path
stringlengths 13
17
| screenshot_names
sequencelengths 1
873
| code
stringlengths 0
40.4k
| cell_type
stringclasses 1
value |
|---|---|---|---|
73079642/cell_13 | [
"text_plain_output_1.png",
"image_output_1.png"
] | from keras.layers import Input,Conv2D,MaxPool2D, UpSampling2D,Dense, Dropout
from keras.models import Model
from tensorflow.keras import losses
inputs = Input(shape=(28, 28, 1))
x = Conv2D(32, 3, activation='relu', padding='same')(inputs)
x = MaxPool2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
encoded = MaxPool2D()(x)
x = Conv2D(32, 3, activation='relu', padding='same')(encoded)
x = UpSampling2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
x = UpSampling2D()(x)
decoded = Conv2D(1, 3, activation='sigmoid', padding='same')(x)
from tensorflow.keras import losses
autoencoder1 = Model(inputs, decoded)
autoencoder1.compile(optimizer='adam', loss=losses.binary_crossentropy)
autoencoder1.summary() | code |
73079642/cell_33 | [
"image_output_1.png"
] | from keras.datasets import mnist, cifar10
from keras.layers import Conv2DTranspose, BatchNormalization, add, LeakyReLU
from keras.layers import Input,Conv2D,MaxPool2D, UpSampling2D,Dense, Dropout
from keras.models import Model
from keras.optimizers import Adam
from tensorflow.keras import layers, losses
import matplotlib.pyplot as plt
import numpy as np
import tensorflow as tf
def preprocess(array1, array2, channel):
"""
Normalizes/scales [0,1], divinding by the supplied array and reshapes
it into the appropriate format.
"""
if channel == 1:
ar1 = array1.astype('float32').reshape([-1, 28, 28, 1]) / 255.0
ar2 = array2.astype('float32').reshape([-1, 28, 28, 1]) / 255.0
else:
ar1 = array1.astype('float32').reshape([-1, 32, 32, 3]) / 255
ar2 = array2.astype('float32').reshape([-1, 32, 32, 3]) / 255
return (ar1, ar2)
def noise(a1, a2, channel):
"""
Adds random noise to each image in the supplied array.
"""
if channel == 1:
noise_factor = 0.2
noisy_arr1 = a1 + noise_factor * np.random.normal(0.0, 1.0, size=a1.shape)
noisy_arr2 = a2 + noise_factor * np.random.normal(0.0, 1.0, size=a2.shape)
else:
noi = 0.1
noisy_arr1 = a1 + noi * np.random.normal(0.0, 1.0, size=a1.shape)
noisy_arr2 = a2 + noi * np.random.normal(0.0, 1.0, size=a2.shape)
ab1 = np.clip(noisy_arr1, 0, 1)
ab2 = np.clip(noisy_arr2, 0, 1)
return (ab1, ab2)
# Visualization for mnist, cifar10, noisy, denoised/predictions data
def display(rows, cols, a, b, check=False ):
'''rows: defining no. of rows in figure
cols: defining no. of colums in figure
a: train images without noise or noisy_image while test
prediction
b: train images with noise or denoised_image based while test
prediction
check: default False for 32*32 cifar10, true for 28*28
mnist dataset and any predictions
'''
# defining a figure
f = plt.figure(figsize=(2*cols,2*rows*2))
for i in range(rows):
for j in range(cols):
# adding subplot to figure on each iteration
f.add_subplot(rows*2,cols, (2*i*cols)+(j+1))
if check:
plt.imshow(a[i*cols + j].reshape([28,28]),cmap="Blues")
else:
plt.imshow(a[i*cols + j])
plt.axis("off")
for j in range(cols):
# adding subplot to figure on each iteration
f.add_subplot(rows*2,cols,((2*i+1)*cols)+(j+1))
if check:
plt.imshow(b[i*cols + j].reshape([28,28]),cmap="Blues")
else:
plt.imshow(b[i*cols + j])
plt.axis("off")
plt.axis("off")
#f.suptitle("Sample Training Data",fontsize=18)
plt.savefig("ss.png")
plt.show()
(train_data, _), (test_data, _) = mnist.load_data()
channel = 1
train_data, test_data = preprocess(train_data, test_data, channel)
noisy_train_data, noisy_test_data = noise(train_data, test_data, channel)
(cifar_train, _), (cifar_test, _) = cifar10.load_data()
rows = 2
cols = 3
channel = 3
cifar_train, cifar_test = preprocess(cifar_train, cifar_test, channel)
cifar_train_noise, cifar_test_noise = noise(cifar_train, cifar_test, channel)
inputs = Input(shape=(28, 28, 1))
x = Conv2D(32, 3, activation='relu', padding='same')(inputs)
x = MaxPool2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
encoded = MaxPool2D()(x)
x = Conv2D(32, 3, activation='relu', padding='same')(encoded)
x = UpSampling2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
x = UpSampling2D()(x)
decoded = Conv2D(1, 3, activation='sigmoid', padding='same')(x)
from tensorflow.keras import layers, losses
class Denoise(Model):
"""__init__ constructor in OOP
This method called when an object is created from the class and
it allow the class to initialize the attributes of a class.
super()function used to give access to
methods and properties of a parent or sibling class
"""
def __init__(self):
super(Denoise, self).__init__()
self.encoder = tf.keras.Sequential([layers.Input(shape=(28, 28, 1)), layers.Conv2D(16, (3, 3), activation='relu', padding='same', strides=2), layers.Conv2D(8, (3, 3), activation='relu', padding='same', strides=2)])
self.decoder = tf.keras.Sequential([layers.Conv2DTranspose(8, kernel_size=3, strides=2, activation='relu', padding='same'), layers.Conv2DTranspose(16, kernel_size=3, strides=2, activation='relu', padding='same'), layers.Conv2D(1, kernel_size=(3, 3), activation='sigmoid', padding='same')])
def call(self, x):
encoded = self.encoder(x)
decoded = self.decoder(encoded)
return decoded
autoencoder = Denoise()
size = 32
channel = 3
inputs = Input(shape=(size, size, channel))
x = Conv2D(32, 3, activation='relu', padding='same')(inputs)
x = BatchNormalization()(x)
x = MaxPool2D()(x)
x = Dropout(0.5)(x)
skip = Conv2D(32, 3, padding='same')(x)
x = LeakyReLU()(skip)
x = BatchNormalization()(x)
x = MaxPool2D()(x)
x = Dropout(0.5)(x)
x = Conv2D(64, 3, activation='relu', padding='same')(x)
x = BatchNormalization()(x)
encoded = MaxPool2D()(x)
x = Conv2DTranspose(64, 3, activation='relu', strides=(2, 2), padding='same')(encoded)
x = BatchNormalization()(x)
x = Dropout(0.5)(x)
x = Conv2DTranspose(32, 3, activation='relu', strides=(2, 2), padding='same')(x)
x = BatchNormalization()(x)
x = Dropout(0.5)(x)
x = Conv2DTranspose(32, 3, padding='same')(x)
x = add([x, skip])
x = LeakyReLU()(x)
x = BatchNormalization()(x)
decoded = Conv2DTranspose(3, 3, activation='sigmoid', strides=(2, 2), padding='same')(x)
autoencoder2 = Model(inputs, decoded)
autoencoder2.compile(optimizer=Adam(lr=0.0001), loss='binary_crossentropy')
autoencoder2.summary()
epochs = 25
batch_size = 256
history2 = autoencoder2.fit(cifar_train_noise, cifar_train, epochs=epochs, batch_size=batch_size, shuffle=True, validation_data=(cifar_test_noise, cifar_test)) | code |
73079642/cell_6 | [
"image_output_1.png"
] | from keras.datasets import mnist, cifar10
import matplotlib.pyplot as plt
import numpy as np
def preprocess(array1, array2, channel):
"""
Normalizes/scales [0,1], divinding by the supplied array and reshapes
it into the appropriate format.
"""
if channel == 1:
ar1 = array1.astype('float32').reshape([-1, 28, 28, 1]) / 255.0
ar2 = array2.astype('float32').reshape([-1, 28, 28, 1]) / 255.0
else:
ar1 = array1.astype('float32').reshape([-1, 32, 32, 3]) / 255
ar2 = array2.astype('float32').reshape([-1, 32, 32, 3]) / 255
return (ar1, ar2)
def noise(a1, a2, channel):
"""
Adds random noise to each image in the supplied array.
"""
if channel == 1:
noise_factor = 0.2
noisy_arr1 = a1 + noise_factor * np.random.normal(0.0, 1.0, size=a1.shape)
noisy_arr2 = a2 + noise_factor * np.random.normal(0.0, 1.0, size=a2.shape)
else:
noi = 0.1
noisy_arr1 = a1 + noi * np.random.normal(0.0, 1.0, size=a1.shape)
noisy_arr2 = a2 + noi * np.random.normal(0.0, 1.0, size=a2.shape)
ab1 = np.clip(noisy_arr1, 0, 1)
ab2 = np.clip(noisy_arr2, 0, 1)
return (ab1, ab2)
# Visualization for mnist, cifar10, noisy, denoised/predictions data
def display(rows, cols, a, b, check=False ):
'''rows: defining no. of rows in figure
cols: defining no. of colums in figure
a: train images without noise or noisy_image while test
prediction
b: train images with noise or denoised_image based while test
prediction
check: default False for 32*32 cifar10, true for 28*28
mnist dataset and any predictions
'''
# defining a figure
f = plt.figure(figsize=(2*cols,2*rows*2))
for i in range(rows):
for j in range(cols):
# adding subplot to figure on each iteration
f.add_subplot(rows*2,cols, (2*i*cols)+(j+1))
if check:
plt.imshow(a[i*cols + j].reshape([28,28]),cmap="Blues")
else:
plt.imshow(a[i*cols + j])
plt.axis("off")
for j in range(cols):
# adding subplot to figure on each iteration
f.add_subplot(rows*2,cols,((2*i+1)*cols)+(j+1))
if check:
plt.imshow(b[i*cols + j].reshape([28,28]),cmap="Blues")
else:
plt.imshow(b[i*cols + j])
plt.axis("off")
plt.axis("off")
#f.suptitle("Sample Training Data",fontsize=18)
plt.savefig("ss.png")
plt.show()
(train_data, _), (test_data, _) = mnist.load_data()
channel = 1
train_data, test_data = preprocess(train_data, test_data, channel)
noisy_train_data, noisy_test_data = noise(train_data, test_data, channel)
display(2, 3, train_data, noisy_train_data, check=True) | code |
73079642/cell_29 | [
"text_plain_output_1.png"
] | from keras.layers import Input,Conv2D,MaxPool2D, UpSampling2D,Dense, Dropout
from keras.models import Model
from keras.models import load_model
from tensorflow.keras import layers, losses
from tensorflow.keras import losses
import matplotlib.pyplot as plt
import numpy as np
import tensorflow as tf
def noise(a1, a2, channel):
"""
Adds random noise to each image in the supplied array.
"""
if channel == 1:
noise_factor = 0.2
noisy_arr1 = a1 + noise_factor * np.random.normal(0.0, 1.0, size=a1.shape)
noisy_arr2 = a2 + noise_factor * np.random.normal(0.0, 1.0, size=a2.shape)
else:
noi = 0.1
noisy_arr1 = a1 + noi * np.random.normal(0.0, 1.0, size=a1.shape)
noisy_arr2 = a2 + noi * np.random.normal(0.0, 1.0, size=a2.shape)
ab1 = np.clip(noisy_arr1, 0, 1)
ab2 = np.clip(noisy_arr2, 0, 1)
return (ab1, ab2)
# Visualization for mnist, cifar10, noisy, denoised/predictions data
def display(rows, cols, a, b, check=False ):
'''rows: defining no. of rows in figure
cols: defining no. of colums in figure
a: train images without noise or noisy_image while test
prediction
b: train images with noise or denoised_image based while test
prediction
check: default False for 32*32 cifar10, true for 28*28
mnist dataset and any predictions
'''
# defining a figure
f = plt.figure(figsize=(2*cols,2*rows*2))
for i in range(rows):
for j in range(cols):
# adding subplot to figure on each iteration
f.add_subplot(rows*2,cols, (2*i*cols)+(j+1))
if check:
plt.imshow(a[i*cols + j].reshape([28,28]),cmap="Blues")
else:
plt.imshow(a[i*cols + j])
plt.axis("off")
for j in range(cols):
# adding subplot to figure on each iteration
f.add_subplot(rows*2,cols,((2*i+1)*cols)+(j+1))
if check:
plt.imshow(b[i*cols + j].reshape([28,28]),cmap="Blues")
else:
plt.imshow(b[i*cols + j])
plt.axis("off")
plt.axis("off")
#f.suptitle("Sample Training Data",fontsize=18)
plt.savefig("ss.png")
plt.show()
inputs = Input(shape=(28, 28, 1))
x = Conv2D(32, 3, activation='relu', padding='same')(inputs)
x = MaxPool2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
encoded = MaxPool2D()(x)
x = Conv2D(32, 3, activation='relu', padding='same')(encoded)
x = UpSampling2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
x = UpSampling2D()(x)
decoded = Conv2D(1, 3, activation='sigmoid', padding='same')(x)
from tensorflow.keras import losses
autoencoder1 = Model(inputs, decoded)
autoencoder1.compile(optimizer='adam', loss=losses.binary_crossentropy)
autoencoder1.summary()
history1 = autoencoder1.fit(noisy_train_data, train_data, epochs=50, batch_size=256, shuffle=True, validation_data=(noisy_test_data, test_data))
autoencoder1.save('autoencoder_model1.h5')
from keras.models import load_model
model1 = load_model('autoencoder_model1.h5')
num_imgs = 45
rand = np.random.randint(1, 100)
test_images = noisy_test_data[rand:rand + num_imgs]
test_denoised = model1.predict(test_images)
from tensorflow.keras import layers, losses
class Denoise(Model):
"""__init__ constructor in OOP
This method called when an object is created from the class and
it allow the class to initialize the attributes of a class.
super()function used to give access to
methods and properties of a parent or sibling class
"""
def __init__(self):
super(Denoise, self).__init__()
self.encoder = tf.keras.Sequential([layers.Input(shape=(28, 28, 1)), layers.Conv2D(16, (3, 3), activation='relu', padding='same', strides=2), layers.Conv2D(8, (3, 3), activation='relu', padding='same', strides=2)])
self.decoder = tf.keras.Sequential([layers.Conv2DTranspose(8, kernel_size=3, strides=2, activation='relu', padding='same'), layers.Conv2DTranspose(16, kernel_size=3, strides=2, activation='relu', padding='same'), layers.Conv2D(1, kernel_size=(3, 3), activation='sigmoid', padding='same')])
def call(self, x):
encoded = self.encoder(x)
decoded = self.decoder(encoded)
return decoded
autoencoder = Denoise()
autoencoder.compile(optimizer='rmsprop', loss=losses.MeanSquaredError())
history = autoencoder.fit(noisy_train_data, train_data, epochs=10, batch_size=256, shuffle=True, validation_data=(noisy_test_data, test_data))
num_imgs = 45
rand = np.random.randint(1, 100)
test_images = noisy_test_data[rand:rand + num_imgs]
test_denoised = autoencoder.predict(test_images)
display(2, 4, test_images, test_denoised, check=True) | code |
73079642/cell_1 | [
"application_vnd.jupyter.stderr_output_1.png"
] | import numpy as np
import matplotlib.pyplot as plt
from keras.layers import Input, Conv2D, MaxPool2D, UpSampling2D, Dense, Dropout
import tensorflow as tf
from keras.models import Model
from keras.datasets import mnist, cifar10 | code |
73079642/cell_7 | [
"image_output_1.png"
] | from keras.datasets import mnist, cifar10
import matplotlib.pyplot as plt
import numpy as np
def preprocess(array1, array2, channel):
"""
Normalizes/scales [0,1], divinding by the supplied array and reshapes
it into the appropriate format.
