'

From d575ef1a1ea141ca0943c7bb83368bc93ec9c46a Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Thu, 22 Apr 2021 20:26:27 +0530
Subject: [PATCH 04/32] cnn classifcation added

---
 computer_vision/cnn_classification.py | 99 +++++++++++++++++++++++++++
 1 file changed, 99 insertions(+)
 create mode 100644 computer_vision/cnn_classification.py

diff --git a/computer_vision/cnn_classification.py b/computer_vision/cnn_classification.py
new file mode 100644
index 000000000000..f5edc137d1e0
--- /dev/null
+++ b/computer_vision/cnn_classification.py
@@ -0,0 +1,99 @@
+# Convolutional Neural Network
+
+ # Installing Tensorflow
+ # pip install tensorflow
+
+ # Installing Keras
+ # pip install --upgrade keras
+
+ # Part 1 - Building the CNN
+
+ # Importing the Keras libraries and packages
+ from keras.models import Sequential
+ from keras.layers import Conv2D
+ from keras.layers import MaxPooling2D
+ from keras.layers import Flatten
+ from keras.layers import Dense
+
+ # Initialising the CNN
+ classifier = Sequential()
+
+ # Step 1 - Convolution
+ classifier.add(Conv2D(32, (3, 3), input_shape = (64, 64, 3), activation = 'relu'))
+
+ # Step 2 - Pooling
+ classifier.add(MaxPooling2D(pool_size = (2, 2)))
+
+ # Adding a second convolutional layer
+ classifier.add(Conv2D(32, (3, 3), activation = 'relu'))
+ classifier.add(MaxPooling2D(pool_size = (2, 2)))
+
+ # Step 3 - Flattening
+ classifier.add(Flatten())
+
+ # Step 4 - Full connection
+ classifier.add(Dense(units = 128, activation = 'relu'))
+ classifier.add(Dense(units = 1, activation = 'sigmoid'))
+
+ # Compiling the CNN
+ classifier.compile(optimizer = 'adam', loss = 'binary_crossentropy', metrics = ['accuracy'])
+
+ # Part 2 - Fitting the CNN to the images
+
+
+# Load Trained model weights
+
+ # from keras.models import load_model
+ # regressor=load_model('cnn.h5')
+
+
+ from keras.preprocessing.image import ImageDataGenerator
+ train_datagen = ImageDataGenerator(rescale = 1./255, shear_range = 0.2, zoom_range = 0.2,
+ horizontal_flip = True)
+
+ train_datagen = ImageDataGenerator(rescale = 1./255,
+                                    shear_range = 0.2,
+                                    zoom_range = 0.2,
+                                    horizontal_flip = True)
+
+ test_datagen = ImageDataGenerator(rescale = 1./255)
+ training_set = train_datagen.flow_from_directory('dataset/training_set',target_size = (64, 64),
+ batch_size = 32, class_mode = 'binary')
+ test_set = test_datagen.flow_from_directory('dataset/test_set', target_size = (64, 64), batch_size =
+ 32, class_mode = 'binary')
+ classifier.fit_generator(training_set, _per_epoch = 23, = 25, validation_data = test_set,
+ validation_steps = 2)
+
+ training_set = train_datagen.flow_from_directory('dataset/training_set',
+                                                  target_size = (64, 64),
+                                                  batch_size = 32,
+                                                  class_mode = 'binary')
+
+ test_set = test_datagen.flow_from_directory('dataset/test_set',
+                                             target_size = (64, 64),
+                                             batch_size = 32,
+                                             class_mode = 'binary')
+
+ classifier.fit_generator(training_set,
+                          steps_per_epoch = 5,
+                          epochs = 30,
+                          validation_data = test_set)
+ #classifier.save("cnn.h5")
+
+ # Part 3 - Making new predictions
+
+ import numpy as np
+ from keras.preprocessing import image
+ test_image = image.load_img('dataset/single_prediction/image.png', target_size = (64, 64))
+ test_image = image.load_img(image, target_size = (64, 64))
+ test_image = image.img_to_array(test_image)
+ test_image = np.expand_dims(test_image, axis = 0)
+ result = classifier.predict(test_image)
+ training_set.class_indices
+ if result[0][0] == 0:
+ 	prediction = 'TB'
+ if result[0][0] == 1:
+     prediction = 'Abnormality detected'
+ else:
+ 	prediction = ' NO TB' 
+     prediction = 'Normal'
\ No newline at end of file

From e24fdf636cc580423fd96f6932684322d7b619bd Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Thu, 22 Apr 2021 20:26:54 +0530
Subject: [PATCH 05/32] cnn classification added


From 62785b780dc268dbbf77d3138b634b8844940c4c Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Thu, 22 Apr 2021 20:35:28 +0530
Subject: [PATCH 06/32] Update cnn_classification.py

---
 computer_vision/cnn_classification.py | 159 +++++++++++++-------------
 1 file changed, 77 insertions(+), 82 deletions(-)

diff --git a/computer_vision/cnn_classification.py b/computer_vision/cnn_classification.py
index f5edc137d1e0..499668458ee9 100644
--- a/computer_vision/cnn_classification.py
+++ b/computer_vision/cnn_classification.py
@@ -1,99 +1,94 @@
 # Convolutional Neural Network
 
- # Installing Tensorflow
- # pip install tensorflow
+# Installing Tensorflow
+# pip install tensorflow
 
- # Installing Keras
- # pip install --upgrade keras
+# Installing Keras
+# pip install --upgrade keras
 
- # Part 1 - Building the CNN
+# Part 1 - Building the CNN
 
- # Importing the Keras libraries and packages
- from keras.models import Sequential
- from keras.layers import Conv2D
- from keras.layers import MaxPooling2D
- from keras.layers import Flatten
- from keras.layers import Dense
+# Importing the Keras libraries and packages
+from keras.models import Sequential
+from keras.layers import Conv2D
+from keras.layers import MaxPooling2D
+from keras.layers import Flatten
+from keras.layers import Dense
 
- # Initialising the CNN
- classifier = Sequential()
+# Initialising the CNN
+classifier = Sequential()
 
- # Step 1 - Convolution
- classifier.add(Conv2D(32, (3, 3), input_shape = (64, 64, 3), activation = 'relu'))
+# Step 1 - Convolution
+classifier.add(Conv2D(32, (3, 3), input_shape = (64, 64, 3), activation = 'relu'))
 
- # Step 2 - Pooling
- classifier.add(MaxPooling2D(pool_size = (2, 2)))
+# Step 2 - Pooling
+classifier.add(MaxPooling2D(pool_size = (2, 2)))
 
- # Adding a second convolutional layer
- classifier.add(Conv2D(32, (3, 3), activation = 'relu'))
- classifier.add(MaxPooling2D(pool_size = (2, 2)))
+# Adding a second convolutional layer
+classifier.add(Conv2D(32, (3, 3), activation = 'relu'))
+classifier.add(MaxPooling2D(pool_size = (2, 2)))
 
- # Step 3 - Flattening
- classifier.add(Flatten())
+# Step 3 - Flattening
+classifier.add(Flatten())
 
- # Step 4 - Full connection
- classifier.add(Dense(units = 128, activation = 'relu'))
- classifier.add(Dense(units = 1, activation = 'sigmoid'))
+# Step 4 - Full connection
+classifier.add(Dense(units = 128, activation = 'relu'))
+classifier.add(Dense(units = 1, activation = 'sigmoid'))
 
- # Compiling the CNN
- classifier.compile(optimizer = 'adam', loss = 'binary_crossentropy', metrics = ['accuracy'])
+# Compiling the CNN
+classifier.compile(optimizer = 'adam', loss = 'binary_crossentropy', metrics = ['accuracy'])
 
- # Part 2 - Fitting the CNN to the images
+# Part 2 - Fitting the CNN to the images
 