"""
if channel == 1:
ar1 = array1.astype('float32').reshape([-1, 28, 28, 1]) / 255.0
ar2 = array2.astype('float32').reshape([-1, 28, 28, 1]) / 255.0
else:
ar1 = array1.astype('float32').reshape([-1, 32, 32, 3]) / 255
ar2 = array2.astype('float32').reshape([-1, 32, 32, 3]) / 255
return (ar1, ar2)
def noise(a1, a2, channel):
"""
Adds random noise to each image in the supplied array.
"""
if channel == 1:
noise_factor = 0.2
noisy_arr1 = a1 + noise_factor * np.random.normal(0.0, 1.0, size=a1.shape)
noisy_arr2 = a2 + noise_factor * np.random.normal(0.0, 1.0, size=a2.shape)
else:
noi = 0.1
noisy_arr1 = a1 + noi * np.random.normal(0.0, 1.0, size=a1.shape)
noisy_arr2 = a2 + noi * np.random.normal(0.0, 1.0, size=a2.shape)
ab1 = np.clip(noisy_arr1, 0, 1)
ab2 = np.clip(noisy_arr2, 0, 1)
return (ab1, ab2)
# Visualization for mnist, cifar10, noisy, denoised/predictions data
def display(rows, cols, a, b, check=False ):
'''rows: defining no. of rows in figure
cols: defining no. of colums in figure
a: train images without noise or noisy_image while test
prediction
b: train images with noise or denoised_image based while test
prediction
check: default False for 32*32 cifar10, true for 28*28
mnist dataset and any predictions
'''
# defining a figure
f = plt.figure(figsize=(2*cols,2*rows*2))
for i in range(rows):
for j in range(cols):
# adding subplot to figure on each iteration
f.add_subplot(rows*2,cols, (2*i*cols)+(j+1))
if check:
plt.imshow(a[i*cols + j].reshape([28,28]),cmap="Blues")
else:
plt.imshow(a[i*cols + j])
plt.axis("off")
for j in range(cols):
# adding subplot to figure on each iteration
f.add_subplot(rows*2,cols,((2*i+1)*cols)+(j+1))
if check:
plt.imshow(b[i*cols + j].reshape([28,28]),cmap="Blues")
else:
plt.imshow(b[i*cols + j])
plt.axis("off")
plt.axis("off")
#f.suptitle("Sample Training Data",fontsize=18)
plt.savefig("ss.png")
plt.show()
(train_data, _), (test_data, _) = mnist.load_data()
channel = 1
train_data, test_data = preprocess(train_data, test_data, channel)
noisy_train_data, noisy_test_data = noise(train_data, test_data, channel)
(cifar_train, _), (cifar_test, _) = cifar10.load_data()
rows = 2
cols = 3
channel = 3
cifar_train, cifar_test = preprocess(cifar_train, cifar_test, channel)
cifar_train_noise, cifar_test_noise = noise(cifar_train, cifar_test, channel)
display(rows, cols, cifar_train, cifar_train_noise) | code |
73079642/cell_32 | [
"image_output_1.png"
] | from keras.datasets import mnist, cifar10
from keras.layers import Conv2DTranspose, BatchNormalization, add, LeakyReLU
from keras.layers import Input,Conv2D,MaxPool2D, UpSampling2D,Dense, Dropout
from keras.models import Model
from keras.optimizers import Adam
from tensorflow.keras import layers, losses
import matplotlib.pyplot as plt
import numpy as np
import tensorflow as tf
def preprocess(array1, array2, channel):
"""
Normalizes/scales [0,1], divinding by the supplied array and reshapes
it into the appropriate format.
"""
if channel == 1:
ar1 = array1.astype('float32').reshape([-1, 28, 28, 1]) / 255.0
ar2 = array2.astype('float32').reshape([-1, 28, 28, 1]) / 255.0
else:
ar1 = array1.astype('float32').reshape([-1, 32, 32, 3]) / 255
ar2 = array2.astype('float32').reshape([-1, 32, 32, 3]) / 255
return (ar1, ar2)
def noise(a1, a2, channel):
"""
Adds random noise to each image in the supplied array.
"""
if channel == 1:
noise_factor = 0.2
noisy_arr1 = a1 + noise_factor * np.random.normal(0.0, 1.0, size=a1.shape)
noisy_arr2 = a2 + noise_factor * np.random.normal(0.0, 1.0, size=a2.shape)
else:
noi = 0.1
noisy_arr1 = a1 + noi * np.random.normal(0.0, 1.0, size=a1.shape)
noisy_arr2 = a2 + noi * np.random.normal(0.0, 1.0, size=a2.shape)
ab1 = np.clip(noisy_arr1, 0, 1)
ab2 = np.clip(noisy_arr2, 0, 1)
return (ab1, ab2)
# Visualization for mnist, cifar10, noisy, denoised/predictions data
def display(rows, cols, a, b, check=False ):
'''rows: defining no. of rows in figure
cols: defining no. of colums in figure
a: train images without noise or noisy_image while test
prediction
b: train images with noise or denoised_image based while test
prediction
check: default False for 32*32 cifar10, true for 28*28
mnist dataset and any predictions
'''
# defining a figure
f = plt.figure(figsize=(2*cols,2*rows*2))
for i in range(rows):
for j in range(cols):
# adding subplot to figure on each iteration
f.add_subplot(rows*2,cols, (2*i*cols)+(j+1))
if check:
plt.imshow(a[i*cols + j].reshape([28,28]),cmap="Blues")
else:
plt.imshow(a[i*cols + j])
plt.axis("off")
for j in range(cols):
# adding subplot to figure on each iteration
f.add_subplot(rows*2,cols,((2*i+1)*cols)+(j+1))
if check:
plt.imshow(b[i*cols + j].reshape([28,28]),cmap="Blues")
else:
plt.imshow(b[i*cols + j])
plt.axis("off")
plt.axis("off")
#f.suptitle("Sample Training Data",fontsize=18)
plt.savefig("ss.png")
plt.show()
(train_data, _), (test_data, _) = mnist.load_data()
channel = 1
train_data, test_data = preprocess(train_data, test_data, channel)
noisy_train_data, noisy_test_data = noise(train_data, test_data, channel)
(cifar_train, _), (cifar_test, _) = cifar10.load_data()
rows = 2
cols = 3
channel = 3
cifar_train, cifar_test = preprocess(cifar_train, cifar_test, channel)
cifar_train_noise, cifar_test_noise = noise(cifar_train, cifar_test, channel)
inputs = Input(shape=(28, 28, 1))
x = Conv2D(32, 3, activation='relu', padding='same')(inputs)
x = MaxPool2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
encoded = MaxPool2D()(x)
x = Conv2D(32, 3, activation='relu', padding='same')(encoded)
x = UpSampling2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
x = UpSampling2D()(x)
decoded = Conv2D(1, 3, activation='sigmoid', padding='same')(x)
from tensorflow.keras import layers, losses
class Denoise(Model):
"""__init__ constructor in OOP
This method called when an object is created from the class and
it allow the class to initialize the attributes of a class.
super()function used to give access to
methods and properties of a parent or sibling class
"""
def __init__(self):
super(Denoise, self).__init__()
self.encoder = tf.keras.Sequential([layers.Input(shape=(28, 28, 1)), layers.Conv2D(16, (3, 3), activation='relu', padding='same', strides=2), layers.Conv2D(8, (3, 3), activation='relu', padding='same', strides=2)])
self.decoder = tf.keras.Sequential([layers.Conv2DTranspose(8, kernel_size=3, strides=2, activation='relu', padding='same'), layers.Conv2DTranspose(16, kernel_size=3, strides=2, activation='relu', padding='same'), layers.Conv2D(1, kernel_size=(3, 3), activation='sigmoid', padding='same')])
def call(self, x):
encoded = self.encoder(x)
decoded = self.decoder(encoded)
return decoded
autoencoder = Denoise()
size = 32
channel = 3
inputs = Input(shape=(size, size, channel))
x = Conv2D(32, 3, activation='relu', padding='same')(inputs)
x = BatchNormalization()(x)
x = MaxPool2D()(x)
x = Dropout(0.5)(x)
skip = Conv2D(32, 3, padding='same')(x)
x = LeakyReLU()(skip)
x = BatchNormalization()(x)
x = MaxPool2D()(x)
x = Dropout(0.5)(x)
x = Conv2D(64, 3, activation='relu', padding='same')(x)
x = BatchNormalization()(x)
encoded = MaxPool2D()(x)
x = Conv2DTranspose(64, 3, activation='relu', strides=(2, 2), padding='same')(encoded)
x = BatchNormalization()(x)
x = Dropout(0.5)(x)
x = Conv2DTranspose(32, 3, activation='relu', strides=(2, 2), padding='same')(x)
x = BatchNormalization()(x)
x = Dropout(0.5)(x)
x = Conv2DTranspose(32, 3, padding='same')(x)
x = add([x, skip])
x = LeakyReLU()(x)
x = BatchNormalization()(x)
decoded = Conv2DTranspose(3, 3, activation='sigmoid', strides=(2, 2), padding='same')(x)
autoencoder2 = Model(inputs, decoded)
autoencoder2.compile(optimizer=Adam(lr=0.0001), loss='binary_crossentropy')
autoencoder2.summary() | code |
73079642/cell_28 | [
"image_output_1.png"
] | from keras.layers import Input,Conv2D,MaxPool2D, UpSampling2D,Dense, Dropout
from keras.models import Model
from tensorflow.keras import layers, losses
from tensorflow.keras import losses
import matplotlib.pyplot as plt
import tensorflow as tf
# Visualization for mnist, cifar10, noisy, denoised/predictions data
def display(rows, cols, a, b, check=False ):
'''rows: defining no. of rows in figure
cols: defining no. of colums in figure
a: train images without noise or noisy_image while test
prediction
b: train images with noise or denoised_image based while test
prediction
check: default False for 32*32 cifar10, true for 28*28
mnist dataset and any predictions
'''
# defining a figure
f = plt.figure(figsize=(2*cols,2*rows*2))
for i in range(rows):
for j in range(cols):
# adding subplot to figure on each iteration
f.add_subplot(rows*2,cols, (2*i*cols)+(j+1))
if check:
plt.imshow(a[i*cols + j].reshape([28,28]),cmap="Blues")
else:
plt.imshow(a[i*cols + j])
plt.axis("off")
for j in range(cols):
# adding subplot to figure on each iteration
f.add_subplot(rows*2,cols,((2*i+1)*cols)+(j+1))
if check:
plt.imshow(b[i*cols + j].reshape([28,28]),cmap="Blues")
else:
plt.imshow(b[i*cols + j])
plt.axis("off")
plt.axis("off")
#f.suptitle("Sample Training Data",fontsize=18)
plt.savefig("ss.png")
plt.show()
inputs = Input(shape=(28, 28, 1))
x = Conv2D(32, 3, activation='relu', padding='same')(inputs)
x = MaxPool2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
encoded = MaxPool2D()(x)
x = Conv2D(32, 3, activation='relu', padding='same')(encoded)
x = UpSampling2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
x = UpSampling2D()(x)
decoded = Conv2D(1, 3, activation='sigmoid', padding='same')(x)
from tensorflow.keras import losses
autoencoder1 = Model(inputs, decoded)
autoencoder1.compile(optimizer='adam', loss=losses.binary_crossentropy)
autoencoder1.summary()
history1 = autoencoder1.fit(noisy_train_data, train_data, epochs=50, batch_size=256, shuffle=True, validation_data=(noisy_test_data, test_data))
def plot_diag(history):
f = plt.figure(figsize=(10,7))
f.add_subplot()
#Adding Subplot
plt.plot(history.epoch, history.history['loss'], label = "loss") # Loss curve for training set
plt.plot(history.epoch, history.history['val_loss'], label = "val_loss") # Loss curve for validation set
plt.title("Loss Curve",fontsize=18)
plt.xlabel("Epochs",fontsize=15)
plt.ylabel("Loss",fontsize=15)
plt.grid(alpha=0.3)
plt.legend()
plt.savefig("Loss_curve.png")
plt.show()
plot_diag(history1)
from tensorflow.keras import layers, losses
class Denoise(Model):
"""__init__ constructor in OOP
This method called when an object is created from the class and
it allow the class to initialize the attributes of a class.