 
 # Load Trained model weights
 
- # from keras.models import load_model
- # regressor=load_model('cnn.h5')
-
-
- from keras.preprocessing.image import ImageDataGenerator
- train_datagen = ImageDataGenerator(rescale = 1./255, shear_range = 0.2, zoom_range = 0.2,
- horizontal_flip = True)
-
- train_datagen = ImageDataGenerator(rescale = 1./255,
-                                    shear_range = 0.2,
-                                    zoom_range = 0.2,
-                                    horizontal_flip = True)
-
- test_datagen = ImageDataGenerator(rescale = 1./255)
- training_set = train_datagen.flow_from_directory('dataset/training_set',target_size = (64, 64),
- batch_size = 32, class_mode = 'binary')
- test_set = test_datagen.flow_from_directory('dataset/test_set', target_size = (64, 64), batch_size =
- 32, class_mode = 'binary')
- classifier.fit_generator(training_set, _per_epoch = 23, = 25, validation_data = test_set,
- validation_steps = 2)
-
- training_set = train_datagen.flow_from_directory('dataset/training_set',
-                                                  target_size = (64, 64),
-                                                  batch_size = 32,
-                                                  class_mode = 'binary')
-
- test_set = test_datagen.flow_from_directory('dataset/test_set',
-                                             target_size = (64, 64),
-                                             batch_size = 32,
-                                             class_mode = 'binary')
-
- classifier.fit_generator(training_set,
-                          steps_per_epoch = 5,
-                          epochs = 30,
-                          validation_data = test_set)
- #classifier.save("cnn.h5")
-
- # Part 3 - Making new predictions
-
- import numpy as np
- from keras.preprocessing import image
- test_image = image.load_img('dataset/single_prediction/image.png', target_size = (64, 64))
- test_image = image.load_img(image, target_size = (64, 64))
- test_image = image.img_to_array(test_image)
- test_image = np.expand_dims(test_image, axis = 0)
- result = classifier.predict(test_image)
- training_set.class_indices
- if result[0][0] == 0:
- 	prediction = 'TB'
- if result[0][0] == 1:
-     prediction = 'Abnormality detected'
- else:
- 	prediction = ' NO TB' 
-     prediction = 'Normal'
\ No newline at end of file
+# from keras.models import load_model
+# regressor=load_model('cnn.h5')
+
+
+from keras.preprocessing.image import ImageDataGenerator
+
+train_datagen = ImageDataGenerator(rescale = 1./255, shear_range = 0.2, zoom_range = 0.2,
+horizontal_flip = True)
+
+train_datagen = ImageDataGenerator(rescale = 1./255,
+                                   shear_range = 0.2,
+                                   zoom_range = 0.2,
+                                   horizontal_flip = True)
+
+test_datagen = ImageDataGenerator(rescale = 1./255)
+
+training_set = train_datagen.flow_from_directory('dataset/training_set',
+                                                 target_size = (64, 64),
+                                                 batch_size = 32,
+                                                 class_mode = 'binary')
+
+test_set = test_datagen.flow_from_directory('dataset/test_set',
+                                            target_size = (64, 64),
+                                            batch_size = 32,
+                                            class_mode = 'binary')
+
+classifier.fit_generator(training_set,
+                         steps_per_epoch = 5,
+                         epochs = 30,
+                         validation_data = test_set)
+#classifier.save("cnn.h5")
+
+# Part 3 - Making new predictions
+
+import numpy as np
+from keras.preprocessing import image
+test_image = image.load_img('dataset/single_prediction/image.png', target_size = (64, 64))
+test_image = image.load_img(image, target_size = (64, 64))
+test_image = image.img_to_array(test_image)
+test_image = np.expand_dims(test_image, axis = 0)
+result = classifier.predict(test_image)
+training_set.class_indices
+if result[0][0] == 0:
+ prediction = 'TB'
+if result[0][0] == 1:
+    prediction = 'Abnormality detected'
+else:
+ prediction = ' NO TB' 
+    prediction = 'Normal'

From 96714a3b2da1849c4cc5db503e8f8f5ec956541a Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Thu, 22 Apr 2021 20:36:12 +0530
Subject: [PATCH 07/32] Update cnn_classification.py

---
 computer_vision/cnn_classification.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/computer_vision/cnn_classification.py b/computer_vision/cnn_classification.py
index 499668458ee9..238ab9a323f6 100644
--- a/computer_vision/cnn_classification.py
+++ b/computer_vision/cnn_classification.py
@@ -1,5 +1,7 @@
 # Convolutional Neural Network
 
+#Resources : https://en.wikipedia.org/wiki/Convolutional_neural_network
+
 # Installing Tensorflow
 # pip install tensorflow
 

From 61a8e6537ced05db6c8c116a9c894319c55d6e44 Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Thu, 22 Apr 2021 20:40:26 +0530
Subject: [PATCH 08/32] Update cnn_classification.py

---
 computer_vision/cnn_classification.py | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/computer_vision/cnn_classification.py b/computer_vision/cnn_classification.py
index 238ab9a323f6..1576f33de430 100644
--- a/computer_vision/cnn_classification.py
+++ b/computer_vision/cnn_classification.py
@@ -88,9 +88,7 @@
 result = classifier.predict(test_image)
 training_set.class_indices
 if result[0][0] == 0:
- prediction = 'TB'
+ prediction = 'Normal'
 if result[0][0] == 1:
     prediction = 'Abnormality detected'
-else:
- prediction = ' NO TB' 
-    prediction = 'Normal'
+

From 9e9aa1cbe0d7e670e0f48c93e3e918e9e435a9cb Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Thu, 22 Apr 2021 20:41:45 +0530
Subject: [PATCH 09/32] Update cnn_classification.py

---
 computer_vision/cnn_classification.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/computer_vision/cnn_classification.py b/computer_vision/cnn_classification.py
index 1576f33de430..d0b84c02e868 100644
--- a/computer_vision/cnn_classification.py
+++ b/computer_vision/cnn_classification.py
@@ -88,7 +88,7 @@
 result = classifier.predict(test_image)
 training_set.class_indices
 if result[0][0] == 0:
- prediction = 'Normal'
+  prediction = 'Normal'
 if result[0][0] == 1:
-    prediction = 'Abnormality detected'
+  prediction = 'Abnormality detected'
 

From 5d93c2766500506d5c200e555a9d3c39473605bf Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Fri, 23 Apr 2021 17:21:52 +0530
Subject: [PATCH 10/32] black formatted

---
 computer_vision/cnn_classification.py | 65 +++++++++++++--------------
 1 file changed, 31 insertions(+), 34 deletions(-)

diff --git a/computer_vision/cnn_classification.py b/computer_vision/cnn_classification.py
index d0b84c02e868..c8344f4c6520 100644
--- a/computer_vision/cnn_classification.py
+++ b/computer_vision/cnn_classification.py
@@ -1,6 +1,6 @@
 # Convolutional Neural Network
 
-#Resources : https://en.wikipedia.org/wiki/Convolutional_neural_network
+# Resources : https://en.wikipedia.org/wiki/Convolutional_neural_network
 
 # Installing Tensorflow
 # pip install tensorflow
@@ -21,24 +21,24 @@
 classifier = Sequential()
 
 # Step 1 - Convolution
-classifier.add(Conv2D(32, (3, 3), input_shape = (64, 64, 3), activation = 'relu'))
+classifier.add(Conv2D(32, (3, 3), input_shape=(64, 64, 3), activation="relu"))
 
 # Step 2 - Pooling
-classifier.add(MaxPooling2D(pool_size = (2, 2)))
+classifier.add(MaxPooling2D(pool_size=(2, 2)))
 
 # Adding a second convolutional layer
-classifier.add(Conv2D(32, (3, 3), activation = 'relu'))
-classifier.add(MaxPooling2D(pool_size = (2, 2)))
+classifier.add(Conv2D(32, (3, 3), activation="relu"))
+classifier.add(MaxPooling2D(pool_size=(2, 2)))
 
 # Step 3 - Flattening
 classifier.add(Flatten())
 
 # Step 4 - Full connection
-classifier.add(Dense(units = 128, activation = 'relu'))
-classifier.add(Dense(units = 1, activation = 'sigmoid'))
+classifier.add(Dense(units=128, activation="relu"))
+classifier.add(Dense(units=1, activation="sigmoid"))
 
 # Compiling the CNN
-classifier.compile(optimizer = 'adam', loss = 'binary_crossentropy', metrics = ['accuracy'])
+classifier.compile(optimizer="adam", loss="binary_crossentropy", metrics=["accuracy"])
 
 # Part 2 - Fitting the CNN to the images
 
@@ -51,44 +51,41 @@
 
 from keras.preprocessing.image import ImageDataGenerator
 
-train_datagen = ImageDataGenerator(rescale = 1./255, shear_range = 0.2, zoom_range = 0.2,
-horizontal_flip = True)
+train_datagen = ImageDataGenerator(
+    rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True
+)
 
-train_datagen = ImageDataGenerator(rescale = 1./255,
-                                   shear_range = 0.2,
-                                   zoom_range = 0.2,
-                                   horizontal_flip = True)
+train_datagen = ImageDataGenerator(
+    rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True
+)
 
-test_datagen = ImageDataGenerator(rescale = 1./255)
+test_datagen = ImageDataGenerator(rescale=1.0 / 255)
 
-training_set = train_datagen.flow_from_directory('dataset/training_set',
-                                                 target_size = (64, 64),
-                                                 batch_size = 32,
-                                                 class_mode = 'binary')
+training_set = train_datagen.flow_from_directory(
+    "dataset/training_set", target_size=(64, 64), batch_size=32, class_mode="binary"
+)
 
-test_set = test_datagen.flow_from_directory('dataset/test_set',
-                                            target_size = (64, 64),
-                                            batch_size = 32,
-                                            class_mode = 'binary')
+test_set = test_datagen.flow_from_directory(
+    "dataset/test_set", target_size=(64, 64), batch_size=32, class_mode="binary"
+)
 