super()function used to give access to
methods and properties of a parent or sibling class
"""
def __init__(self):
super(Denoise, self).__init__()
self.encoder = tf.keras.Sequential([layers.Input(shape=(28, 28, 1)), layers.Conv2D(16, (3, 3), activation='relu', padding='same', strides=2), layers.Conv2D(8, (3, 3), activation='relu', padding='same', strides=2)])
self.decoder = tf.keras.Sequential([layers.Conv2DTranspose(8, kernel_size=3, strides=2, activation='relu', padding='same'), layers.Conv2DTranspose(16, kernel_size=3, strides=2, activation='relu', padding='same'), layers.Conv2D(1, kernel_size=(3, 3), activation='sigmoid', padding='same')])
def call(self, x):
encoded = self.encoder(x)
decoded = self.decoder(encoded)
return decoded
autoencoder = Denoise()
autoencoder.compile(optimizer='rmsprop', loss=losses.MeanSquaredError())
history = autoencoder.fit(noisy_train_data, train_data, epochs=10, batch_size=256, shuffle=True, validation_data=(noisy_test_data, test_data))
plot_diag(history) | code |
73079642/cell_15 | [
"text_plain_output_1.png",
"image_output_1.png"
] | from keras.layers import Input,Conv2D,MaxPool2D, UpSampling2D,Dense, Dropout
from keras.models import Model
from tensorflow.keras import losses
inputs = Input(shape=(28, 28, 1))
x = Conv2D(32, 3, activation='relu', padding='same')(inputs)
x = MaxPool2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
encoded = MaxPool2D()(x)
x = Conv2D(32, 3, activation='relu', padding='same')(encoded)
x = UpSampling2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
x = UpSampling2D()(x)
decoded = Conv2D(1, 3, activation='sigmoid', padding='same')(x)
from tensorflow.keras import losses
autoencoder1 = Model(inputs, decoded)
autoencoder1.compile(optimizer='adam', loss=losses.binary_crossentropy)
autoencoder1.summary()
history1 = autoencoder1.fit(noisy_train_data, train_data, epochs=50, batch_size=256, shuffle=True, validation_data=(noisy_test_data, test_data)) | code |
73079642/cell_17 | [
"text_plain_output_1.png"
] | from keras.layers import Input,Conv2D,MaxPool2D, UpSampling2D,Dense, Dropout
from keras.models import Model
from tensorflow.keras import losses
import matplotlib.pyplot as plt
# Visualization for mnist, cifar10, noisy, denoised/predictions data
def display(rows, cols, a, b, check=False ):
'''rows: defining no. of rows in figure
cols: defining no. of colums in figure
a: train images without noise or noisy_image while test
prediction
b: train images with noise or denoised_image based while test
prediction
check: default False for 32*32 cifar10, true for 28*28
mnist dataset and any predictions
'''
# defining a figure
f = plt.figure(figsize=(2*cols,2*rows*2))
for i in range(rows):
for j in range(cols):
# adding subplot to figure on each iteration
f.add_subplot(rows*2,cols, (2*i*cols)+(j+1))
if check:
plt.imshow(a[i*cols + j].reshape([28,28]),cmap="Blues")
else:
plt.imshow(a[i*cols + j])
plt.axis("off")
for j in range(cols):
# adding subplot to figure on each iteration
f.add_subplot(rows*2,cols,((2*i+1)*cols)+(j+1))
if check:
plt.imshow(b[i*cols + j].reshape([28,28]),cmap="Blues")
else:
plt.imshow(b[i*cols + j])
plt.axis("off")
plt.axis("off")
#f.suptitle("Sample Training Data",fontsize=18)
plt.savefig("ss.png")
plt.show()
inputs = Input(shape=(28, 28, 1))
x = Conv2D(32, 3, activation='relu', padding='same')(inputs)
x = MaxPool2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
encoded = MaxPool2D()(x)
x = Conv2D(32, 3, activation='relu', padding='same')(encoded)
x = UpSampling2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
x = UpSampling2D()(x)
decoded = Conv2D(1, 3, activation='sigmoid', padding='same')(x)
from tensorflow.keras import losses
autoencoder1 = Model(inputs, decoded)
autoencoder1.compile(optimizer='adam', loss=losses.binary_crossentropy)
autoencoder1.summary()
history1 = autoencoder1.fit(noisy_train_data, train_data, epochs=50, batch_size=256, shuffle=True, validation_data=(noisy_test_data, test_data))
def plot_diag(history):
f = plt.figure(figsize=(10, 7))
f.add_subplot()
plt.plot(history.epoch, history.history['loss'], label='loss')
plt.plot(history.epoch, history.history['val_loss'], label='val_loss')
plt.title('Loss Curve', fontsize=18)
plt.xlabel('Epochs', fontsize=15)
plt.ylabel('Loss', fontsize=15)
plt.grid(alpha=0.3)
plt.legend()
plt.savefig('Loss_curve.png')
plt.show()
plot_diag(history1) | code |
73079642/cell_35 | [
"text_plain_output_1.png"
] | from keras.datasets import mnist, cifar10
from keras.layers import Conv2DTranspose, BatchNormalization, add, LeakyReLU
from keras.layers import Input,Conv2D,MaxPool2D, UpSampling2D,Dense, Dropout
from keras.models import Model
from keras.optimizers import Adam
from tensorflow.keras import layers, losses
from tensorflow.keras import losses
import matplotlib.pyplot as plt
import numpy as np
import tensorflow as tf
def preprocess(array1, array2, channel):
"""
Normalizes/scales [0,1], divinding by the supplied array and reshapes
it into the appropriate format.
"""
if channel == 1:
ar1 = array1.astype('float32').reshape([-1, 28, 28, 1]) / 255.0
ar2 = array2.astype('float32').reshape([-1, 28, 28, 1]) / 255.0
else:
ar1 = array1.astype('float32').reshape([-1, 32, 32, 3]) / 255
ar2 = array2.astype('float32').reshape([-1, 32, 32, 3]) / 255
return (ar1, ar2)
def noise(a1, a2, channel):
"""
Adds random noise to each image in the supplied array.
"""
if channel == 1:
noise_factor = 0.2
noisy_arr1 = a1 + noise_factor * np.random.normal(0.0, 1.0, size=a1.shape)
noisy_arr2 = a2 + noise_factor * np.random.normal(0.0, 1.0, size=a2.shape)
else:
noi = 0.1
noisy_arr1 = a1 + noi * np.random.normal(0.0, 1.0, size=a1.shape)
noisy_arr2 = a2 + noi * np.random.normal(0.0, 1.0, size=a2.shape)
ab1 = np.clip(noisy_arr1, 0, 1)
ab2 = np.clip(noisy_arr2, 0, 1)
return (ab1, ab2)
# Visualization for mnist, cifar10, noisy, denoised/predictions data
def display(rows, cols, a, b, check=False ):
'''rows: defining no. of rows in figure
cols: defining no. of colums in figure
a: train images without noise or noisy_image while test
prediction
b: train images with noise or denoised_image based while test
prediction
check: default False for 32*32 cifar10, true for 28*28
mnist dataset and any predictions
'''
# defining a figure
f = plt.figure(figsize=(2*cols,2*rows*2))
for i in range(rows):
for j in range(cols):
# adding subplot to figure on each iteration
f.add_subplot(rows*2,cols, (2*i*cols)+(j+1))
if check:
plt.imshow(a[i*cols + j].reshape([28,28]),cmap="Blues")
else:
plt.imshow(a[i*cols + j])
plt.axis("off")
for j in range(cols):
# adding subplot to figure on each iteration
f.add_subplot(rows*2,cols,((2*i+1)*cols)+(j+1))
if check:
plt.imshow(b[i*cols + j].reshape([28,28]),cmap="Blues")
else:
plt.imshow(b[i*cols + j])
plt.axis("off")
plt.axis("off")
#f.suptitle("Sample Training Data",fontsize=18)
plt.savefig("ss.png")
plt.show()
(train_data, _), (test_data, _) = mnist.load_data()
channel = 1
train_data, test_data = preprocess(train_data, test_data, channel)
noisy_train_data, noisy_test_data = noise(train_data, test_data, channel)
(cifar_train, _), (cifar_test, _) = cifar10.load_data()
rows = 2
cols = 3
channel = 3
cifar_train, cifar_test = preprocess(cifar_train, cifar_test, channel)
cifar_train_noise, cifar_test_noise = noise(cifar_train, cifar_test, channel)
inputs = Input(shape=(28, 28, 1))
x = Conv2D(32, 3, activation='relu', padding='same')(inputs)
x = MaxPool2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
encoded = MaxPool2D()(x)
x = Conv2D(32, 3, activation='relu', padding='same')(encoded)
x = UpSampling2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
x = UpSampling2D()(x)
decoded = Conv2D(1, 3, activation='sigmoid', padding='same')(x)
from tensorflow.keras import losses
autoencoder1 = Model(inputs, decoded)
autoencoder1.compile(optimizer='adam', loss=losses.binary_crossentropy)
autoencoder1.summary()
history1 = autoencoder1.fit(noisy_train_data, train_data, epochs=50, batch_size=256, shuffle=True, validation_data=(noisy_test_data, test_data))
def plot_diag(history):
f = plt.figure(figsize=(10,7))
f.add_subplot()
#Adding Subplot
plt.plot(history.epoch, history.history['loss'], label = "loss") # Loss curve for training set
plt.plot(history.epoch, history.history['val_loss'], label = "val_loss") # Loss curve for validation set
plt.title("Loss Curve",fontsize=18)
plt.xlabel("Epochs",fontsize=15)
plt.ylabel("Loss",fontsize=15)
plt.grid(alpha=0.3)
plt.legend()
plt.savefig("Loss_curve.png")
plt.show()
plot_diag(history1)
from tensorflow.keras import layers, losses
class Denoise(Model):
"""__init__ constructor in OOP
This method called when an object is created from the class and
it allow the class to initialize the attributes of a class.
super()function used to give access to
methods and properties of a parent or sibling class
"""
def __init__(self):
super(Denoise, self).__init__()
self.encoder = tf.keras.Sequential([layers.Input(shape=(28, 28, 1)), layers.Conv2D(16, (3, 3), activation='relu', padding='same', strides=2), layers.Conv2D(8, (3, 3), activation='relu', padding='same', strides=2)])
self.decoder = tf.keras.Sequential([layers.Conv2DTranspose(8, kernel_size=3, strides=2, activation='relu', padding='same'), layers.Conv2DTranspose(16, kernel_size=3, strides=2, activation='relu', padding='same'), layers.Conv2D(1, kernel_size=(3, 3), activation='sigmoid', padding='same')])
def call(self, x):
encoded = self.encoder(x)
decoded = self.decoder(encoded)
return decoded
autoencoder = Denoise()
size = 32
channel = 3
inputs = Input(shape=(size, size, channel))
x = Conv2D(32, 3, activation='relu', padding='same')(inputs)
x = BatchNormalization()(x)
x = MaxPool2D()(x)
x = Dropout(0.5)(x)
skip = Conv2D(32, 3, padding='same')(x)
x = LeakyReLU()(skip)
x = BatchNormalization()(x)
x = MaxPool2D()(x)
x = Dropout(0.5)(x)
x = Conv2D(64, 3, activation='relu', padding='same')(x)
x = BatchNormalization()(x)
encoded = MaxPool2D()(x)
x = Conv2DTranspose(64, 3, activation='relu', strides=(2, 2), padding='same')(encoded)
x = BatchNormalization()(x)
x = Dropout(0.5)(x)
x = Conv2DTranspose(32, 3, activation='relu', strides=(2, 2), padding='same')(x)
x = BatchNormalization()(x)
x = Dropout(0.5)(x)
x = Conv2DTranspose(32, 3, padding='same')(x)
x = add([x, skip])
x = LeakyReLU()(x)
x = BatchNormalization()(x)
decoded = Conv2DTranspose(3, 3, activation='sigmoid', strides=(2, 2), padding='same')(x)
autoencoder2 = Model(inputs, decoded)
autoencoder2.compile(optimizer=Adam(lr=0.0001), loss='binary_crossentropy')
autoencoder2.summary()
epochs = 25
batch_size = 256
history2 = autoencoder2.fit(cifar_train_noise, cifar_train, epochs=epochs, batch_size=batch_size, shuffle=True, validation_data=(cifar_test_noise, cifar_test))
plot_diag(history2) | code |
73079642/cell_27 | [
"image_output_1.png"
] | from keras.layers import Input,Conv2D,MaxPool2D, UpSampling2D,Dense, Dropout
from keras.models import Model
from tensorflow.keras import layers, losses
from tensorflow.keras import losses
import tensorflow as tf
inputs = Input(shape=(28, 28, 1))
x = Conv2D(32, 3, activation='relu', padding='same')(inputs)
x = MaxPool2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
encoded = MaxPool2D()(x)
x = Conv2D(32, 3, activation='relu', padding='same')(encoded)
x = UpSampling2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
x = UpSampling2D()(x)
decoded = Conv2D(1, 3, activation='sigmoid', padding='same')(x)
from tensorflow.keras import losses
autoencoder1 = Model(inputs, decoded)
autoencoder1.compile(optimizer='adam', loss=losses.binary_crossentropy)
autoencoder1.summary()
from tensorflow.keras import layers, losses
class Denoise(Model):
"""__init__ constructor in OOP
This method called when an object is created from the class and
it allow the class to initialize the attributes of a class.
super()function used to give access to
methods and properties of a parent or sibling class
"""
def __init__(self):
super(Denoise, self).__init__()
self.encoder = tf.keras.Sequential([layers.Input(shape=(28, 28, 1)), layers.Conv2D(16, (3, 3), activation='relu', padding='same', strides=2), layers.Conv2D(8, (3, 3), activation='relu', padding='same', strides=2)])
self.decoder = tf.keras.Sequential([layers.Conv2DTranspose(8, kernel_size=3, strides=2, activation='relu', padding='same'), layers.Conv2DTranspose(16, kernel_size=3, strides=2, activation='relu', padding='same'), layers.Conv2D(1, kernel_size=(3, 3), activation='sigmoid', padding='same')])
def call(self, x):
encoded = self.encoder(x)
decoded = self.decoder(encoded)
return decoded
autoencoder = Denoise()
autoencoder.compile(optimizer='rmsprop', loss=losses.MeanSquaredError())
history = autoencoder.fit(noisy_train_data, train_data, epochs=10, batch_size=256, shuffle=True, validation_data=(noisy_test_data, test_data)) | code |
73079642/cell_37 | [
"text_plain_output_1.png"
] | from keras.datasets import mnist, cifar10
from keras.layers import Conv2DTranspose, BatchNormalization, add, LeakyReLU
from keras.layers import Input,Conv2D,MaxPool2D, UpSampling2D,Dense, Dropout
from keras.models import Model
from keras.models import load_model
from keras.models import load_model
from keras.optimizers import Adam
from tensorflow.keras import layers, losses
from tensorflow.keras import losses
import matplotlib.pyplot as plt
import numpy as np
import tensorflow as tf
def preprocess(array1, array2, channel):
"""
Normalizes/scales [0,1], divinding by the supplied array and reshapes
it into the appropriate format.
"""
if channel == 1:
ar1 = array1.astype('float32').reshape([-1, 28, 28, 1]) / 255.0
ar2 = array2.astype('float32').reshape([-1, 28, 28, 1]) / 255.0
else:
ar1 = array1.astype('float32').reshape([-1, 32, 32, 3]) / 255
ar2 = array2.astype('float32').reshape([-1, 32, 32, 3]) / 255
return (ar1, ar2)
def noise(a1, a2, channel):
"""
Adds random noise to each image in the supplied array.
"""
if channel == 1:
noise_factor = 0.2
noisy_arr1 = a1 + noise_factor * np.random.normal(0.0, 1.0, size=a1.shape)
noisy_arr2 = a2 + noise_factor * np.random.normal(0.0, 1.0, size=a2.shape)
else:
noi = 0.1
noisy_arr1 = a1 + noi * np.random.normal(0.0, 1.0, size=a1.shape)
noisy_arr2 = a2 + noi * np.random.normal(0.0, 1.0, size=a2.shape)
ab1 = np.clip(noisy_arr1, 0, 1)
ab2 = np.clip(noisy_arr2, 0, 1)
return (ab1, ab2)
# Visualization for mnist, cifar10, noisy, denoised/predictions data
def display(rows, cols, a, b, check=False ):
'''rows: defining no. of rows in figure
cols: defining no. of colums in figure
a: train images without noise or noisy_image while test
prediction
b: train images with noise or denoised_image based while test
prediction
check: default False for 32*32 cifar10, true for 28*28
mnist dataset and any predictions
'''
# defining a figure
f = plt.figure(figsize=(2*cols,2*rows*2))
for i in range(rows):
for j in range(cols):
# adding subplot to figure on each iteration
f.add_subplot(rows*2,cols, (2*i*cols)+(j+1))
if check:
plt.imshow(a[i*cols + j].reshape([28,28]),cmap="Blues")
else:
plt.imshow(a[i*cols + j])
plt.axis("off")
for j in range(cols):
# adding subplot to figure on each iteration
f.add_subplot(rows*2,cols,((2*i+1)*cols)+(j+1))
if check:
plt.imshow(b[i*cols + j].reshape([28,28]),cmap="Blues")
else:
plt.imshow(b[i*cols + j])
plt.axis("off")
plt.axis("off")
#f.suptitle("Sample Training Data",fontsize=18)
plt.savefig("ss.png")
plt.show()
(train_data, _), (test_data, _) = mnist.load_data()
channel = 1
train_data, test_data = preprocess(train_data, test_data, channel)
noisy_train_data, noisy_test_data = noise(train_data, test_data, channel)
(cifar_train, _), (cifar_test, _) = cifar10.load_data()
rows = 2
cols = 3
channel = 3
cifar_train, cifar_test = preprocess(cifar_train, cifar_test, channel)
cifar_train_noise, cifar_test_noise = noise(cifar_train, cifar_test, channel)
inputs = Input(shape=(28, 28, 1))
x = Conv2D(32, 3, activation='relu', padding='same')(inputs)
x = MaxPool2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
encoded = MaxPool2D()(x)
x = Conv2D(32, 3, activation='relu', padding='same')(encoded)
x = UpSampling2D()(x)
x = Dropout(0.2)(x)
x = Conv2D(32, 3, activation='relu', padding='same')(x)
x = UpSampling2D()(x)
decoded = Conv2D(1, 3, activation='sigmoid', padding='same')(x)
from tensorflow.keras import losses
autoencoder1 = Model(inputs, decoded)
autoencoder1.compile(optimizer='adam', loss=losses.binary_crossentropy)
autoencoder1.summary()
history1 = autoencoder1.fit(noisy_train_data, train_data, epochs=50, batch_size=256, shuffle=True, validation_data=(noisy_test_data, test_data))
autoencoder1.save('autoencoder_model1.h5')
from keras.models import load_model
model1 = load_model('autoencoder_model1.h5')
num_imgs = 45
rand = np.random.randint(1, 100)
test_images = noisy_test_data[rand:rand + num_imgs]
test_denoised = model1.predict(test_images)
from tensorflow.keras import layers, losses
class Denoise(Model):
"""__init__ constructor in OOP
This method called when an object is created from the class and
it allow the class to initialize the attributes of a class.
super()function used to give access to
methods and properties of a parent or sibling class
"""
def __init__(self):
super(Denoise, self).__init__()
self.encoder = tf.keras.Sequential([layers.Input(shape=(28, 28, 1)), layers.Conv2D(16, (3, 3), activation='relu', padding='same', strides=2), layers.Conv2D(8, (3, 3), activation='relu', padding='same', strides=2)])
self.decoder = tf.keras.Sequential([layers.Conv2DTranspose(8, kernel_size=3, strides=2, activation='relu', padding='same'), layers.Conv2DTranspose(16, kernel_size=3, strides=2, activation='relu', padding='same'), layers.Conv2D(1, kernel_size=(3, 3), activation='sigmoid', padding='same')])
def call(self, x):
encoded = self.encoder(x)
decoded = self.decoder(encoded)
return decoded
autoencoder = Denoise()
autoencoder.compile(optimizer='rmsprop', loss=losses.MeanSquaredError())
history = autoencoder.fit(noisy_train_data, train_data, epochs=10, batch_size=256, shuffle=True, validation_data=(noisy_test_data, test_data))
num_imgs = 45
rand = np.random.randint(1, 100)
test_images = noisy_test_data[rand:rand + num_imgs]
test_denoised = autoencoder.predict(test_images)
size = 32
channel = 3
inputs = Input(shape=(size, size, channel))
x = Conv2D(32, 3, activation='relu', padding='same')(inputs)
x = BatchNormalization()(x)
x = MaxPool2D()(x)
x = Dropout(0.5)(x)
skip = Conv2D(32, 3, padding='same')(x)
x = LeakyReLU()(skip)
x = BatchNormalization()(x)
x = MaxPool2D()(x)
x = Dropout(0.5)(x)
x = Conv2D(64, 3, activation='relu', padding='same')(x)
x = BatchNormalization()(x)
encoded = MaxPool2D()(x)
x = Conv2DTranspose(64, 3, activation='relu', strides=(2, 2), padding='same')(encoded)
x = BatchNormalization()(x)
x = Dropout(0.5)(x)
x = Conv2DTranspose(32, 3, activation='relu', strides=(2, 2), padding='same')(x)
x = BatchNormalization()(x)
x = Dropout(0.5)(x)
x = Conv2DTranspose(32, 3, padding='same')(x)
x = add([x, skip])
x = LeakyReLU()(x)
x = BatchNormalization()(x)
decoded = Conv2DTranspose(3, 3, activation='sigmoid', strides=(2, 2), padding='same')(x)
autoencoder2 = Model(inputs, decoded)
autoencoder2.compile(optimizer=Adam(lr=0.0001), loss='binary_crossentropy')
autoencoder2.summary()
epochs = 25
batch_size = 256
history2 = autoencoder2.fit(cifar_train_noise, cifar_train, epochs=epochs, batch_size=batch_size, shuffle=True, validation_data=(cifar_test_noise, cifar_test))
autoencoder2.save('autoencoder_model.h5')
from keras.models import load_model
model2 = load_model('autoencoder_model.h5')
num_imgs = 48
rand = np.random.randint(1, cifar_test_noise.shape[0] - 48)
cifar_test_images = cifar_test_noise[rand:rand + num_imgs]
cifar_test_denoised = autoencoder2.predict(cifar_test_images)
display(3, 4, cifar_test_images, cifar_test_denoised) | code |
129014335/cell_9 | [
"application_vnd.jupyter.stderr_output_2.png",
"text_plain_output_1.png"
] | import pandas as pd
iris = pd.read_csv('/kaggle/input/iris/Iris.csv')
df = iris.drop(['Id'], axis=1)
df['Species'].value_counts() | code |
129014335/cell_23 | [
"text_html_output_1.png"
] | from sklearn.linear_model import LogisticRegression
from sklearn.linear_model import LogisticRegression
model_LR = LogisticRegression()
model_LR.fit(X_train, y_train) | code |
129014335/cell_30 | [
"text_html_output_1.png"
] | from sklearn.metrics import accuracy_score
from sklearn.metrics import accuracy_score
from sklearn.metrics import accuracy_score
from sklearn.tree import DecisionTreeClassifier
from sklearn.tree import DecisionTreeClassifier
model_DTC = DecisionTreeClassifier()
model_DTC.fit(X_train, y_train)
predictionDTC = model_DTC.predict(X_test)
from sklearn.metrics import accuracy_score
print(accuracy_score(y_test, predictionDTC) * 100) | code |
129014335/cell_6 | [
"text_plain_output_1.png"
] | import pandas as pd
iris = pd.read_csv('/kaggle/input/iris/Iris.csv')
df = iris.drop(['Id'], axis=1)
df.head() | code |
129014335/cell_29 | [
"text_plain_output_1.png"
] | from sklearn.tree import DecisionTreeClassifier
from sklearn.tree import DecisionTreeClassifier
model_DTC = DecisionTreeClassifier()
model_DTC.fit(X_train, y_train) | code |
129014335/cell_26 | [
"text_plain_output_1.png"
] | from sklearn.svm import SVC
from sklearn.svm import SVC
model_SVC = SVC()
model_SVC.fit(X_train, y_train) | code |
129014335/cell_11 | [
"application_vnd.jupyter.stderr_output_1.png"
] | import pandas as pd
import seaborn as sns
iris = pd.read_csv('/kaggle/input/iris/Iris.csv')
df = iris.drop(['Id'], axis=1)
sns.pairplot(df, hue='Species') | code |
129014335/cell_7 | [
"text_html_output_1.png"
] | import pandas as pd
iris = pd.read_csv('/kaggle/input/iris/Iris.csv')
df = iris.drop(['Id'], axis=1)
df.describe() | code |
129014335/cell_32 | [
"text_plain_output_1.png"
] | from sklearn.metrics import accuracy_score
from sklearn.metrics import accuracy_score
from sklearn.metrics import accuracy_score
from sklearn.svm import SVC
import numpy as np
from sklearn.svm import SVC
model_SVC = SVC()
model_SVC.fit(X_train, y_train)
predictionSVC = model_SVC.predict(X_test)
from sklearn.metrics import accuracy_score
X_new = np.array([[3.4, 2.2, 1.5, 0.5], [4.8, 2.3, 3.7, 1.3], [5.1, 2.6, 4.9, 2]])
prediction_new = model_SVC.predict(X_new)
print('Prediction of new species : {}'.format(prediction_new)) | code |
129014335/cell_8 | [
"text_plain_output_1.png"
] | import pandas as pd
iris = pd.read_csv('/kaggle/input/iris/Iris.csv')
df = iris.drop(['Id'], axis=1)
df.info() | code |
129014335/cell_3 | [
"text_plain_output_1.png"
] | import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns | code |
129014335/cell_17 | [
"text_html_output_1.png"
] | from sklearn.preprocessing import LabelEncoder
import pandas as pd
iris = pd.read_csv('/kaggle/input/iris/Iris.csv')
df = iris.drop(['Id'], axis=1)
df.isnull().sum()
from sklearn.preprocessing import LabelEncoder
le = LabelEncoder()
df['Species'] = le.fit_transform(df['Species'])
print(df.head())
print(df[50:55])
print(df.tail()) | code |
129014335/cell_24 | [
"text_plain_output_1.png"
] | from sklearn.linear_model import LogisticRegression
from sklearn.metrics import accuracy_score
from sklearn.linear_model import LogisticRegression
model_LR = LogisticRegression()
model_LR.fit(X_train, y_train)
predictionLR = model_LR.predict(X_test)
from sklearn.metrics import accuracy_score
print(accuracy_score(y_test, predictionLR) * 100) | code |
129014335/cell_14 | [
"text_html_output_1.png"
] | import pandas as pd
iris = pd.read_csv('/kaggle/input/iris/Iris.csv')
df = iris.drop(['Id'], axis=1)
df.isnull().sum() | code |
129014335/cell_27 | [
"text_plain_output_1.png",
"image_output_1.png"
] | from sklearn.metrics import accuracy_score
from sklearn.metrics import accuracy_score
from sklearn.svm import SVC
from sklearn.svm import SVC
model_SVC = SVC()
model_SVC.fit(X_train, y_train)
predictionSVC = model_SVC.predict(X_test)
from sklearn.metrics import accuracy_score
print(accuracy_score(y_test, predictionSVC) * 100) | code |
129014335/cell_5 | [
"text_html_output_1.png"
] | import pandas as pd
iris = pd.read_csv('/kaggle/input/iris/Iris.csv')
iris.head() | code |
128047546/cell_13 | [
"text_plain_output_1.png"
] | from sklearn.model_selection import KFold | code |
128047546/cell_6 | [
"text_plain_output_1.png"
] | import pandas as pd
boston = pd.read_pickle('https://cf-courses-data.s3.us.cloud-object-storage.appdomain.cloud/IBM-ML240EN-SkillsNetwork/labs/data/boston_housing_clean.pickle')
boston.keys()
boston['description'] | code |
128047546/cell_11 | [
"text_html_output_1.png"
] | import pandas as pd
boston = pd.read_pickle('https://cf-courses-data.s3.us.cloud-object-storage.appdomain.cloud/IBM-ML240EN-SkillsNetwork/labs/data/boston_housing_clean.pickle')
boston.keys()
boston_df = boston['dataframe']
X = boston_df.drop('MEDV', axis=1)
y = boston_df['MEDV']
y | code |
128047546/cell_19 | [
"application_vnd.jupyter.stderr_output_1.png"
] | from sklearn.linear_model import LinearRegression
from sklearn.metrics import r2_score,mean_squared_error
from sklearn.model_selection import KFold
import pandas as pd
boston = pd.read_pickle('https://cf-courses-data.s3.us.cloud-object-storage.appdomain.cloud/IBM-ML240EN-SkillsNetwork/labs/data/boston_housing_clean.pickle')
boston.keys()
boston_df = boston['dataframe']
X = boston_df.drop('MEDV', axis=1)
y = boston_df['MEDV']
X.shape
y.shape
kf = KFold(n_splits=3, shuffle=True)
type(kf.split(X))
LR = LinearRegression()
scores = []
for train_index, test_index in kf.split(X):
X_train, X_test, y_train, y_test = (X.iloc[train_index, :], X.iloc[test_index, :], y[train_index], y[test_index])
print('X_train:', X.iloc[train_index, :], 'y_train:', y[train_index], 'X_test:', X.iloc[test_index, :], 'y_test:', y[test_index])
LR.fit(X_train, y_train)
y_pred = LR.predict(X_test)
score = r2_score(y_test.values, y_pred)
scores.append(score)
scores | code |
128047546/cell_7 | [
"text_plain_output_1.png"
] | column_desc = "Boston House Prices dataset\n===========================\n\nNotes\n------\nData Set Characteristics: \n\n :Number of Instances: 506 \n\n :Number of Attributes: 13 numeric/categorical predictive\n \n :Median Value (attribute 14) is usually the target\n\n :Attribute Information (in order):\n - CRIM per capita crime rate by town\n - ZN proportion of residential land zoned for lots over 25,000 sq.ft.\n - INDUS proportion of non-retail business acres per town\n - CHAS Charles River dummy variable (= 1 if tract bounds river; 0 otherwise)\n - NOX nitric oxides concentration (parts per 10 million)\n - RM average number of rooms per dwelling\n - AGE proportion of owner-occupied units built prior to 1940\n - DIS weighted distances to five Boston employment centres\n - RAD index of accessibility to radial highways\n - TAX full-value property-tax rate per $10,000\n - PTRATIO pupil-teacher ratio by town\n - B 1000(Bk - 0.63)^2 where Bk is the proportion of blacks by town\n - LSTAT % lower status of the population\n - MEDV Median value of owner-occupied homes in $1000's\n\n :Missing Attribute Values: None\n\n :Creator: Harrison, D. and Rubinfeld, D.L.\n\nThis is a copy of UCI ML housing dataset.\nhttp://archive.ics.uci.edu/ml/datasets/Housing\n\n\nThis dataset was taken from the StatLib library which is maintained at Carnegie Mellon University.\n\nThe Boston house-price data of Harrison, D. and Rubinfeld, D.L. 'Hedonic\nprices and the demand for clean air', J. Environ. Economics & Management,\nvol.5, 81-102, 1978. Used in Belsley, Kuh & Welsch, 'Regression diagnostics\n...', Wiley, 1980. N.B. Various transformations are used in the table on\npages 244-261 of the latter.\n\nThe Boston house-price data has been used in many machine learning papers that address regression\nproblems. \n \n**References**\n\n - Belsley, Kuh & Welsch, 'Regression diagnostics: Identifying Influential Data and Sources of Collinearity', Wiley, 1980. 244-261.\n - Quinlan,R. (1993). Combining Instance-Based and Model-Based Learning. In Proceedings on the Tenth International Conference of Machine Learning, 236-243, University of Massachusetts, Amherst. Morgan Kaufmann.\n - many more! (see http://archive.ics.uci.edu/ml/datasets/Housing)\n"
lines = column_desc.split('\n')
lines | code |
128047546/cell_15 | [
"text_plain_output_1.png"
] | from sklearn.model_selection import KFold
kf = KFold(n_splits=3, shuffle=True)
kf | code |
128047546/cell_16 | [
"text_plain_output_1.png"
] | from sklearn.model_selection import KFold
import pandas as pd