-classifier.fit_generator(training_set,
-                         steps_per_epoch = 5,
-                         epochs = 30,
-                         validation_data = test_set)
-#classifier.save("cnn.h5")
+classifier.fit_generator(
+    training_set, steps_per_epoch=5, epochs=30, validation_data=test_set
+)
+# classifier.save("cnn.h5")
 
 # Part 3 - Making new predictions
 
 import numpy as np
 from keras.preprocessing import image
-test_image = image.load_img('dataset/single_prediction/image.png', target_size = (64, 64))
-test_image = image.load_img(image, target_size = (64, 64))
+
+test_image = image.load_img("dataset/single_prediction/image.png", target_size=(64, 64))
+test_image = image.load_img(image, target_size=(64, 64))
 test_image = image.img_to_array(test_image)
-test_image = np.expand_dims(test_image, axis = 0)
+test_image = np.expand_dims(test_image, axis=0)
 result = classifier.predict(test_image)
 training_set.class_indices
 if result[0][0] == 0:
-  prediction = 'Normal'
+    prediction = "Normal"
 if result[0][0] == 1:
-  prediction = 'Abnormality detected'
-
+    prediction = "Abnormality detected"

From b27e7af7152e75275a64f3869d3e47cdd058902c Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Fri, 23 Apr 2021 17:28:52 +0530
Subject: [PATCH 11/32] flake8 corrected

---
 computer_vision/cnn_classification.py | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/computer_vision/cnn_classification.py b/computer_vision/cnn_classification.py
index c8344f4c6520..0b1500a083a1 100644
--- a/computer_vision/cnn_classification.py
+++ b/computer_vision/cnn_classification.py
@@ -16,6 +16,11 @@
 from keras.layers import MaxPooling2D
 from keras.layers import Flatten
 from keras.layers import Dense
+from keras.preprocessing import image
+from keras.preprocessing.image import ImageDataGenerator
+
+import numpy as np
+
 
 # Initialising the CNN
 classifier = Sequential()
@@ -49,8 +54,6 @@
 # regressor=load_model('cnn.h5')
 
 
-from keras.preprocessing.image import ImageDataGenerator
-
 train_datagen = ImageDataGenerator(
     rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True
 )
@@ -76,8 +79,6 @@
 
 # Part 3 - Making new predictions
 
-import numpy as np
-from keras.preprocessing import image
 
 test_image = image.load_img("dataset/single_prediction/image.png", target_size=(64, 64))
 test_image = image.load_img(image, target_size=(64, 64))

From 24bb00e24602a5ed937ec1b51d1f02072264a093 Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Fri, 23 Apr 2021 17:44:15 +0530
Subject: [PATCH 12/32] Add files via upload

---
 computer_vision/cnn_classification.py | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/computer_vision/cnn_classification.py b/computer_vision/cnn_classification.py
index 0b1500a083a1..90797fb4cb9a 100644
--- a/computer_vision/cnn_classification.py
+++ b/computer_vision/cnn_classification.py
@@ -43,7 +43,8 @@
 classifier.add(Dense(units=1, activation="sigmoid"))
 
 # Compiling the CNN
-classifier.compile(optimizer="adam", loss="binary_crossentropy", metrics=["accuracy"])
+classifier.compile(optimizer="adam", loss="binary_crossentropy",
+                   metrics=["accuracy"])
 
 # Part 2 - Fitting the CNN to the images
 
@@ -65,11 +66,13 @@
 test_datagen = ImageDataGenerator(rescale=1.0 / 255)
 
 training_set = train_datagen.flow_from_directory(
-    "dataset/training_set", target_size=(64, 64), batch_size=32, class_mode="binary"
+    "dataset/training_set", target_size=(64, 64), batch_size=32,
+    class_mode="binary"
 )
 
 test_set = test_datagen.flow_from_directory(
-    "dataset/test_set", target_size=(64, 64), batch_size=32, class_mode="binary"
+    "dataset/test_set", target_size=(64, 64), batch_size=32,
+    class_mode="binary"
 )
 
 classifier.fit_generator(
@@ -80,7 +83,8 @@
 # Part 3 - Making new predictions
 
 
-test_image = image.load_img("dataset/single_prediction/image.png", target_size=(64, 64))
+test_image = image.load_img("dataset/single_prediction/image.png",
+                            target_size=(64, 64))
 test_image = image.load_img(image, target_size=(64, 64))
 test_image = image.img_to_array(test_image)
 test_image = np.expand_dims(test_image, axis=0)

From 96a770e7f9b295b23d9ecd47e04cb534e1e45ec7 Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Tue, 27 Apr 2021 17:10:22 +0530
Subject: [PATCH 13/32] added dataset link

---
 computer_vision/cnn_classification.py | 50 ++-------------------------
 1 file changed, 2 insertions(+), 48 deletions(-)

diff --git a/computer_vision/cnn_classification.py b/computer_vision/cnn_classification.py
index 90797fb4cb9a..78f4507d5dda 100644
--- a/computer_vision/cnn_classification.py
+++ b/computer_vision/cnn_classification.py
@@ -1,6 +1,7 @@
 # Convolutional Neural Network
 
 # Resources : https://en.wikipedia.org/wiki/Convolutional_neural_network
+# Download dataset from : https://lhncbc.nlm.nih.gov/LHC-publications/pubs/TuberculosisChestXrayImageDataSets.html
 
 # Installing Tensorflow
 # pip install tensorflow
@@ -46,51 +47,4 @@
 classifier.compile(optimizer="adam", loss="binary_crossentropy",
                    metrics=["accuracy"])
 
-# Part 2 - Fitting the CNN to the images
-
-
-# Load Trained model weights
-
-# from keras.models import load_model
-# regressor=load_model('cnn.h5')
-
-
-train_datagen = ImageDataGenerator(
-    rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True
-)
-
-train_datagen = ImageDataGenerator(
-    rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True
-)
-
-test_datagen = ImageDataGenerator(rescale=1.0 / 255)
-
-training_set = train_datagen.flow_from_directory(
-    "dataset/training_set", target_size=(64, 64), batch_size=32,
-    class_mode="binary"
-)
-
-test_set = test_datagen.flow_from_directory(
-    "dataset/test_set", target_size=(64, 64), batch_size=32,
-    class_mode="binary"
-)
-
-classifier.fit_generator(
-    training_set, steps_per_epoch=5, epochs=30, validation_data=test_set
-)
-# classifier.save("cnn.h5")
-
-# Part 3 - Making new predictions
-
-
-test_image = image.load_img("dataset/single_prediction/image.png",
-                            target_size=(64, 64))
-test_image = image.load_img(image, target_size=(64, 64))
-test_image = image.img_to_array(test_image)
-test_image = np.expand_dims(test_image, axis=0)
-result = classifier.predict(test_image)
-training_set.class_indices
-if result[0][0] == 0:
-    prediction = "Normal"
-if result[0][0] == 1:
-    prediction = "Abnormality detected"
+# Part 

From 9f0064be93b1e727c191f018d5cb09dada78af51 Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Tue, 27 Apr 2021 17:30:08 +0530
Subject: [PATCH 14/32] Delete cnn_classification.py

---
 computer_vision/cnn_classification.py | 50 ---------------------------
 1 file changed, 50 deletions(-)
 delete mode 100644 computer_vision/cnn_classification.py

diff --git a/computer_vision/cnn_classification.py b/computer_vision/cnn_classification.py
deleted file mode 100644
index 78f4507d5dda..000000000000
--- a/computer_vision/cnn_classification.py
+++ /dev/null
@@ -1,50 +0,0 @@
-# Convolutional Neural Network
-
-# Resources : https://en.wikipedia.org/wiki/Convolutional_neural_network
-# Download dataset from : https://lhncbc.nlm.nih.gov/LHC-publications/pubs/TuberculosisChestXrayImageDataSets.html
-
-# Installing Tensorflow
-# pip install tensorflow
-
-# Installing Keras
-# pip install --upgrade keras
-
-# Part 1 - Building the CNN
-
-# Importing the Keras libraries and packages
-from keras.models import Sequential
-from keras.layers import Conv2D
-from keras.layers import MaxPooling2D
-from keras.layers import Flatten
-from keras.layers import Dense
-from keras.preprocessing import image
-from keras.preprocessing.image import ImageDataGenerator
-
-import numpy as np
-
-
-# Initialising the CNN
-classifier = Sequential()
-
-# Step 1 - Convolution
-classifier.add(Conv2D(32, (3, 3), input_shape=(64, 64, 3), activation="relu"))
-
-# Step 2 - Pooling
-classifier.add(MaxPooling2D(pool_size=(2, 2)))
-
-# Adding a second convolutional layer
-classifier.add(Conv2D(32, (3, 3), activation="relu"))
-classifier.add(MaxPooling2D(pool_size=(2, 2)))
-
-# Step 3 - Flattening
-classifier.add(Flatten())
-
-# Step 4 - Full connection
-classifier.add(Dense(units=128, activation="relu"))
-classifier.add(Dense(units=1, activation="sigmoid"))
-
-# Compiling the CNN
-classifier.compile(optimizer="adam", loss="binary_crossentropy",
-                   metrics=["accuracy"])
-
-# Part 