boston = pd.read_pickle('https://cf-courses-data.s3.us.cloud-object-storage.appdomain.cloud/IBM-ML240EN-SkillsNetwork/labs/data/boston_housing_clean.pickle')
boston.keys()
boston_df = boston['dataframe']
X = boston_df.drop('MEDV', axis=1)
X.shape
kf = KFold(n_splits=3, shuffle=True)
type(kf.split(X)) | code |
128047546/cell_3 | [
"text_plain_output_1.png"
] | import pandas as pd
boston = pd.read_pickle('https://cf-courses-data.s3.us.cloud-object-storage.appdomain.cloud/IBM-ML240EN-SkillsNetwork/labs/data/boston_housing_clean.pickle')
boston.keys() | code |
128047546/cell_17 | [
"text_plain_output_1.png"
] | from sklearn.model_selection import KFold
import pandas as pd
boston = pd.read_pickle('https://cf-courses-data.s3.us.cloud-object-storage.appdomain.cloud/IBM-ML240EN-SkillsNetwork/labs/data/boston_housing_clean.pickle')
boston.keys()
boston_df = boston['dataframe']
X = boston_df.drop('MEDV', axis=1)
X.shape
kf = KFold(n_splits=3, shuffle=True)
type(kf.split(X))
for training_set, test_set in kf.split(X):
print('Training set', training_set, '\n', 'Total:', len(training_set))
print('Test set', test_set, '\n', 'Total:', len(test_set))
print(' ') | code |
128047546/cell_24 | [
"text_plain_output_1.png"
] | from sklearn.linear_model import LinearRegression
from sklearn.metrics import r2_score,mean_squared_error
from sklearn.model_selection import KFold
from sklearn.model_selection import cross_val_predict
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler
import pandas as pd
boston = pd.read_pickle('https://cf-courses-data.s3.us.cloud-object-storage.appdomain.cloud/IBM-ML240EN-SkillsNetwork/labs/data/boston_housing_clean.pickle')
boston.keys()
boston_df = boston['dataframe']
X = boston_df.drop('MEDV', axis=1)
y = boston_df['MEDV']
X.shape
y.shape
kf = KFold(n_splits=3, shuffle=True)
type(kf.split(X))
LR = LinearRegression()
scores = []
for train_index, test_index in kf.split(X):
X_train, X_test, y_train, y_test = (X.iloc[train_index, :], X.iloc[test_index, :], y[train_index], y[test_index])
LR.fit(X_train, y_train)
y_pred = LR.predict(X_test)
score = r2_score(y_test.values, y_pred)
scores.append(score)
scores
steps = Pipeline([('std_scaler', StandardScaler()), ('Lin_Reg', LinearRegression())])
from sklearn.model_selection import cross_val_predict
predict = cross_val_predict(steps, X, y, cv=kf)
r2_score(y, predict) | code |
128047546/cell_22 | [
"text_plain_output_1.png"
] | from sklearn.linear_model import LinearRegression
from sklearn.metrics import r2_score,mean_squared_error
from sklearn.model_selection import KFold
import pandas as pd
boston = pd.read_pickle('https://cf-courses-data.s3.us.cloud-object-storage.appdomain.cloud/IBM-ML240EN-SkillsNetwork/labs/data/boston_housing_clean.pickle')
boston.keys()
boston_df = boston['dataframe']
X = boston_df.drop('MEDV', axis=1)
y = boston_df['MEDV']
X.shape
y.shape
kf = KFold(n_splits=3, shuffle=True)
type(kf.split(X))
LR = LinearRegression()
scores = []
for train_index, test_index in kf.split(X):
X_train, X_test, y_train, y_test = (X.iloc[train_index, :], X.iloc[test_index, :], y[train_index], y[test_index])
LR.fit(X_train, y_train)
y_pred = LR.predict(X_test)
score = r2_score(y_test.values, y_pred)
scores.append(score)
scores
kf | code |
128047546/cell_10 | [
"text_plain_output_1.png"
] | import pandas as pd
boston = pd.read_pickle('https://cf-courses-data.s3.us.cloud-object-storage.appdomain.cloud/IBM-ML240EN-SkillsNetwork/labs/data/boston_housing_clean.pickle')
boston.keys()
boston_df = boston['dataframe']
X = boston_df.drop('MEDV', axis=1)
X.shape | code |
128047546/cell_12 | [
"text_plain_output_1.png"
] | import pandas as pd
boston = pd.read_pickle('https://cf-courses-data.s3.us.cloud-object-storage.appdomain.cloud/IBM-ML240EN-SkillsNetwork/labs/data/boston_housing_clean.pickle')
boston.keys()
boston_df = boston['dataframe']
X = boston_df.drop('MEDV', axis=1)
y = boston_df['MEDV']
y.shape | code |
128047546/cell_5 | [
"text_plain_output_2.png",
"text_plain_output_1.png"
] | import pandas as pd
boston = pd.read_pickle('https://cf-courses-data.s3.us.cloud-object-storage.appdomain.cloud/IBM-ML240EN-SkillsNetwork/labs/data/boston_housing_clean.pickle')
boston.keys()
boston_df = boston['dataframe']
boston_df | code |
90107754/cell_13 | [
"text_plain_output_1.png"
] | import cudf as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/train.tsv.zip', sep='\t')
test_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/test.tsv.zip', sep='\t')
test_df.describe() | code |
90107754/cell_9 | [
"text_plain_output_1.png"
] | import cudf as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/train.tsv.zip', sep='\t')
test_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/test.tsv.zip', sep='\t')
train_df.describe() | code |
90107754/cell_34 | [
"text_plain_output_1.png",
"image_output_1.png"
] | from keras.layers import Bidirectional
from keras.layers import Dense
from keras.layers import Dense,Dropout,Embedding,LSTM
from keras.layers import Embedding
from keras.layers import LSTM
from keras.models import Sequential
from keras.models import Sequential
from keras.preprocessing.text import Tokenizer
from tensorflow.keras.layers import Dense, Embedding
from tensorflow.keras.layers import LSTM
from tensorflow.keras.models import Sequential
from tensorflow.keras.preprocessing.sequence import pad_sequences
from tensorflow.keras.preprocessing.text import Tokenizer
import cudf as pd
import keras
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import seaborn as sns
train_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/train.tsv.zip', sep='\t')
test_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/test.tsv.zip', sep='\t')
train_df.shape
test_df.shape
train_df.isnull().sum()
test_df.isnull().sum()
train_df.isnull().any().any()
test_df.isnull().any().any()
train = train_df.to_pandas()
test = test_df.to_pandas()
X_train = train.Phrase
Y_train = train.Sentiment
tokenize = Tokenizer()
tokenize.fit_on_texts(X_train.values)
X_test = test.Phrase
X_train = tokenize.texts_to_sequences(X_train)
X_test = tokenize.texts_to_sequences(X_test)
max_lenght = max([len(s.split()) for s in train['Phrase']])
X_train = pad_sequences(X_train, max_lenght)
X_test = pad_sequences(X_test, max_lenght)
model = Sequential()
inputs = keras.Input(shape=(None,), dtype='int32')
model.add(inputs)
model.add(Embedding(50000, 128))
model.add(Bidirectional(LSTM(64, return_sequences=True)))
model.add(Bidirectional(LSTM(64)))
model.add(Dense(5, activation='softmax'))
model.summary()
model.compile('adam', 'sparse_categorical_crossentropy', metrics=['accuracy'])
model.fit(X_train, Y_train, batch_size=256, epochs=30)
y_pred = model.predict(X_test)
y_pred | code |
90107754/cell_33 | [
"text_plain_output_1.png"
] | from keras.layers import Bidirectional
from keras.layers import Dense
from keras.layers import Dense,Dropout,Embedding,LSTM
from keras.layers import Embedding
from keras.layers import LSTM
from keras.models import Sequential
from keras.models import Sequential
from keras.preprocessing.text import Tokenizer
from tensorflow.keras.layers import Dense, Embedding
from tensorflow.keras.layers import LSTM
from tensorflow.keras.models import Sequential
from tensorflow.keras.preprocessing.sequence import pad_sequences
from tensorflow.keras.preprocessing.text import Tokenizer
import cudf as pd
import keras
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import seaborn as sns
train_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/train.tsv.zip', sep='\t')
test_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/test.tsv.zip', sep='\t')
train_df.shape
test_df.shape
train_df.isnull().sum()
test_df.isnull().sum()
train_df.isnull().any().any()
test_df.isnull().any().any()
train = train_df.to_pandas()
test = test_df.to_pandas()
X_train = train.Phrase
Y_train = train.Sentiment
tokenize = Tokenizer()
tokenize.fit_on_texts(X_train.values)
X_test = test.Phrase
X_train = tokenize.texts_to_sequences(X_train)
X_test = tokenize.texts_to_sequences(X_test)
max_lenght = max([len(s.split()) for s in train['Phrase']])
X_train = pad_sequences(X_train, max_lenght)
X_test = pad_sequences(X_test, max_lenght)
model = Sequential()
inputs = keras.Input(shape=(None,), dtype='int32')
model.add(inputs)
model.add(Embedding(50000, 128))
model.add(Bidirectional(LSTM(64, return_sequences=True)))
model.add(Bidirectional(LSTM(64)))
model.add(Dense(5, activation='softmax'))
model.summary()
model.compile('adam', 'sparse_categorical_crossentropy', metrics=['accuracy'])
model.fit(X_train, Y_train, batch_size=256, epochs=30) | code |
90107754/cell_11 | [
"text_plain_output_1.png"
] | import cudf as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/train.tsv.zip', sep='\t')
test_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/test.tsv.zip', sep='\t')
test_df.head() | code |
90107754/cell_19 | [
"text_plain_output_1.png"
] | import cudf as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/train.tsv.zip', sep='\t')
test_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/test.tsv.zip', sep='\t')
test_df.shape
test_df.isnull().sum()
test_df.isnull().any().any() | code |
90107754/cell_1 | [
"text_plain_output_1.png"
] | import os
import numpy as np
import pandas as pd
import os
for dirname, _, filenames in os.walk('/kaggle/input'):
for filename in filenames:
print(os.path.join(dirname, filename)) | code |
90107754/cell_7 | [
"text_plain_output_2.png",
"application_vnd.jupyter.stderr_output_1.png"
] | import cudf as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/train.tsv.zip', sep='\t')
test_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/test.tsv.zip', sep='\t')
train_df.head() | code |
90107754/cell_18 | [
"text_plain_output_1.png"
] | import cudf as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/train.tsv.zip', sep='\t')
test_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/test.tsv.zip', sep='\t')
train_df.shape
train_df.isnull().sum()
train_df.isnull().any().any() | code |
90107754/cell_32 | [
"text_plain_output_1.png"
] | from keras.layers import Bidirectional
from keras.layers import Dense
from keras.layers import Dense,Dropout,Embedding,LSTM
from keras.layers import Embedding
from keras.layers import LSTM
from keras.models import Sequential
from keras.models import Sequential
from tensorflow.keras.layers import Dense, Embedding
from tensorflow.keras.layers import LSTM
from tensorflow.keras.models import Sequential
import keras
model = Sequential()
inputs = keras.Input(shape=(None,), dtype='int32')
model.add(inputs)
model.add(Embedding(50000, 128))
model.add(Bidirectional(LSTM(64, return_sequences=True)))
model.add(Bidirectional(LSTM(64)))
model.add(Dense(5, activation='softmax'))
model.summary() | code |
90107754/cell_8 | [
"text_plain_output_5.png",
"text_plain_output_4.png",
"application_vnd.jupyter.stderr_output_3.png",
"text_plain_output_2.png",
"application_vnd.jupyter.stderr_output_1.png"
] | import cudf as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/train.tsv.zip', sep='\t')
test_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/test.tsv.zip', sep='\t')
train_df.info() | code |
90107754/cell_16 | [
"text_plain_output_1.png"
] | import cudf as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/train.tsv.zip', sep='\t')
test_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/test.tsv.zip', sep='\t')
train_df.shape
train_df.isnull().sum() | code |
90107754/cell_17 | [
"text_html_output_1.png"
] | import cudf as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/train.tsv.zip', sep='\t')
test_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/test.tsv.zip', sep='\t')
test_df.shape
test_df.isnull().sum() | code |
90107754/cell_24 | [
"text_plain_output_1.png"
] | import cudf as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import seaborn as sns
train_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/train.tsv.zip', sep='\t')
test_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/test.tsv.zip', sep='\t')
train_df.shape
train_df.isnull().sum()
train_df.isnull().any().any()
train = train_df.to_pandas()
sns.countplot(x='Sentiment', data=train) | code |
90107754/cell_14 | [
"text_html_output_1.png"
] | import cudf as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/train.tsv.zip', sep='\t')
test_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/test.tsv.zip', sep='\t')
test_df.shape | code |
90107754/cell_22 | [
"text_plain_output_1.png"
] | import cudf as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/train.tsv.zip', sep='\t')
test_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/test.tsv.zip', sep='\t')
train_df.shape
train_df.isnull().sum()
train_df.isnull().any().any()
train_df['Sentiment'].value_counts() | code |
90107754/cell_10 | [
"text_html_output_1.png"
] | import cudf as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/train.tsv.zip', sep='\t')
test_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/test.tsv.zip', sep='\t')
train_df.shape | code |
90107754/cell_12 | [
"text_html_output_1.png"
] | import cudf as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/train.tsv.zip', sep='\t')
test_df = pd.read_csv('/kaggle/input/sentiment-analysis-on-movie-reviews/test.tsv.zip', sep='\t')
test_df.info() | code |
90116977/cell_23 | [
"text_html_output_1.png"
] | import matplotlib.pyplot as plt
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import seaborn as sns
data = '../input/latest-covid19-india-statewise-data/Latest Covid-19 India Status.csv'
data2 = '../input/covid-time-series-data-india-till-31oct21/case_time_series.csv'
state = '../input/latest-covid19-cases-maharashtra-india/Maharashtra Latest Covid Cases.csv'
dataset = pd.read_csv(data)
dataset1 = pd.read_csv(data2)
maha = pd.read_csv(state)
dataset.shape
dataset.isna().sum().sort_values(ascending=False)
state_population = dataset.sort_values(by=['Population'], ascending=False)
plt.figure(figsize=(12, 10))
population_cnt = sns.barplot(state_population['Population'], state_population['State/UTs'], palette='Blues_d')
plt.title('State/UTs vs. Population')
plt.show() | code |
90116977/cell_20 | [
"text_plain_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
data = '../input/latest-covid19-india-statewise-data/Latest Covid-19 India Status.csv'
data2 = '../input/covid-time-series-data-india-till-31oct21/case_time_series.csv'
state = '../input/latest-covid19-cases-maharashtra-india/Maharashtra Latest Covid Cases.csv'
dataset = pd.read_csv(data)
dataset1 = pd.read_csv(data2)
maha = pd.read_csv(state)
dataset.shape
dataset.isna().sum().sort_values(ascending=False) | code |
90116977/cell_6 | [