From f6b14d8dfc308aa82239b6da2f440997300aa3c5 Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Tue, 27 Apr 2021 17:30:24 +0530
Subject: [PATCH 15/32] added cnn classification

---
 .../cnn_classification/cnn_classification.py  | 97 +++++++++++++++++++
 .../cnn_classification/requirements.txt       |  3 +
 2 files changed, 100 insertions(+)
 create mode 100644 computer_vision/cnn_classification/cnn_classification.py
 create mode 100644 computer_vision/cnn_classification/requirements.txt

diff --git a/computer_vision/cnn_classification/cnn_classification.py b/computer_vision/cnn_classification/cnn_classification.py
new file mode 100644
index 000000000000..c3b94b8dd769
--- /dev/null
+++ b/computer_vision/cnn_classification/cnn_classification.py
@@ -0,0 +1,97 @@
+
+"""
+Convolutional Neural Network
+
+Resources : https://en.wikipedia.org/wiki/Convolutional_neural_network
+
+Download dataset from :
+https://lhncbc.nlm.nih.gov/LHC-publications/pubs/TuberculosisChestXrayImageDataSets.html
+
+"""
+
+# Part 1 - Building the CNN
+
+# Importing the Keras libraries and packages
+from keras.models import Sequential
+from keras.layers import Conv2D
+from keras.layers import MaxPooling2D
+from keras.layers import Flatten
+from keras.layers import Dense
+from keras.preprocessing import image
+from keras.preprocessing.image import ImageDataGenerator
+
+import numpy as np
+
+
+# Initialising the CNN
+classifier = Sequential()
+
+# Step 1 - Convolution
+classifier.add(Conv2D(32, (3, 3), input_shape=(64, 64, 3), activation="relu"))
+
+# Step 2 - Pooling
+classifier.add(MaxPooling2D(pool_size=(2, 2)))
+
+# Adding a second convolutional layer
+classifier.add(Conv2D(32, (3, 3), activation="relu"))
+classifier.add(MaxPooling2D(pool_size=(2, 2)))
+
+# Step 3 - Flattening
+classifier.add(Flatten())
+
+# Step 4 - Full connection
+classifier.add(Dense(units=128, activation="relu"))
+classifier.add(Dense(units=1, activation="sigmoid"))
+
+# Compiling the CNN
+classifier.compile(optimizer="adam",
+                   loss="binary_crossentropy", metrics=["accuracy"])
+
+# Part 2 - Fitting the CNN to the images
+
+
+# Load Trained model weights
+
+# from keras.models import load_model
+# regressor=load_model('cnn.h5')
+
+train_datagen = ImageDataGenerator(
+    rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True
+)
+
+train_datagen = ImageDataGenerator(
+    rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True
+)
+
+test_datagen = ImageDataGenerator(rescale=1.0 / 255)
+
+training_set = train_datagen.flow_from_directory(
+    "dataset/training_set", target_size=(64, 64), batch_size=32,
+    class_mode="binary"
+)
+
+test_set = test_datagen.flow_from_directory(
+    "dataset/test_set", target_size=(64, 64), batch_size=32,
+    class_mode="binary"
+)
+
+classifier.fit_generator(
+    training_set, steps_per_epoch=5, epochs=30, validation_data=test_set
+)
+
+classifier.save("cnn.h5")
+
+# Part 3 - Making new predictions
+
+
+test_image = image.load_img("dataset/single_prediction/image.png",
+                            target_size=(64, 64))
+test_image = image.load_img(image, target_size=(64, 64))
+test_image = image.img_to_array(test_image)
+test_image = np.expand_dims(test_image, axis=0)
+result = classifier.predict(test_image)
+training_set.class_indices
+if result[0][0] == 0:
+    prediction = "Normal"
+if result[0][0] == 1:
+    prediction = "Abnormality detected"
diff --git a/computer_vision/cnn_classification/requirements.txt b/computer_vision/cnn_classification/requirements.txt
new file mode 100644
index 000000000000..8f65fd9896c1
--- /dev/null
+++ b/computer_vision/cnn_classification/requirements.txt
@@ -0,0 +1,3 @@
+keras
+numpy
+tensorflow

From 2e5ccbd989a7ef379935322a24ac7346e49d47db Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Tue, 27 Apr 2021 17:35:35 +0530
Subject: [PATCH 16/32] Update cnn_classification.py

---
 computer_vision/cnn_classification/cnn_classification.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/computer_vision/cnn_classification/cnn_classification.py b/computer_vision/cnn_classification/cnn_classification.py
index c3b94b8dd769..9d6f756a9086 100644
--- a/computer_vision/cnn_classification/cnn_classification.py
+++ b/computer_vision/cnn_classification/cnn_classification.py
@@ -2,6 +2,8 @@
 """
 Convolutional Neural Network
 
+Objective : To train a CNN model detect if TB is present in Lung X-ray or not.
+
 Resources : https://en.wikipedia.org/wiki/Convolutional_neural_network
 
 Download dataset from :

From 7981c4a99961fd6df86e2a16b299ef8b96c017fb Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Tue, 27 Apr 2021 17:38:46 +0530
Subject: [PATCH 17/32] Update cnn_classification.py

---
 computer_vision/cnn_classification/cnn_classification.py | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/computer_vision/cnn_classification/cnn_classification.py b/computer_vision/cnn_classification/cnn_classification.py
index 9d6f756a9086..c5a4a3ca0ff8 100644
--- a/computer_vision/cnn_classification/cnn_classification.py
+++ b/computer_vision/cnn_classification/cnn_classification.py
@@ -61,10 +61,6 @@
     rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True
 )
 
-train_datagen = ImageDataGenerator(
-    rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True
-)
-
 test_datagen = ImageDataGenerator(rescale=1.0 / 255)
 
 training_set = train_datagen.flow_from_directory(

From ab1e53db530d2512753047f8a59375948a7789dd Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Tue, 27 Apr 2021 17:39:49 +0530
Subject: [PATCH 18/32] Update cnn_classification.py

---
 computer_vision/cnn_classification/cnn_classification.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/computer_vision/cnn_classification/cnn_classification.py b/computer_vision/cnn_classification/cnn_classification.py
index c5a4a3ca0ff8..41767c91e954 100644
--- a/computer_vision/cnn_classification/cnn_classification.py
+++ b/computer_vision/cnn_classification/cnn_classification.py
@@ -4,7 +4,8 @@
 
 Objective : To train a CNN model detect if TB is present in Lung X-ray or not.
 
-Resources : https://en.wikipedia.org/wiki/Convolutional_neural_network
+Resources CNN Theory : https://en.wikipedia.org/wiki/Convolutional_neural_network
+Resources Tensorflow : https://www.tensorflow.org/tutorials/images/cnn
 
 Download dataset from :
 https://lhncbc.nlm.nih.gov/LHC-publications/pubs/TuberculosisChestXrayImageDataSets.html

From ca54ea03bcafcab605cd52f6765c8e5a3bf524a9 Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Tue, 27 Apr 2021 18:13:17 +0530
Subject: [PATCH 19/32] Delete requirements.txt

---
 computer_vision/cnn_classification/requirements.txt | 3 ---
 1 file changed, 3 deletions(-)
 delete mode 100644 computer_vision/cnn_classification/requirements.txt

diff --git a/computer_vision/cnn_classification/requirements.txt b/computer_vision/cnn_classification/requirements.txt
deleted file mode 100644
index 8f65fd9896c1..000000000000
--- a/computer_vision/cnn_classification/requirements.txt
+++ /dev/null
@@ -1,3 +0,0 @@
-keras
-numpy
-tensorflow

From c8ff1759d4cd00f6390e2585d4d3fded6c5e40c4 Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Tue, 27 Apr 2021 19:17:41 +0530
Subject: [PATCH 20/32] Update cnn_classification.py

---
 .../cnn_classification/cnn_classification.py  | 26 +++++++++----------
 1 file changed, 12 insertions(+), 14 deletions(-)

diff --git a/computer_vision/cnn_classification/cnn_classification.py b/computer_vision/cnn_classification/cnn_classification.py
index 41767c91e954..59eca1734cb9 100644
--- a/computer_vision/cnn_classification/cnn_classification.py
+++ b/computer_vision/cnn_classification/cnn_classification.py
@@ -14,12 +14,10 @@
 
 # Part 1 - Building the CNN
 
-# Importing the Keras libraries and packages
-from keras.models import Sequential
-from keras.layers import Conv2D
-from keras.layers import MaxPooling2D
-from keras.layers import Flatten
-from keras.layers import Dense
+# Importing the  libraries and packages
+import tensorflow as tf
+from tensorflow.keras import layers, models
+
 from keras.preprocessing import image
 from keras.preprocessing.image import ImageDataGenerator
 