"text_plain_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
data = '../input/latest-covid19-india-statewise-data/Latest Covid-19 India Status.csv'
data2 = '../input/covid-time-series-data-india-till-31oct21/case_time_series.csv'
state = '../input/latest-covid19-cases-maharashtra-india/Maharashtra Latest Covid Cases.csv'
dataset = pd.read_csv(data)
dataset1 = pd.read_csv(data2)
maha = pd.read_csv(state)
dataset.tail(10) | code |
90116977/cell_11 | [
"text_html_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
data = '../input/latest-covid19-india-statewise-data/Latest Covid-19 India Status.csv'
data2 = '../input/covid-time-series-data-india-till-31oct21/case_time_series.csv'
state = '../input/latest-covid19-cases-maharashtra-india/Maharashtra Latest Covid Cases.csv'
dataset = pd.read_csv(data)
dataset1 = pd.read_csv(data2)
maha = pd.read_csv(state)
maha.head() | code |
90116977/cell_1 | [
"text_plain_output_1.png"
] | import os
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import datetime as dt
import os
for dirname, _, filenames in os.walk('/kaggle/input'):
for filename in filenames:
print(os.path.join(dirname, filename)) | code |
90116977/cell_7 | [
"text_plain_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
data = '../input/latest-covid19-india-statewise-data/Latest Covid-19 India Status.csv'
data2 = '../input/covid-time-series-data-india-till-31oct21/case_time_series.csv'
state = '../input/latest-covid19-cases-maharashtra-india/Maharashtra Latest Covid Cases.csv'
dataset = pd.read_csv(data)
dataset1 = pd.read_csv(data2)
maha = pd.read_csv(state)
dataset1.head() | code |
90116977/cell_8 | [
"application_vnd.jupyter.stderr_output_1.png",
"image_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
data = '../input/latest-covid19-india-statewise-data/Latest Covid-19 India Status.csv'
data2 = '../input/covid-time-series-data-india-till-31oct21/case_time_series.csv'
state = '../input/latest-covid19-cases-maharashtra-india/Maharashtra Latest Covid Cases.csv'
dataset = pd.read_csv(data)
dataset1 = pd.read_csv(data2)
maha = pd.read_csv(state)
dataset1.tail() | code |
90116977/cell_15 | [
"text_html_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
data = '../input/latest-covid19-india-statewise-data/Latest Covid-19 India Status.csv'
data2 = '../input/covid-time-series-data-india-till-31oct21/case_time_series.csv'
state = '../input/latest-covid19-cases-maharashtra-india/Maharashtra Latest Covid Cases.csv'
dataset = pd.read_csv(data)
dataset1 = pd.read_csv(data2)
maha = pd.read_csv(state)
dataset.shape
dataset.describe(include='all') | code |
90116977/cell_17 | [
"text_html_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
data = '../input/latest-covid19-india-statewise-data/Latest Covid-19 India Status.csv'
data2 = '../input/covid-time-series-data-india-till-31oct21/case_time_series.csv'
state = '../input/latest-covid19-cases-maharashtra-india/Maharashtra Latest Covid Cases.csv'
dataset = pd.read_csv(data)
dataset1 = pd.read_csv(data2)
maha = pd.read_csv(state)
dataset.shape
dataset.info() | code |
90116977/cell_14 | [
"text_html_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
data = '../input/latest-covid19-india-statewise-data/Latest Covid-19 India Status.csv'
data2 = '../input/covid-time-series-data-india-till-31oct21/case_time_series.csv'
state = '../input/latest-covid19-cases-maharashtra-india/Maharashtra Latest Covid Cases.csv'
dataset = pd.read_csv(data)
dataset1 = pd.read_csv(data2)
maha = pd.read_csv(state)
dataset.shape
dataset['State/UTs'].unique() | code |
90116977/cell_22 | [
"text_plain_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
data = '../input/latest-covid19-india-statewise-data/Latest Covid-19 India Status.csv'
data2 = '../input/covid-time-series-data-india-till-31oct21/case_time_series.csv'
state = '../input/latest-covid19-cases-maharashtra-india/Maharashtra Latest Covid Cases.csv'
dataset = pd.read_csv(data)
dataset1 = pd.read_csv(data2)
maha = pd.read_csv(state)
dataset.shape
dataset.isna().sum().sort_values(ascending=False)
state_population = dataset.sort_values(by=['Population'], ascending=False)
print(state_population) | code |
90116977/cell_10 | [
"text_html_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
data = '../input/latest-covid19-india-statewise-data/Latest Covid-19 India Status.csv'
data2 = '../input/covid-time-series-data-india-till-31oct21/case_time_series.csv'
state = '../input/latest-covid19-cases-maharashtra-india/Maharashtra Latest Covid Cases.csv'
dataset = pd.read_csv(data)
dataset1 = pd.read_csv(data2)
maha = pd.read_csv(state)
today = dataset1[dataset1.Date_YMD == '2021-04-11']
today | code |
90116977/cell_12 | [
"text_html_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
data = '../input/latest-covid19-india-statewise-data/Latest Covid-19 India Status.csv'
data2 = '../input/covid-time-series-data-india-till-31oct21/case_time_series.csv'
state = '../input/latest-covid19-cases-maharashtra-india/Maharashtra Latest Covid Cases.csv'
dataset = pd.read_csv(data)
dataset1 = pd.read_csv(data2)
maha = pd.read_csv(state)
dataset.shape | code |
90116977/cell_5 | [
"text_plain_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
data = '../input/latest-covid19-india-statewise-data/Latest Covid-19 India Status.csv'
data2 = '../input/covid-time-series-data-india-till-31oct21/case_time_series.csv'
state = '../input/latest-covid19-cases-maharashtra-india/Maharashtra Latest Covid Cases.csv'
dataset = pd.read_csv(data)
dataset1 = pd.read_csv(data2)
maha = pd.read_csv(state)
dataset.head(10) | code |
128046602/cell_4 | [
"text_plain_output_1.png",
"image_output_1.png"
] | from sklearn.utils import shuffle
import pandas as pd
df = pd.read_csv('/kaggle/input/kullanlcaklar/Womens_Clothing_E-Commerce_Reviews_1.csv')
df = shuffle(df)
df = df.reset_index(drop=True)
df.head() | code |
128046602/cell_20 | [
"text_html_output_1.png"
] | from sklearn.utils import shuffle
import matplotlib.pyplot as plt
import pandas as pd
import random
import re
df = pd.read_csv('/kaggle/input/kullanlcaklar/Womens_Clothing_E-Commerce_Reviews_1.csv')
df = shuffle(df)
df = df.reset_index(drop=True)
N = 10
population_size = 100
mutation_rate = 0.3
generations = 200
text_column_name = 'Title'
rating_column_name = 'Rating'
def clean_text(text):
text = text.lower()
text = re.sub('[^\\w\\s]', '', text)
return text
def tokenize(text):
return text.split()
def prepare_population(df, text_column_name):
all_words = []
for text in df[text_column_name]:
clean_review = clean_text(text)
tokens = tokenize(clean_review)
all_words.extend(tokens)
random.shuffle(all_words)
sublists = [all_words[i:i + N] for i in range(0, len(all_words), N)]
population = sublists
return (population, all_words)
def count_words(text, word_list):
count = 0
for word in word_list:
count += text.count(word)
return count
def fitness(individual):
correct_classification = 0
for i in range(len(df)):
row = df.iloc[i]
text, rating = (row[text_column_name], row[rating_column_name])
pozitif = count_words(text, individual[:N // 2])
negatif = count_words(text, individual[N // 2:])
if pozitif == negatif:
if random.random() < 0.5:
classification = 1
else:
classification = 5
elif pozitif > negatif:
classification = 5
else:
classification = 1
if classification == rating:
correct_classification += 1
return correct_classification / len(df)
def fitness2(individual):
correct_classification = 0
for i in range(len(df)):
row = df.iloc[i]
text, rating = (row[text_column_name], row[rating_column_name])
pozitif = count_words(text, individual[:N // 2])
negatif = count_words(text, individual[N // 2:])
if pozitif == negatif:
if random.random() < 0.5:
classification = 'negative'
else:
classification = 'positive'
elif pozitif > negatif:
classification = 'positive'
else:
classification = 'negative'
if classification == rating:
correct_classification += 1
return correct_classification / len(df)
def crossover(parent1, parent2):
n = N // 2
child1 = parent1[:n] + parent2[n:]
child2 = parent2[:n] + parent1[n:]
return (child1, child2)
def mutate(individual, all_words):
if random.random() < mutation_rate:
i = random.randint(0, len(individual) - 1)
new_word = random.choice(all_words)
individual[i] = new_word
return individual
def display_results(population, success_rates, average_success_rates):
best_individual = max(population, key=fitness)
print("En iyi birey:", best_individual)
print("Başarı oranı:", fitness(best_individual))
fig, ax = plt.subplots()
ax.plot(success_rates, label="En iyi birey")
ax.plot(average_success_rates, label="Ortalama başarı")
ax.set_xlabel('Generation')
ax.set_ylabel('Success Rate')
ax.set_title('Success Rate per Generation')
ax.legend()
plt.show()
def display_results2(population, success_rates, average_success_rates):
best_individual = max(population, key=fitness2)
print("En iyi birey:", best_individual)
print("Başarı oranı:", fitness2(best_individual))
fig, ax = plt.subplots()
ax.plot(success_rates, label="En iyi birey")
ax.plot(average_success_rates, label="Ortalama başarı")
ax.set_xlabel('Generation')
ax.set_ylabel('Success Rate')
ax.set_title('Success Rate per Generation')
ax.legend()
plt.show()
success_rates = []
average_success_rates = []
for generation in range(generations):
fitness_values = []
for individual in population:
fitness_values.append(fitness(individual))
average_success_rate = sum(fitness_values) / population_size
average_success_rates.append(average_success_rate)
children = []
for i in range(population_size // 2):
parents = random.choices(population, weights=fitness_values, k=2)
child1, child2 = crossover(parents[0], parents[1])
children.append(child1)
children.append(child2)
mutations = []
for child in children:
mutations.append(mutate(child, all_words))
population = mutations
best_individual = max(population, key=fitness)
success_rate = fitness(best_individual)
success_rates.append(success_rate)
display_results(population, success_rates, average_success_rates) | code |
128046602/cell_2 | [
"text_html_output_1.png"
] | import nltk
import nltk
from nltk.corpus import stopwords
from nltk.tokenize import word_tokenize
nltk.download('stopwords')
nltk.download('punkt') | code |
128046602/cell_3 | [
"text_plain_output_1.png",
"image_output_1.png"
] | import pandas as pd
df = pd.read_csv('/kaggle/input/kullanlcaklar/Womens_Clothing_E-Commerce_Reviews_1.csv')
df.head() | code |
128046602/cell_17 | [
"text_plain_output_2.png",
"text_plain_output_1.png"
] | from sklearn.utils import shuffle
import pandas as pd
import random
import re
df = pd.read_csv('/kaggle/input/kullanlcaklar/Womens_Clothing_E-Commerce_Reviews_1.csv')
df = shuffle(df)
df = df.reset_index(drop=True)
N = 10
population_size = 100
mutation_rate = 0.3
generations = 200
text_column_name = 'Title'
rating_column_name = 'Rating'
def clean_text(text):
text = text.lower()
text = re.sub('[^\\w\\s]', '', text)
return text
def tokenize(text):
return text.split()
def prepare_population(df, text_column_name):
all_words = []
for text in df[text_column_name]:
clean_review = clean_text(text)
tokens = tokenize(clean_review)
all_words.extend(tokens)
random.shuffle(all_words)
sublists = [all_words[i:i + N] for i in range(0, len(all_words), N)]
population = sublists
return (population, all_words)
def count_words(text, word_list):
count = 0
for word in word_list:
count += text.count(word)
return count
def fitness(individual):
correct_classification = 0
for i in range(len(df)):
row = df.iloc[i]
text, rating = (row[text_column_name], row[rating_column_name])
pozitif = count_words(text, individual[:N // 2])
negatif = count_words(text, individual[N // 2:])
if pozitif == negatif:
if random.random() < 0.5:
classification = 1
else:
classification = 5
elif pozitif > negatif:
classification = 5
else:
classification = 1
if classification == rating:
correct_classification += 1
return correct_classification / len(df)
def fitness2(individual):
correct_classification = 0
for i in range(len(df)):
row = df.iloc[i]
text, rating = (row[text_column_name], row[rating_column_name])
pozitif = count_words(text, individual[:N // 2])
negatif = count_words(text, individual[N // 2:])
if pozitif == negatif:
if random.random() < 0.5:
classification = 'negative'
else:
classification = 'positive'
elif pozitif > negatif:
classification = 'positive'
else:
classification = 'negative'
if classification == rating:
correct_classification += 1
return correct_classification / len(df)
population, all_words = prepare_population(df, text_column_name)
print(population) | code |
88076328/cell_6 | [
"text_html_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
omicron_data = pd.read_csv('/kaggle/input/omicron-covid19-variant-daily-cases/covid-variants.csv')
omicron_data = pd.DataFrame(omicron_data)
omicron_data.loc[(omicron_data['location'] == 'Thailand') & (omicron_data['variant'] == 'Omicron')] | code |
88076328/cell_8 | [
"application_vnd.jupyter.stderr_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
omicron_data = pd.read_csv('/kaggle/input/omicron-covid19-variant-daily-cases/covid-variants.csv')
omicron_data = pd.DataFrame(omicron_data)
omicron_data.loc[(omicron_data['location'] == 'Thailand') & (omicron_data['variant'] == 'Omicron')]
omicron_num_seq_Thailand = omicron_data[omicron_data['location'] == 'Thailand']['num_sequences_total']
omicron_date_Thailand = omicron_data[omicron_data['location'] == 'Thailand']['date']
plt.figure(figsize=(50, 10))
plt.xlabel('date')
plt.ylabel('total cases')
plt.title('Omicron Cases in Thailand')
plt.plot(omicron_date_Thailand, omicron_num_seq_Thailand)
plt.show() | code |
88076328/cell_5 | [
"text_html_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
omicron_data = pd.read_csv('/kaggle/input/omicron-covid19-variant-daily-cases/covid-variants.csv')
omicron_data = pd.DataFrame(omicron_data)
omicron_data.head() | code |
74041571/cell_18 | [
"text_html_output_1.png"
] | from IPython.core.display import display, Markdown
from bs4 import BeautifulSoup
from pathlib import Path
import pandas as pd
PUB = Path('../input/30dmlleaderboards/public_lb.html')
PRIV = Path('../input/30dmlleaderboards/private_lb.html')
CSV_PUB = Path('../input/30dmlleaderboards/30-days-of-ml-publicleaderboard/30-days-of-ml-publicleaderboard.csv')
def _strip_all_spaces(series):
return series.replace('\\s+', ' ', regex=True).str.strip()
def _extract_kernel(element):
try:
anchor = element.find('a')
title = anchor.get('title')
href = anchor.get('href')
except AttributeError:
title, href = ('', '')
return (title, href)
def _load_pub_dataframe(pathname=PUB):
"""
Scrape the key data items from the HTML version of the public leaderboard.