@@ -27,24 +25,24 @@
 
 
 # Initialising the CNN
-classifier = Sequential()
+classifier = models.Sequential()
 
 # Step 1 - Convolution
-classifier.add(Conv2D(32, (3, 3), input_shape=(64, 64, 3), activation="relu"))
+classifier.add(layers.Conv2D(32, (3, 3), input_shape=(64, 64, 3), activation="relu"))
 
 # Step 2 - Pooling
-classifier.add(MaxPooling2D(pool_size=(2, 2)))
+classifier.add(layers.MaxPooling2D(pool_size=(2, 2)))
 
 # Adding a second convolutional layer
-classifier.add(Conv2D(32, (3, 3), activation="relu"))
-classifier.add(MaxPooling2D(pool_size=(2, 2)))
+classifier.add(layers.Conv2D(32, (3, 3), activation="relu"))
+classifier.add(layers.MaxPooling2D(pool_size=(2, 2)))
 
 # Step 3 - Flattening
-classifier.add(Flatten())
+classifier.add(layers.Flatten())
 
 # Step 4 - Full connection
-classifier.add(Dense(units=128, activation="relu"))
-classifier.add(Dense(units=1, activation="sigmoid"))
+classifier.add(layers.Dense(units=128, activation="relu"))
+classifier.add(layers.Dense(units=1, activation="sigmoid"))
 
 # Compiling the CNN
 classifier.compile(optimizer="adam",

From 8d12d8d8081beb5a2ac9358ae2115aa0e44a88ed Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Tue, 27 Apr 2021 19:17:50 +0530
Subject: [PATCH 21/32] Create cnn_classification.py


From ecd416147041faa441cf65d6f48f15e768ee0ffa Mon Sep 17 00:00:00 2001
From: 9harshit <43147421+9harshit@users.noreply.github.com>
Date: Tue, 27 Apr 2021 19:21:54 +0530
Subject: [PATCH 22/32] Add files via upload

---
 .../cnn_classification/cnn_classification.py          | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

diff --git a/computer_vision/cnn_classification/cnn_classification.py b/computer_vision/cnn_classification/cnn_classification.py
index 59eca1734cb9..b0e37ebf5c33 100644
--- a/computer_vision/cnn_classification/cnn_classification.py
+++ b/computer_vision/cnn_classification/cnn_classification.py
@@ -1,4 +1,3 @@
-
 """
 Convolutional Neural Network
 
@@ -14,8 +13,7 @@
 
 # Part 1 - Building the CNN
 
-# Importing the  libraries and packages
-import tensorflow as tf
+# Importing the Keras libraries and packages
 from tensorflow.keras import layers, models
 
 from keras.preprocessing import image
@@ -28,7 +26,8 @@
 classifier = models.Sequential()
 
 # Step 1 - Convolution
-classifier.add(layers.Conv2D(32, (3, 3), input_shape=(64, 64, 3), activation="relu"))
+classifier.add(layers.Conv2D(32, (3, 3), input_shape=(64, 64, 3),
+                             activation="relu"))
 
 # Step 2 - Pooling
 classifier.add(layers.MaxPooling2D(pool_size=(2, 2)))
@@ -45,8 +44,8 @@
 classifier.add(layers.Dense(units=1, activation="sigmoid"))
 
 # Compiling the CNN
-classifier.compile(optimizer="adam",
-                   loss="binary_crossentropy", metrics=["accuracy"])
+classifier.compile(optimizer="adam", loss="binary_crossentropy",
+                   metrics=["accuracy"])
 
 # Part 2 - Fitting the CNN to the images
 

From 201d6363c5640a95906eee5bda6fc751a569fd4a Mon Sep 17 00:00:00 2001
From: Harshit Agarwal <43147421+9harshit@users.noreply.github.com>
Date: Fri, 30 Apr 2021 10:34:54 +0530
Subject: [PATCH 23/32] using keras from tensorflow only

---
 .../cnn_classification/cnn_classification.py       | 14 ++++++--------
 1 file changed, 6 insertions(+), 8 deletions(-)

diff --git a/computer_vision/cnn_classification/cnn_classification.py b/computer_vision/cnn_classification/cnn_classification.py
index b0e37ebf5c33..615c2ff8bf90 100644
--- a/computer_vision/cnn_classification/cnn_classification.py
+++ b/computer_vision/cnn_classification/cnn_classification.py
@@ -14,11 +14,9 @@
 # Part 1 - Building the CNN
 
 # Importing the Keras libraries and packages
+import tensorflow as tf
 from tensorflow.keras import layers, models
 
-from keras.preprocessing import image
-from keras.preprocessing.image import ImageDataGenerator
-
 import numpy as np
 
 
@@ -55,11 +53,11 @@
 # from keras.models import load_model
 # regressor=load_model('cnn.h5')
 
-train_datagen = ImageDataGenerator(
+train_datagen = tf.keras.preprocessing.image.ImageDataGenerator(
     rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True
 )
 
-test_datagen = ImageDataGenerator(rescale=1.0 / 255)
+test_datagen = tf.keras.preprocessing.image.ImageDataGenerator(rescale=1.0 / 255)
 
 training_set = train_datagen.flow_from_directory(
     "dataset/training_set", target_size=(64, 64), batch_size=32,
@@ -80,10 +78,10 @@
 # Part 3 - Making new predictions
 
 
-test_image = image.load_img("dataset/single_prediction/image.png",
+test_image = tf.keras.preprocessing.image.load_img("dataset/single_prediction/image.png",
                             target_size=(64, 64))
-test_image = image.load_img(image, target_size=(64, 64))
-test_image = image.img_to_array(test_image)
+test_image = tf.keras.preprocessing.image.load_img(image, target_size=(64, 64))
+test_image = tf.keras.preprocessing.image.img_to_array(test_image)
 test_image = np.expand_dims(test_image, axis=0)
 result = classifier.predict(test_image)
 training_set.class_indices

From 89e0107fe592fa9fa925399efc027e57e687e16a Mon Sep 17 00:00:00 2001
From: Harshit Agarwal <43147421+9harshit@users.noreply.github.com>
Date: Fri, 30 Apr 2021 10:48:50 +0530
Subject: [PATCH 24/32] update tensorflow

---
 .../cnn_classification/cnn_classification.py     | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/computer_vision/cnn_classification/cnn_classification.py b/computer_vision/cnn_classification/cnn_classification.py
index 615c2ff8bf90..7c310b4ab4af 100644
--- a/computer_vision/cnn_classification/cnn_classification.py
+++ b/computer_vision/cnn_classification/cnn_classification.py
@@ -57,16 +57,17 @@
     rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True
 )
 
-test_datagen = tf.keras.preprocessing.image.ImageDataGenerator(rescale=1.0 / 255)
+test_datagen = tf.keras.preprocessing.image.ImageDataGenerator(
+    rescale=1.0 / 255)
 
 training_set = train_datagen.flow_from_directory(
-    "dataset/training_set", target_size=(64, 64), batch_size=32,
-    class_mode="binary"
+    "dataset/training_set", target_size=(64, 64),
+    batch_size=32, class_mode="binary"
 )
 
 test_set = test_datagen.flow_from_directory(
-    "dataset/test_set", target_size=(64, 64), batch_size=32,
-    class_mode="binary"
+    "dataset/test_set", target_size=(64, 64),
+    batch_size=32, class_mode="binary"
 )
 
 classifier.fit_generator(
@@ -78,9 +79,8 @@
 # Part 3 - Making new predictions
 
 
-test_image = tf.keras.preprocessing.image.load_img("dataset/single_prediction/image.png",
-                            target_size=(64, 64))
-test_image = tf.keras.preprocessing.image.load_img(image, target_size=(64, 64))
+test_image = tf.keras.preprocessing.image.load_img(
+    "dataset/single_prediction/image.png", target_size=(64, 64))
 test_image = tf.keras.preprocessing.image.img_to_array(test_image)
 test_image = np.expand_dims(test_image, axis=0)
 result = classifier.predict(test_image)

From 2c83aafb633a632697949ccd74e592bce1acc77c Mon Sep 17 00:00:00 2001
From: Harshit Agarwal <43147421+9harshit@users.noreply.github.com>
Date: Mon, 24 May 2021 13:20:51 +0530
Subject: [PATCH 25/32] Add files via upload

---
 .../cnn_classification/cnn_classification.py  | 110 +++++++++---------
 1 file changed, 58 insertions(+), 52 deletions(-)

diff --git a/computer_vision/cnn_classification/cnn_classification.py b/computer_vision/cnn_classification/cnn_classification.py
index 7c310b4ab4af..cf5909474839 100644
--- a/computer_vision/cnn_classification/cnn_classification.py
+++ b/computer_vision/cnn_classification/cnn_classification.py
@@ -9,83 +9,89 @@
 Download dataset from :
 https://lhncbc.nlm.nih.gov/LHC-publications/pubs/TuberculosisChestXrayImageDataSets.html
 
+1. Download the dataset folder and create two folder training set and test set in the parent dataste folder
+2. Move 30-40 image from both TB positive and TB Negative folder in the test set folder
+3. The labels of the iamges will be extracted from the folder name the image is present in.
+
 """
 