Return a pandas DataFrame
"""
with pathname.open() as f:
soup = BeautifulSoup(f, 'lxml-xml')
recs = soup.find_all('tr', class_=['competition-leaderboard__row', 'competition-leaderboard__row competition-leaderboard__row--user-scored'])
rows = []
for rec in recs:
record = dict()
keys = ['Rank', 'Team Name', 'Kernel', 'Score', 'Number of Entries']
for key in keys:
if key != 'Kernel':
record[key] = rec.find('td', {'data-th': key}).text
else:
element = rec.find('td', {'data-th': key})
record['KernelTitle'], record['KernelHref'] = _extract_kernel(element)
record['Last Entry'] = rec.find_all('span')[-1]['title']
rows.append(record)
df = pd.DataFrame(rows)
df['Last Entry'] = pd.to_datetime(df['Last Entry'].str.split().str[0:6].str.join(' '))
df['Team Name'] = _strip_all_spaces(df['Team Name'])
df = df[df['Team Name'].notna()]
df.columns = ['Rank', 'TeamName', 'KernelTitle', 'KernelHref', 'Score', 'Entries', 'Latest']
df = df.drop(['Score', 'Latest'], axis='columns')
return df
def _load_priv_dataframe(pathname=PRIV):
"""
Scrape the key data items from the HTML version of the private leaderboard.
Return a pandas DataFrame
"""
with pathname.open() as f:
soup = BeautifulSoup(f, 'lxml-xml')
recs = soup.find_all('tr', class_=['competition-leaderboard__row', 'competition-leaderboard__row competition-leaderboard__row--user-scored'])
rows = []
for rec in recs:
record = dict()
keys = ['Rank', 'Team Name', 'Score']
for key in keys:
record[key] = rec.find('td', {'data-th': key}).text
change_span = rec.find('td', {'data-th': 'Change'}).find('span')
if change_span.find('span', class_='position-change__none'):
change = (0, 0)
elif change_span.find('span', class_='position-change__risen'):
change = (1, int(change_span.find('span', class_='position-change__risen').text))
else:
change = (-1, int(change_span.find('span', class_='position-change__fallen').text))
record['ChangeDirection'], record['ChangeNo'] = change
rows.append(record)
df = pd.DataFrame(rows)
df['Team Name'] = _strip_all_spaces(df['Team Name'])
df.columns = ['PrivRank', 'TeamName', 'PrivScore', 'ChangeDirection', 'ChangeNo']
return df
def _load_csv_dataframe(pathname=CSV_PUB):
"""
Read the CSV version of the public leaderboard as downloaded from Kaggle.
(This will be merged with the scraped version).
Return a pandas DataFrame
"""
df = pd.read_csv(pathname)
df.TeamName = df.TeamName.replace('\\s+', ' ', regex=True).str.strip()
return df
def _load_dataframe(pathname):
if pathname == PUB:
df = _load_pub_dataframe()
elif pathname == CSV_PUB:
df = _load_csv_dataframe()
return df
def _load_and_merge_public_lb():
"""
Load and clean the two versions (HTML and CSV) of the public leaderboard.
Then merge the two on TeamName.
Return a merged pandas DataFrame.
"""
df = _load_dataframe(PUB)
df_pub = _load_dataframe(CSV_PUB)
df = df[df['TeamName'].isin(df_pub['TeamName'])]
df_pub = df_pub[df_pub['TeamName'].isin(df['TeamName'])]
final = df.merge(df_pub, on='TeamName', how='left')
final.columns = ['PubRank', 'TeamName', 'KernelTitle', 'KernelHref', 'Entries', 'TeamId', 'SubmissionDate', 'PubScore']
return final
def load_data():
"""
Load all data (two versions of public and scraped version of private leaderboard).
Merge into one DataFrame and set dtypes correctly.
Return a pandas DataFrame
"""
pub_df = _load_and_merge_public_lb()
priv_df = _load_priv_dataframe()
df = pub_df.merge(priv_df, on='TeamName', how='left')
type_dict = {'PubRank': 'int32', 'Entries': 'int32', 'SubmissionDate': 'datetime64', 'PubScore': 'float64', 'PrivRank': 'int32', 'PrivScore': 'float64', 'ChangeDirection': 'category', 'ChangeNo': 'int32'}
for key, value in type_dict.items():
df[key] = df[key].astype(value)
new_order = ['TeamId', 'TeamName', 'PubRank', 'PubScore', 'PrivRank', 'PrivScore', 'ChangeDirection', 'ChangeNo', 'SubmissionDate', 'Entries', 'KernelTitle', 'KernelHref']
df = df[new_order]
return df
DF = load_data()
DF.sample(5)
kernels = DF[DF.KernelHref.str.len() > 0].sort_values(by='PrivRank')
kernels = kernels[['PrivRank', 'PrivScore', 'KernelTitle', 'KernelHref']]
markdown = '\n| Private Rank | Private Score | Notebook |\n|--------------|--------------:|---------:|\n'
for row in kernels.iterrows():
rec = row[1]
priv_rank = rec['PrivRank']
priv_score = rec['PrivScore']
title = ' '.join(rec['KernelTitle'].split())
title = title.replace('|', '/')
url = rec['KernelHref']
anchor = f'[{title}](https://kaggle.com{url})'
line = f'| {priv_rank} | {priv_score} | {anchor} |\n'
markdown += line
display(Markdown(markdown)) | code |
74041571/cell_16 | [
"text_html_output_1.png"
] | from bs4 import BeautifulSoup
from pathlib import Path
import pandas as pd
import plotly.express as px
PUB = Path('../input/30dmlleaderboards/public_lb.html')
PRIV = Path('../input/30dmlleaderboards/private_lb.html')
CSV_PUB = Path('../input/30dmlleaderboards/30-days-of-ml-publicleaderboard/30-days-of-ml-publicleaderboard.csv')
def _strip_all_spaces(series):
return series.replace('\\s+', ' ', regex=True).str.strip()
def _extract_kernel(element):
try:
anchor = element.find('a')
title = anchor.get('title')
href = anchor.get('href')
except AttributeError:
title, href = ('', '')
return (title, href)
def _load_pub_dataframe(pathname=PUB):
"""
Scrape the key data items from the HTML version of the public leaderboard.
Return a pandas DataFrame
"""
with pathname.open() as f:
soup = BeautifulSoup(f, 'lxml-xml')
recs = soup.find_all('tr', class_=['competition-leaderboard__row', 'competition-leaderboard__row competition-leaderboard__row--user-scored'])
rows = []
for rec in recs:
record = dict()
keys = ['Rank', 'Team Name', 'Kernel', 'Score', 'Number of Entries']
for key in keys:
if key != 'Kernel':
record[key] = rec.find('td', {'data-th': key}).text
else:
element = rec.find('td', {'data-th': key})
record['KernelTitle'], record['KernelHref'] = _extract_kernel(element)
record['Last Entry'] = rec.find_all('span')[-1]['title']
rows.append(record)
df = pd.DataFrame(rows)
df['Last Entry'] = pd.to_datetime(df['Last Entry'].str.split().str[0:6].str.join(' '))
df['Team Name'] = _strip_all_spaces(df['Team Name'])
df = df[df['Team Name'].notna()]
df.columns = ['Rank', 'TeamName', 'KernelTitle', 'KernelHref', 'Score', 'Entries', 'Latest']
df = df.drop(['Score', 'Latest'], axis='columns')
return df
def _load_priv_dataframe(pathname=PRIV):
"""
Scrape the key data items from the HTML version of the private leaderboard.
Return a pandas DataFrame
"""
with pathname.open() as f:
soup = BeautifulSoup(f, 'lxml-xml')
recs = soup.find_all('tr', class_=['competition-leaderboard__row', 'competition-leaderboard__row competition-leaderboard__row--user-scored'])
rows = []
for rec in recs:
record = dict()
keys = ['Rank', 'Team Name', 'Score']
for key in keys:
record[key] = rec.find('td', {'data-th': key}).text
change_span = rec.find('td', {'data-th': 'Change'}).find('span')
if change_span.find('span', class_='position-change__none'):
change = (0, 0)
elif change_span.find('span', class_='position-change__risen'):
change = (1, int(change_span.find('span', class_='position-change__risen').text))
else:
change = (-1, int(change_span.find('span', class_='position-change__fallen').text))
record['ChangeDirection'], record['ChangeNo'] = change
rows.append(record)
df = pd.DataFrame(rows)
df['Team Name'] = _strip_all_spaces(df['Team Name'])
df.columns = ['PrivRank', 'TeamName', 'PrivScore', 'ChangeDirection', 'ChangeNo']
return df
def _load_csv_dataframe(pathname=CSV_PUB):
"""
Read the CSV version of the public leaderboard as downloaded from Kaggle.
(This will be merged with the scraped version).
Return a pandas DataFrame
"""
df = pd.read_csv(pathname)
df.TeamName = df.TeamName.replace('\\s+', ' ', regex=True).str.strip()
return df
def _load_dataframe(pathname):
if pathname == PUB:
df = _load_pub_dataframe()
elif pathname == CSV_PUB:
df = _load_csv_dataframe()
return df
def _load_and_merge_public_lb():
"""
Load and clean the two versions (HTML and CSV) of the public leaderboard.
Then merge the two on TeamName.
Return a merged pandas DataFrame.
"""
df = _load_dataframe(PUB)
df_pub = _load_dataframe(CSV_PUB)
df = df[df['TeamName'].isin(df_pub['TeamName'])]
df_pub = df_pub[df_pub['TeamName'].isin(df['TeamName'])]
final = df.merge(df_pub, on='TeamName', how='left')
final.columns = ['PubRank', 'TeamName', 'KernelTitle', 'KernelHref', 'Entries', 'TeamId', 'SubmissionDate', 'PubScore']
return final
def load_data():
"""
Load all data (two versions of public and scraped version of private leaderboard).
Merge into one DataFrame and set dtypes correctly.
Return a pandas DataFrame
"""
pub_df = _load_and_merge_public_lb()
priv_df = _load_priv_dataframe()
df = pub_df.merge(priv_df, on='TeamName', how='left')
type_dict = {'PubRank': 'int32', 'Entries': 'int32', 'SubmissionDate': 'datetime64', 'PubScore': 'float64', 'PrivRank': 'int32', 'PrivScore': 'float64', 'ChangeDirection': 'category', 'ChangeNo': 'int32'}
for key, value in type_dict.items():
df[key] = df[key].astype(value)
new_order = ['TeamId', 'TeamName', 'PubRank', 'PubScore', 'PrivRank', 'PrivScore', 'ChangeDirection', 'ChangeNo', 'SubmissionDate', 'Entries', 'KernelTitle', 'KernelHref']
df = df[new_order]
return df
DF = load_data()
DF.sample(5)
RANK = 500
SOURCE = DF[DF['PrivRank'] <= RANK].copy()
x_data = SOURCE["Entries"]
y_data = SOURCE["PrivRank"]
color_data = SOURCE["ChangeDirection"]
size_data = SOURCE["Entries"]
fig = px.scatter(SOURCE, x=x_data, y=y_data, color=color_data,
size=size_data, opacity=0.75, hover_data=['TeamName'])
fig.show()
x_data = SOURCE["SubmissionDate"]
y_data = SOURCE["Entries"]
color_data = SOURCE["PrivRank"]
size_data = SOURCE["Entries"]
fig = px.scatter(SOURCE, x=x_data, y=y_data, color=color_data, size=size_data, opacity=0.75, hover_data=['TeamName'])
fig.show()
x_data = SOURCE["PrivRank"]
y_data = SOURCE["PrivScore"]
color_data = SOURCE["ChangeDirection"]
size_data = SOURCE["Entries"]
fig = px.scatter(SOURCE, x=x_data, y=y_data, color=color_data, size=size_data, opacity=0.75, hover_data=['TeamName'])
fig.show()
x_data = SOURCE['PubRank']
y_data = SOURCE['PubScore']
color_data = SOURCE['ChangeDirection']
size_data = SOURCE['Entries']
fig = px.scatter(SOURCE, x=x_data, y=y_data, color=color_data, size=size_data, opacity=0.75, hover_data=['TeamName'])
fig.show() | code |
74041571/cell_14 | [
"text_html_output_1.png"
] | from bs4 import BeautifulSoup
from pathlib import Path
import pandas as pd
import plotly.express as px
PUB = Path('../input/30dmlleaderboards/public_lb.html')
PRIV = Path('../input/30dmlleaderboards/private_lb.html')
CSV_PUB = Path('../input/30dmlleaderboards/30-days-of-ml-publicleaderboard/30-days-of-ml-publicleaderboard.csv')
def _strip_all_spaces(series):
return series.replace('\\s+', ' ', regex=True).str.strip()
def _extract_kernel(element):
try:
anchor = element.find('a')
title = anchor.get('title')
href = anchor.get('href')
except AttributeError:
title, href = ('', '')
return (title, href)
def _load_pub_dataframe(pathname=PUB):
"""
Scrape the key data items from the HTML version of the public leaderboard.
Return a pandas DataFrame
"""
with pathname.open() as f:
soup = BeautifulSoup(f, 'lxml-xml')
recs = soup.find_all('tr', class_=['competition-leaderboard__row', 'competition-leaderboard__row competition-leaderboard__row--user-scored'])
rows = []
for rec in recs:
record = dict()
keys = ['Rank', 'Team Name', 'Kernel', 'Score', 'Number of Entries']
for key in keys:
if key != 'Kernel':
record[key] = rec.find('td', {'data-th': key}).text
else:
element = rec.find('td', {'data-th': key})
record['KernelTitle'], record['KernelHref'] = _extract_kernel(element)
record['Last Entry'] = rec.find_all('span')[-1]['title']
rows.append(record)
df = pd.DataFrame(rows)
df['Last Entry'] = pd.to_datetime(df['Last Entry'].str.split().str[0:6].str.join(' '))
df['Team Name'] = _strip_all_spaces(df['Team Name'])
df = df[df['Team Name'].notna()]
df.columns = ['Rank', 'TeamName', 'KernelTitle', 'KernelHref', 'Score', 'Entries', 'Latest']
df = df.drop(['Score', 'Latest'], axis='columns')
return df
def _load_priv_dataframe(pathname=PRIV):
"""
Scrape the key data items from the HTML version of the private leaderboard.