 # Part 1 - Building the CNN
 
 # Importing the Keras libraries and packages
-import tensorflow as tf
+import tensorflow as tf 
 from tensorflow.keras import layers, models
-
 import numpy as np
 
 
-# Initialising the CNN
-classifier = models.Sequential()
+if __name__ == '__main__':
+
+
+	# Initialising the CNN
+	classifier = models.Sequential()
 
-# Step 1 - Convolution
-classifier.add(layers.Conv2D(32, (3, 3), input_shape=(64, 64, 3),
-                             activation="relu"))
+	# Step 1 - Convolution
+	classifier.add(layers.Conv2D(32, (3, 3), input_shape=(64, 64, 3),
+	                             activation="relu"))
 
-# Step 2 - Pooling
-classifier.add(layers.MaxPooling2D(pool_size=(2, 2)))
+	# Step 2 - Pooling
+	classifier.add(layers.MaxPooling2D(pool_size=(2, 2)))
 
-# Adding a second convolutional layer
-classifier.add(layers.Conv2D(32, (3, 3), activation="relu"))
-classifier.add(layers.MaxPooling2D(pool_size=(2, 2)))
+	# Adding a second convolutional layer
+	classifier.add(layers.Conv2D(32, (3, 3), activation="relu"))
+	classifier.add(layers.MaxPooling2D(pool_size=(2, 2)))
 
-# Step 3 - Flattening
-classifier.add(layers.Flatten())
+	# Step 3 - Flattening
+	classifier.add(layers.Flatten())
 
-# Step 4 - Full connection
-classifier.add(layers.Dense(units=128, activation="relu"))
-classifier.add(layers.Dense(units=1, activation="sigmoid"))
+	# Step 4 - Full connection
+	classifier.add(layers.Dense(units=128, activation="relu"))
+	classifier.add(layers.Dense(units=1, activation="sigmoid"))
 
-# Compiling the CNN
-classifier.compile(optimizer="adam", loss="binary_crossentropy",
-                   metrics=["accuracy"])
+	# Compiling the CNN
+	classifier.compile(optimizer="adam", loss="binary_crossentropy",
+	                   metrics=["accuracy"])
 
-# Part 2 - Fitting the CNN to the images
+	# Part 2 - Fitting the CNN to the images
 
 
-# Load Trained model weights
+	# Load Trained model weights
 
-# from keras.models import load_model
-# regressor=load_model('cnn.h5')
+	# from keras.models import load_model
+	# regressor=load_model('cnn.h5')
 
-train_datagen = tf.keras.preprocessing.image.ImageDataGenerator(
-    rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True
-)
+	train_datagen = tf.keras.preprocessing.image.ImageDataGenerator(
+	    rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True
+	)
 
-test_datagen = tf.keras.preprocessing.image.ImageDataGenerator(
-    rescale=1.0 / 255)
+	test_datagen = tf.keras.preprocessing.image.ImageDataGenerator(
+	    rescale=1.0 / 255)
 
-training_set = train_datagen.flow_from_directory(
-    "dataset/training_set", target_size=(64, 64),
-    batch_size=32, class_mode="binary"
-)
+	training_set = train_datagen.flow_from_directory(
+	    "dataset/training_set", target_size=(64, 64),
+	    batch_size=32, class_mode="binary"
+	)
 
-test_set = test_datagen.flow_from_directory(
-    "dataset/test_set", target_size=(64, 64),
-    batch_size=32, class_mode="binary"
-)
+	test_set = test_datagen.flow_from_directory(
+	    "dataset/test_set", target_size=(64, 64),
+	    batch_size=32, class_mode="binary"
+	)
 
-classifier.fit_generator(
-    training_set, steps_per_epoch=5, epochs=30, validation_data=test_set
-)
+	classifier.fit_generator(
+	    training_set, steps_per_epoch=5, epochs=30, validation_data=test_set
+	)
 
-classifier.save("cnn.h5")
+	classifier.save("cnn.h5")
 
-# Part 3 - Making new predictions
+	# Part 3 - Making new predictions
 
 
-test_image = tf.keras.preprocessing.image.load_img(
-    "dataset/single_prediction/image.png", target_size=(64, 64))
-test_image = tf.keras.preprocessing.image.img_to_array(test_image)
-test_image = np.expand_dims(test_image, axis=0)
-result = classifier.predict(test_image)
-training_set.class_indices
-if result[0][0] == 0:
-    prediction = "Normal"
-if result[0][0] == 1:
-    prediction = "Abnormality detected"
+	test_image = tf.keras.preprocessing.image.load_img(
+	    "dataset/single_prediction/image.png", target_size=(64, 64))
+	test_image = tf.keras.preprocessing.image.img_to_array(test_image)
+	test_image = np.expand_dims(test_image, axis=0)
+	result = classifier.predict(test_image)
+	training_set.class_indices
+	if result[0][0] == 0:
+	    prediction = "Normal"
+	if result[0][0] == 1:
+	    prediction = "Abnormality detected"

From c43489d5f409785d6dd6a8afd7319542a31bf813 Mon Sep 17 00:00:00 2001
From: Harshit Agarwal <43147421+9harshit@users.noreply.github.com>
Date: Mon, 24 May 2021 13:23:40 +0530
Subject: [PATCH 26/32] Add files via upload

---
 .../cnn_classification/cnn_classification.py  | 107 +++++++++---------
 1 file changed, 52 insertions(+), 55 deletions(-)

diff --git a/computer_vision/cnn_classification/cnn_classification.py b/computer_vision/cnn_classification/cnn_classification.py
index cf5909474839..0f48f34c620d 100644
--- a/computer_vision/cnn_classification/cnn_classification.py
+++ b/computer_vision/cnn_classification/cnn_classification.py
@@ -18,80 +18,77 @@
 # Part 1 - Building the CNN
 
 # Importing the Keras libraries and packages
-import tensorflow as tf 
+import tensorflow as tf
 from tensorflow.keras import layers, models
 import numpy as np
 
 
-if __name__ == '__main__':
+if __name__ == "__main__":
 
+    # Initialising the CNN
+    classifier = models.Sequential()
 
-	# Initialising the CNN
-	classifier = models.Sequential()
+    # Step 1 - Convolution
+    classifier.add(
+        layers.Conv2D(32, (3, 3), input_shape=(64, 64, 3), activation="relu")
+    )
 
-	# Step 1 - Convolution
-	classifier.add(layers.Conv2D(32, (3, 3), input_shape=(64, 64, 3),
-	                             activation="relu"))
+    # Step 2 - Pooling
+    classifier.add(layers.MaxPooling2D(pool_size=(2, 2)))
 
-	# Step 2 - Pooling
-	classifier.add(layers.MaxPooling2D(pool_size=(2, 2)))
+    # Adding a second convolutional layer
+    classifier.add(layers.Conv2D(32, (3, 3), activation="relu"))
+    classifier.add(layers.MaxPooling2D(pool_size=(2, 2)))
 
-	# Adding a second convolutional layer
-	classifier.add(layers.Conv2D(32, (3, 3), activation="relu"))
-	classifier.add(layers.MaxPooling2D(pool_size=(2, 2)))
+    # Step 3 - Flattening
+    classifier.add(layers.Flatten())
 
-	# Step 3 - Flattening
-	classifier.add(layers.Flatten())
+    # Step 4 - Full connection
+    classifier.add(layers.Dense(units=128, activation="relu"))
+    classifier.add(layers.Dense(units=1, activation="sigmoid"))
 
-	# Step 4 - Full connection
-	classifier.add(layers.Dense(units=128, activation="relu"))
-	classifier.add(layers.Dense(units=1, activation="sigmoid"))
+    # Compiling the CNN
+    classifier.compile(
+        optimizer="adam", loss="binary_crossentropy", metrics=["accuracy"]
+    )
 
-	# Compiling the CNN
-	classifier.compile(optimizer="adam", loss="binary_crossentropy",
-	                   metrics=["accuracy"])
+    # Part 2 - Fitting the CNN to the images
 
-	# Part 2 - Fitting the CNN to the images
+    # Load Trained model weights
 
+    # from keras.models import load_model
+    # regressor=load_model('cnn.h5')
 
-	# Load Trained model weights
+    train_datagen = tf.keras.preprocessing.image.ImageDataGenerator(
+        rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True
+    )
 
-	# from keras.models import load_model
-	# regressor=load_model('cnn.h5')
+    test_datagen = tf.keras.preprocessing.image.ImageDataGenerator(rescale=1.0 / 255)
 
-	train_datagen = tf.keras.preprocessing.image.ImageDataGenerator(
-	    rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True
-	)
+    training_set = train_datagen.flow_from_directory(
+        "dataset/training_set", target_size=(64, 64), batch_size=32, class_mode="binary"
+    )
 
-	test_datagen = tf.keras.preprocessing.image.ImageDataGenerator(
-	    rescale=1.0 / 255)
+    test_set = test_datagen.flow_from_directory(
+        "dataset/test_set", target_size=(64, 64), batch_size=32, class_mode="binary"
+    )
 
-	training_set = train_datagen.flow_from_directory(
-	    "dataset/training_set", target_size=(64, 64),
-	    batch_size=32, class_mode="binary"
-	)
+    classifier.fit_generator(
+        training_set, steps_per_epoch=5, epochs=30, validation_data=test_set
+    )
 
-	test_set = test_datagen.flow_from_directory(
-	    "dataset/test_set", target_size=(64, 64),
-	    batch_size=32, class_mode="binary"
-	)
+    classifier.save("cnn.h5")
 
-	classifier.fit_generator(
-	    training_set, steps_per_epoch=5, epochs=30, validation_data=test_set
-	)
+    # Part 3 - Making new predictions
 
-	classifier.save("cnn.h5")
-
-	# Part 3 - Making new predictions
-
-
-	test_image = tf.keras.preprocessing.image.load_img(
-	    "dataset/single_prediction/image.png", target_size=(64, 64))
-	test_image = tf.keras.preprocessing.image.img_to_array(test_image)
-	test_image = np.expand_dims(test_image, axis=0)
-	result = classifier.predict(test_image)
-	training_set.class_indices
-	if result[0][0] == 0:
-	    prediction = "Normal"
-	if result[0][0] == 1:
-	    prediction = "Abnormality detected"
+    test_image = tf.keras.preprocessing.image.load_img(
+        "dataset/single_prediction/image.png", target_size=(64, 64)
+    )
+    test_image = tf.keras.preprocessing.image.img_to_array(test_image)
+    test_image = np.expand_dims(test_image, axis=0)
+    result = classifier.predict(test_image)
+    training_set.class_indices
+    if result[0][0] == 0:
+        prediction = "Normal"
+    if result[0][0] == 1:
+        prediction = "Abnormality detected"

From 4ac3d3bcae56744ffd45be68f2bccc0f99e15192 Mon Sep 17 00:00:00 2001
From: Harshit Agarwal <43147421+9harshit@users.noreply.github.com>
Date: Mon, 24 May 2021 16:55:18 +0530
Subject: [PATCH 27/32] Update cnn_classification.py

---
 computer_vision/cnn_classification/cnn_classification.py | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/computer_vision/cnn_classification/cnn_classification.py b/computer_vision/cnn_classification/cnn_classification.py
index 0f48f34c620d..9a8a9e05bd31 100644
--- a/computer_vision/cnn_classification/cnn_classification.py
+++ b/computer_vision/cnn_classification/cnn_classification.py
@@ -9,9 +9,12 @@
 Download dataset from :
 https://lhncbc.nlm.nih.gov/LHC-publications/pubs/TuberculosisChestXrayImageDataSets.html
 
-1. Download the dataset folder and create two folder training set and test set in the parent dataste folder
-2. Move 30-40 image from both TB positive and TB Negative folder in the test set folder
-3. The labels of the iamges will be extracted from the folder name the image is present in.
+1. Download the dataset folder and create two folder training set and test set in the parent 
+dataste folder
+2. Move 30-40 image from both TB positive and TB Negative folder in the test 
+set folder
+3. The labels of the iamges will be extracted from the folder name the image
+is present in.
 
 """
 

From a3b550cc0472110fa56c2e8033004cc0b376bd78 Mon Sep 17 00:00:00 2001
From: Harshit Agarwal <43147421+9harshit@users.noreply.github.com>
Date: Mon, 24 May 2021 17:01:43 +0530
Subject: [PATCH 28/32] Add files via upload

---
 .../cnn_classification/cnn_classification.py  | 27 +++++++++++--------
 1 file changed, 16 insertions(+), 11 deletions(-)

diff --git a/computer_vision/cnn_classification/cnn_classification.py b/computer_vision/cnn_classification/cnn_classification.py
index 9a8a9e05bd31..01d83fd26937 100644
--- a/computer_vision/cnn_classification/cnn_classification.py
+++ b/computer_vision/cnn_classification/cnn_classification.py
@@ -3,18 +3,19 @@
 
 Objective : To train a CNN model detect if TB is present in Lung X-ray or not.
 
-Resources CNN Theory : https://en.wikipedia.org/wiki/Convolutional_neural_network
+Resources CNN Theory :
+    https://en.wikipedia.org/wiki/Convolutional_neural_network
 Resources Tensorflow : https://www.tensorflow.org/tutorials/images/cnn
 
 Download dataset from :
 https://lhncbc.nlm.nih.gov/LHC-publications/pubs/TuberculosisChestXrayImageDataSets.html
 
-1. Download the dataset folder and create two folder training set and test set in the parent 
-dataste folder
-2. Move 30-40 image from both TB positive and TB Negative folder in the test 
-set folder
-3. The labels of the iamges will be extracted from the folder name the image
-is present in.
+1. Download the dataset folder and create two folder training set and test set
+in the parent dataste folder
+2. Move 30-40 image from both TB positive and TB Negative folder
+in the test set folder
+3. The labels of the iamges will be extracted from the folder name
+the image is present in.
 
 """
 
@@ -63,17 +64,21 @@
     # regressor=load_model('cnn.h5')
 
     train_datagen = tf.keras.preprocessing.image.ImageDataGenerator(
-        rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True
+        rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2,
+        horizontal_flip=True
     )
 
-    test_datagen = tf.keras.preprocessing.image.ImageDataGenerator(rescale=1.0 / 255)
+    test_datagen = tf.keras.preprocessing.image.ImageDataGenerator(
+        rescale=1.0 / 255)
 
     training_set = train_datagen.flow_from_directory(
-        "dataset/training_set", target_size=(64, 64), batch_size=32, class_mode="binary"
+        "dataset/training_set", target_size=(64, 64),
+        batch_size=32, class_mode="binary"
     )
 
     test_set = test_datagen.flow_from_directory(
-        "dataset/test_set", target_size=(64, 64), batch_size=32, class_mode="binary"
+        "dataset/test_set", target_size=(64, 64),
+        batch_size=32, class_mode="binary"
     )
 
     classifier.fit_generator(

From 3b100ec439ca10ba62870e2b6a6b34c2981d7f38 Mon Sep 17 00:00:00 2001
From: Harshit Agarwal <43147421+9harshit@users.noreply.github.com>
Date: Mon, 24 May 2021 17:10:11 +0530
Subject: [PATCH 29/32] Add files via upload

---
 computer_vision/cnn_classification/cnn_classification.py | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/computer_vision/cnn_classification/cnn_classification.py b/computer_vision/cnn_classification/cnn_classification.py
index 01d83fd26937..c07e255ef199 100644
--- a/computer_vision/cnn_classification/cnn_classification.py
+++ b/computer_vision/cnn_classification/cnn_classification.py
@@ -72,13 +72,13 @@
         rescale=1.0 / 255)
 
     training_set = train_datagen.flow_from_directory(
-        "dataset/training_set", target_size=(64, 64),
-        batch_size=32, class_mode="binary"
+        "dataset/training_set", target_size=(64, 64), batch_size=32,
+        class_mode="binary"
     )
 
     test_set = test_datagen.flow_from_directory(
-        "dataset/test_set", target_size=(64, 64),
-        batch_size=32, class_mode="binary"
+        "dataset/test_set", target_size=(64, 64), batch_size=32,
+        class_mode="binary"
     )
 
     classifier.fit_generator(

From 9c3b895fcb7f2913911f3e0c3fa63701ec90be23 Mon Sep 17 00:00:00 2001
From: Harshit Agarwal <43147421+9harshit@users.noreply.github.com>
Date: Mon, 24 May 2021 17:31:51 +0530
Subject: [PATCH 30/32] changes made

---
 .../cnn_classification/cnn_classification.py     | 16 ++++++----------
 1 file changed, 6 insertions(+), 10 deletions(-)

diff --git a/computer_vision/cnn_classification/cnn_classification.py b/computer_vision/cnn_classification/cnn_classification.py
index c07e255ef199..6d4f19639c24 100644
--- a/computer_vision/cnn_classification/cnn_classification.py
+++ b/computer_vision/cnn_classification/cnn_classification.py
@@ -21,11 +21,11 @@
 
 # Part 1 - Building the CNN
 
+import numpy as np
+
 # Importing the Keras libraries and packages
 import tensorflow as tf
 from tensorflow.keras import layers, models
-import numpy as np
-
 
 if __name__ == "__main__":
 
@@ -64,21 +64,17 @@
     # regressor=load_model('cnn.h5')
 
     train_datagen = tf.keras.preprocessing.image.ImageDataGenerator(
-        rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2,
-        horizontal_flip=True
+        rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True
     )
 
-    test_datagen = tf.keras.preprocessing.image.ImageDataGenerator(
-        rescale=1.0 / 255)
+    test_datagen = tf.keras.preprocessing.image.ImageDataGenerator(rescale=1.0 / 255)
 
     training_set = train_datagen.flow_from_directory(
-        "dataset/training_set", target_size=(64, 64), batch_size=32,
-        class_mode="binary"
+        "dataset/training_set", target_size=(64, 64), batch_size=32, class_mode="binary"
     )
 
     test_set = test_datagen.flow_from_directory(
-        "dataset/test_set", target_size=(64, 64), batch_size=32,
-        class_mode="binary"
+        "dataset/test_set", target_size=(64, 64), batch_size=32, class_mode="binary"
     )
 
     classifier.fit_generator(

From 57f85c6a8b5d55a3ded0e9543d860f59c854e176 Mon Sep 17 00:00:00 2001
From: Harshit Agarwal <43147421+9harshit@users.noreply.github.com>
Date: Thu, 24 Jun 2021 10:22:31 +0530
Subject: [PATCH 31/32] Delete computer_vision/cnn_classification directory

---
 .../cnn_classification/cnn_classification.py  | 98 -------------------
 1 file changed, 98 deletions(-)
 delete mode 100644 computer_vision/cnn_classification/cnn_classification.py

diff --git a/computer_vision/cnn_classification/cnn_classification.py b/computer_vision/cnn_classification/cnn_classification.py
deleted file mode 100644
index 6d4f19639c24..000000000000
--- a/computer_vision/cnn_classification/cnn_classification.py
+++ /dev/null
@@ -1,98 +0,0 @@
-"""
-Convolutional Neural Network
-
-Objective : To train a CNN model detect if TB is present in Lung X-ray or not.
-
-Resources CNN Theory :
-    https://en.wikipedia.org/wiki/Convolutional_neural_network
-Resources Tensorflow : https://www.tensorflow.org/tutorials/images/cnn
-
-Download dataset from :
-https://lhncbc.nlm.nih.gov/LHC-publications/pubs/TuberculosisChestXrayImageDataSets.html
-
-1. Download the dataset folder and create two folder training set and test set
-in the parent dataste folder
-2. Move 30-40 image from both TB positive and TB Negative folder
-in the test set folder
-3. The labels of the iamges will be extracted from the folder name
-the image is present in.
-
-"""
-
-# Part 1 - Building the CNN
-
-import numpy as np
-
-# Importing the Keras libraries and packages
-import tensorflow as tf
-from tensorflow.keras import layers, models
-
-if __name__ == "__main__":
-
-    # Initialising the CNN
-    classifier = models.Sequential()
-
-    # Step 1 - Convolution
-    classifier.add(
-        layers.Conv2D(32, (3, 3), input_shape=(64, 64, 3), activation="relu")
-    )
-
-    # Step 2 - Pooling
-    classifier.add(layers.MaxPooling2D(pool_size=(2, 2)))
-
-    # Adding a second convolutional layer
-    classifier.add(layers.Conv2D(32, (3, 3), activation="relu"))
-    classifier.add(layers.MaxPooling2D(pool_size=(2, 2)))
-
-    # Step 3 - Flattening
-    classifier.add(layers.Flatten())
-
-    # Step 4 - Full connection
-    classifier.add(layers.Dense(units=128, activation="relu"))
-    classifier.add(layers.Dense(units=1, activation="sigmoid"))
-
-    # Compiling the CNN
-    classifier.compile(
-        optimizer="adam", loss="binary_crossentropy", metrics=["accuracy"]
-    )
-
-    # Part 2 - Fitting the CNN to the images
-
-    # Load Trained model weights
-
-    # from keras.models import load_model
-    # regressor=load_model('cnn.h5')
-
-    train_datagen = tf.keras.preprocessing.image.ImageDataGenerator(
-        rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True
-    )
-
-    test_datagen = tf.keras.preprocessing.image.ImageDataGenerator(rescale=1.0 / 255)
-
-    training_set = train_datagen.flow_from_directory(
-        "dataset/training_set", target_size=(64, 64), batch_size=32, class_mode="binary"
-    )
-
-    test_set = test_datagen.flow_from_directory(
-        "dataset/test_set", target_size=(64, 64), batch_size=32, class_mode="binary"
-    )
-
-    classifier.fit_generator(
-        training_set, steps_per_epoch=5, epochs=30, validation_data=test_set
-    )
-
-    classifier.save("cnn.h5")
-
-    # Part 3 - Making new predictions
-
-    test_image = tf.keras.preprocessing.image.load_img(
-        "dataset/single_prediction/image.png", target_size=(64, 64)
-    )
-    test_image = tf.keras.preprocessing.image.img_to_array(test_image)
-    test_image = np.expand_dims(test_image, axis=0)
-    result = classifier.predict(test_image)
-    training_set.class_indices
-    if result[0][0] == 0:
-        prediction = "Normal"
-    if result[0][0] == 1:
-        prediction = "Abnormality detected"

From 0e2880129e52eda2c8d97bee2c81f6c1a1a72f30 Mon Sep 17 00:00:00 2001
From: Harshit Agarwal <43147421+9harshit@users.noreply.github.com>
Date: Thu, 24 Jun 2021 10:22:40 +0530
Subject: [PATCH 32/32] Add files via upload

---
 computer_vision/cnn_classification.py | 98 +++++++++++++++++++++++++++
 1 file changed, 98 insertions(+)
 create mode 100644 computer_vision/cnn_classification.py

diff --git a/computer_vision/cnn_classification.py b/computer_vision/cnn_classification.py
new file mode 100644
index 000000000000..6d4f19639c24
--- /dev/null
+++ b/computer_vision/cnn_classification.py
@@ -0,0 +1,98 @@
+"""
+Convolutional Neural Network
+
+Objective : To train a CNN model detect if TB is present in Lung X-ray or not.
+
+Resources CNN Theory :
+    https://en.wikipedia.org/wiki/Convolutional_neural_network
+Resources Tensorflow : https://www.tensorflow.org/tutorials/images/cnn
+
+Download dataset from :
+https://lhncbc.nlm.nih.gov/LHC-publications/pubs/TuberculosisChestXrayImageDataSets.html
+
+1. Download the dataset folder and create two folder training set and test set
+in the parent dataste folder
+2. Move 30-40 image from both TB positive and TB Negative folder
+in the test set folder
+3. The labels of the iamges will be extracted from the folder name
+the image is present in.
+
+"""
+
+# Part 1 - Building the CNN
+
+import numpy as np
+
+# Importing the Keras libraries and packages
+import tensorflow as tf
+from tensorflow.keras import layers, models
+
+if __name__ == "__main__":
+
+    # Initialising the CNN
+    classifier = models.Sequential()
+
+    # Step 1 - Convolution
+    classifier.add(
+        layers.Conv2D(32, (3, 3), input_shape=(64, 64, 3), activation="relu")
+    )
+
+    # Step 2 - Pooling
+    classifier.add(layers.MaxPooling2D(pool_size=(2, 2)))
+
+    # Adding a second convolutional layer
+    classifier.add(layers.Conv2D(32, (3, 3), activation="relu"))
+    classifier.add(layers.MaxPooling2D(pool_size=(2, 2)))
+
+    # Step 3 - Flattening
+    classifier.add(layers.Flatten())
+
+    # Step 4 - Full connection
+    classifier.add(layers.Dense(units=128, activation="relu"))
+    classifier.add(layers.Dense(units=1, activation="sigmoid"))
+
+    # Compiling the CNN
+    classifier.compile(
+        optimizer="adam", loss="binary_crossentropy", metrics=["accuracy"]
+    )
+
+    # Part 2 - Fitting the CNN to the images
+
+    # Load Trained model weights
+
+    # from keras.models import load_model
+    # regressor=load_model('cnn.h5')
+
+    train_datagen = tf.keras.preprocessing.image.ImageDataGenerator(
+        rescale=1.0 / 255, shear_range=0.2, zoom_range=0.2, horizontal_flip=True
+    )
+
+    test_datagen = tf.keras.preprocessing.image.ImageDataGenerator(rescale=1.0 / 255)
+
+    training_set = train_datagen.flow_from_directory(
+        "dataset/training_set", target_size=(64, 64), batch_size=32, class_mode="binary"
+    )
+
+    test_set = test_datagen.flow_from_directory(
+        "dataset/test_set", target_size=(64, 64), batch_size=32, class_mode="binary"
+    )
+
+    classifier.fit_generator(
+        training_set, steps_per_epoch=5, epochs=30, validation_data=test_set
+    )
+
+    classifier.save("cnn.h5")
+
+    # Part 3 - Making new predictions
+
+    test_image = tf.keras.preprocessing.image.load_img(
+        "dataset/single_prediction/image.png", target_size=(64, 64)
+    )
+    test_image = tf.keras.preprocessing.image.img_to_array(test_image)
+    test_image = np.expand_dims(test_image, axis=0)
+    result = classifier.predict(test_image)
+    training_set.class_indices
+    if result[0][0] == 0:
+        prediction = "Normal"
+    if result[0][0] == 1:
+        prediction = "Abnormality detected"