Return a pandas DataFrame
"""
with pathname.open() as f:
soup = BeautifulSoup(f, 'lxml-xml')
recs = soup.find_all('tr', class_=['competition-leaderboard__row', 'competition-leaderboard__row competition-leaderboard__row--user-scored'])
rows = []
for rec in recs:
record = dict()
keys = ['Rank', 'Team Name', 'Score']
for key in keys:
record[key] = rec.find('td', {'data-th': key}).text
change_span = rec.find('td', {'data-th': 'Change'}).find('span')
if change_span.find('span', class_='position-change__none'):
change = (0, 0)
elif change_span.find('span', class_='position-change__risen'):
change = (1, int(change_span.find('span', class_='position-change__risen').text))
else:
change = (-1, int(change_span.find('span', class_='position-change__fallen').text))
record['ChangeDirection'], record['ChangeNo'] = change
rows.append(record)
df = pd.DataFrame(rows)
df['Team Name'] = _strip_all_spaces(df['Team Name'])
df.columns = ['PrivRank', 'TeamName', 'PrivScore', 'ChangeDirection', 'ChangeNo']
return df
def _load_csv_dataframe(pathname=CSV_PUB):
"""
Read the CSV version of the public leaderboard as downloaded from Kaggle.
(This will be merged with the scraped version).
Return a pandas DataFrame
"""
df = pd.read_csv(pathname)
df.TeamName = df.TeamName.replace('\\s+', ' ', regex=True).str.strip()
return df
def _load_dataframe(pathname):
if pathname == PUB:
df = _load_pub_dataframe()
elif pathname == CSV_PUB:
df = _load_csv_dataframe()
return df
def _load_and_merge_public_lb():
"""
Load and clean the two versions (HTML and CSV) of the public leaderboard.
Then merge the two on TeamName.
Return a merged pandas DataFrame.
"""
df = _load_dataframe(PUB)
df_pub = _load_dataframe(CSV_PUB)
df = df[df['TeamName'].isin(df_pub['TeamName'])]
df_pub = df_pub[df_pub['TeamName'].isin(df['TeamName'])]
final = df.merge(df_pub, on='TeamName', how='left')
final.columns = ['PubRank', 'TeamName', 'KernelTitle', 'KernelHref', 'Entries', 'TeamId', 'SubmissionDate', 'PubScore']
return final
def load_data():
"""
Load all data (two versions of public and scraped version of private leaderboard).
Merge into one DataFrame and set dtypes correctly.
Return a pandas DataFrame
"""
pub_df = _load_and_merge_public_lb()
priv_df = _load_priv_dataframe()
df = pub_df.merge(priv_df, on='TeamName', how='left')
type_dict = {'PubRank': 'int32', 'Entries': 'int32', 'SubmissionDate': 'datetime64', 'PubScore': 'float64', 'PrivRank': 'int32', 'PrivScore': 'float64', 'ChangeDirection': 'category', 'ChangeNo': 'int32'}
for key, value in type_dict.items():
df[key] = df[key].astype(value)
new_order = ['TeamId', 'TeamName', 'PubRank', 'PubScore', 'PrivRank', 'PrivScore', 'ChangeDirection', 'ChangeNo', 'SubmissionDate', 'Entries', 'KernelTitle', 'KernelHref']
df = df[new_order]
return df
DF = load_data()
DF.sample(5)
RANK = 500
SOURCE = DF[DF['PrivRank'] <= RANK].copy()
x_data = SOURCE["Entries"]
y_data = SOURCE["PrivRank"]
color_data = SOURCE["ChangeDirection"]
size_data = SOURCE["Entries"]
fig = px.scatter(SOURCE, x=x_data, y=y_data, color=color_data,
size=size_data, opacity=0.75, hover_data=['TeamName'])
fig.show()
x_data = SOURCE["SubmissionDate"]
y_data = SOURCE["Entries"]
color_data = SOURCE["PrivRank"]
size_data = SOURCE["Entries"]
fig = px.scatter(SOURCE, x=x_data, y=y_data, color=color_data, size=size_data, opacity=0.75, hover_data=['TeamName'])
fig.show()
x_data = SOURCE['PrivRank']
y_data = SOURCE['PrivScore']
color_data = SOURCE['ChangeDirection']
size_data = SOURCE['Entries']
fig = px.scatter(SOURCE, x=x_data, y=y_data, color=color_data, size=size_data, opacity=0.75, hover_data=['TeamName'])
fig.show() | code |
74041571/cell_10 | [
"text_html_output_1.png"
] | from bs4 import BeautifulSoup
from pathlib import Path
import pandas as pd
import plotly.express as px
PUB = Path('../input/30dmlleaderboards/public_lb.html')
PRIV = Path('../input/30dmlleaderboards/private_lb.html')
CSV_PUB = Path('../input/30dmlleaderboards/30-days-of-ml-publicleaderboard/30-days-of-ml-publicleaderboard.csv')
def _strip_all_spaces(series):
return series.replace('\\s+', ' ', regex=True).str.strip()
def _extract_kernel(element):
try:
anchor = element.find('a')
title = anchor.get('title')
href = anchor.get('href')
except AttributeError:
title, href = ('', '')
return (title, href)
def _load_pub_dataframe(pathname=PUB):
"""
Scrape the key data items from the HTML version of the public leaderboard.
Return a pandas DataFrame
"""
with pathname.open() as f:
soup = BeautifulSoup(f, 'lxml-xml')
recs = soup.find_all('tr', class_=['competition-leaderboard__row', 'competition-leaderboard__row competition-leaderboard__row--user-scored'])
rows = []
for rec in recs:
record = dict()
keys = ['Rank', 'Team Name', 'Kernel', 'Score', 'Number of Entries']
for key in keys:
if key != 'Kernel':
record[key] = rec.find('td', {'data-th': key}).text
else:
element = rec.find('td', {'data-th': key})
record['KernelTitle'], record['KernelHref'] = _extract_kernel(element)
record['Last Entry'] = rec.find_all('span')[-1]['title']
rows.append(record)
df = pd.DataFrame(rows)
df['Last Entry'] = pd.to_datetime(df['Last Entry'].str.split().str[0:6].str.join(' '))
df['Team Name'] = _strip_all_spaces(df['Team Name'])
df = df[df['Team Name'].notna()]
df.columns = ['Rank', 'TeamName', 'KernelTitle', 'KernelHref', 'Score', 'Entries', 'Latest']
df = df.drop(['Score', 'Latest'], axis='columns')
return df
def _load_priv_dataframe(pathname=PRIV):
"""
Scrape the key data items from the HTML version of the private leaderboard.
Return a pandas DataFrame
"""
with pathname.open() as f:
soup = BeautifulSoup(f, 'lxml-xml')
recs = soup.find_all('tr', class_=['competition-leaderboard__row', 'competition-leaderboard__row competition-leaderboard__row--user-scored'])
rows = []
for rec in recs:
record = dict()
keys = ['Rank', 'Team Name', 'Score']
for key in keys:
record[key] = rec.find('td', {'data-th': key}).text
change_span = rec.find('td', {'data-th': 'Change'}).find('span')
if change_span.find('span', class_='position-change__none'):
change = (0, 0)
elif change_span.find('span', class_='position-change__risen'):
change = (1, int(change_span.find('span', class_='position-change__risen').text))
else:
change = (-1, int(change_span.find('span', class_='position-change__fallen').text))
record['ChangeDirection'], record['ChangeNo'] = change
rows.append(record)
df = pd.DataFrame(rows)
df['Team Name'] = _strip_all_spaces(df['Team Name'])
df.columns = ['PrivRank', 'TeamName', 'PrivScore', 'ChangeDirection', 'ChangeNo']
return df
def _load_csv_dataframe(pathname=CSV_PUB):
"""
Read the CSV version of the public leaderboard as downloaded from Kaggle.
(This will be merged with the scraped version).
Return a pandas DataFrame
"""
df = pd.read_csv(pathname)
df.TeamName = df.TeamName.replace('\\s+', ' ', regex=True).str.strip()
return df
def _load_dataframe(pathname):
if pathname == PUB:
df = _load_pub_dataframe()
elif pathname == CSV_PUB:
df = _load_csv_dataframe()
return df
def _load_and_merge_public_lb():
"""
Load and clean the two versions (HTML and CSV) of the public leaderboard.
Then merge the two on TeamName.
Return a merged pandas DataFrame.
"""
df = _load_dataframe(PUB)
df_pub = _load_dataframe(CSV_PUB)
df = df[df['TeamName'].isin(df_pub['TeamName'])]
df_pub = df_pub[df_pub['TeamName'].isin(df['TeamName'])]
final = df.merge(df_pub, on='TeamName', how='left')
final.columns = ['PubRank', 'TeamName', 'KernelTitle', 'KernelHref', 'Entries', 'TeamId', 'SubmissionDate', 'PubScore']
return final
def load_data():
"""
Load all data (two versions of public and scraped version of private leaderboard).
Merge into one DataFrame and set dtypes correctly.
Return a pandas DataFrame
"""
pub_df = _load_and_merge_public_lb()
priv_df = _load_priv_dataframe()
df = pub_df.merge(priv_df, on='TeamName', how='left')
type_dict = {'PubRank': 'int32', 'Entries': 'int32', 'SubmissionDate': 'datetime64', 'PubScore': 'float64', 'PrivRank': 'int32', 'PrivScore': 'float64', 'ChangeDirection': 'category', 'ChangeNo': 'int32'}
for key, value in type_dict.items():
df[key] = df[key].astype(value)
new_order = ['TeamId', 'TeamName', 'PubRank', 'PubScore', 'PrivRank', 'PrivScore', 'ChangeDirection', 'ChangeNo', 'SubmissionDate', 'Entries', 'KernelTitle', 'KernelHref']
df = df[new_order]
return df
DF = load_data()
DF.sample(5)
RANK = 500
SOURCE = DF[DF['PrivRank'] <= RANK].copy()
x_data = SOURCE['Entries']
y_data = SOURCE['PrivRank']
color_data = SOURCE['ChangeDirection']
size_data = SOURCE['Entries']
fig = px.scatter(SOURCE, x=x_data, y=y_data, color=color_data, size=size_data, opacity=0.75, hover_data=['TeamName'])
fig.show() | code |
74041571/cell_12 | [
"text_html_output_2.png"
] | from bs4 import BeautifulSoup
from pathlib import Path
import pandas as pd
import plotly.express as px
PUB = Path('../input/30dmlleaderboards/public_lb.html')
PRIV = Path('../input/30dmlleaderboards/private_lb.html')
CSV_PUB = Path('../input/30dmlleaderboards/30-days-of-ml-publicleaderboard/30-days-of-ml-publicleaderboard.csv')
def _strip_all_spaces(series):
return series.replace('\\s+', ' ', regex=True).str.strip()
def _extract_kernel(element):
try:
anchor = element.find('a')
title = anchor.get('title')
href = anchor.get('href')
except AttributeError:
title, href = ('', '')
return (title, href)
def _load_pub_dataframe(pathname=PUB):
"""
Scrape the key data items from the HTML version of the public leaderboard.
Return a pandas DataFrame
"""
with pathname.open() as f:
soup = BeautifulSoup(f, 'lxml-xml')
recs = soup.find_all('tr', class_=['competition-leaderboard__row', 'competition-leaderboard__row competition-leaderboard__row--user-scored'])
rows = []
for rec in recs:
record = dict()
keys = ['Rank', 'Team Name', 'Kernel', 'Score', 'Number of Entries']
for key in keys:
if key != 'Kernel':
record[key] = rec.find('td', {'data-th': key}).text
else:
element = rec.find('td', {'data-th': key})
record['KernelTitle'], record['KernelHref'] = _extract_kernel(element)
record['Last Entry'] = rec.find_all('span')[-1]['title']
rows.append(record)
df = pd.DataFrame(rows)
df['Last Entry'] = pd.to_datetime(df['Last Entry'].str.split().str[0:6].str.join(' '))
df['Team Name'] = _strip_all_spaces(df['Team Name'])
df = df[df['Team Name'].notna()]
df.columns = ['Rank', 'TeamName', 'KernelTitle', 'KernelHref', 'Score', 'Entries', 'Latest']
df = df.drop(['Score', 'Latest'], axis='columns')
return df
def _load_priv_dataframe(pathname=PRIV):
"""
Scrape the key data items from the HTML version of the private leaderboard.
Return a pandas DataFrame
"""
with pathname.open() as f:
soup = BeautifulSoup(f, 'lxml-xml')
recs = soup.find_all('tr', class_=['competition-leaderboard__row', 'competition-leaderboard__row competition-leaderboard__row--user-scored'])
rows = []
for rec in recs:
record = dict()
keys = ['Rank', 'Team Name', 'Score']
for key in keys:
record[key] = rec.find('td', {'data-th': key}).text
change_span = rec.find('td', {'data-th': 'Change'}).find('span')
if change_span.find('span', class_='position-change__none'):
change = (0, 0)
elif change_span.find('span', class_='position-change__risen'):
change = (1, int(change_span.find('span', class_='position-change__risen').text))
else:
change = (-1, int(change_span.find('span', class_='position-change__fallen').text))
record['ChangeDirection'], record['ChangeNo'] = change
rows.append(record)
df = pd.DataFrame(rows)
df['Team Name'] = _strip_all_spaces(df['Team Name'])
df.columns = ['PrivRank', 'TeamName', 'PrivScore', 'ChangeDirection', 'ChangeNo']
return df
def _load_csv_dataframe(pathname=CSV_PUB):
"""
Read the CSV version of the public leaderboard as downloaded from Kaggle.
(This will be merged with the scraped version).
Return a pandas DataFrame
"""
df = pd.read_csv(pathname)
df.TeamName = df.TeamName.replace('\\s+', ' ', regex=True).str.strip()
return df
def _load_dataframe(pathname):
if pathname == PUB:
df = _load_pub_dataframe()
elif pathname == CSV_PUB:
df = _load_csv_dataframe()
return df
def _load_and_merge_public_lb():
"""
Load and clean the two versions (HTML and CSV) of the public leaderboard.
Then merge the two on TeamName.
Return a merged pandas DataFrame.
"""
df = _load_dataframe(PUB)
df_pub = _load_dataframe(CSV_PUB)
df = df[df['TeamName'].isin(df_pub['TeamName'])]
df_pub = df_pub[df_pub['TeamName'].isin(df['TeamName'])]
final = df.merge(df_pub, on='TeamName', how='left')
final.columns = ['PubRank', 'TeamName', 'KernelTitle', 'KernelHref', 'Entries', 'TeamId', 'SubmissionDate', 'PubScore']
return final
def load_data():
"""
Load all data (two versions of public and scraped version of private leaderboard).
Merge into one DataFrame and set dtypes correctly.
Return a pandas DataFrame
"""
pub_df = _load_and_merge_public_lb()
priv_df = _load_priv_dataframe()
df = pub_df.merge(priv_df, on='TeamName', how='left')
type_dict = {'PubRank': 'int32', 'Entries': 'int32', 'SubmissionDate': 'datetime64', 'PubScore': 'float64', 'PrivRank': 'int32', 'PrivScore': 'float64', 'ChangeDirection': 'category', 'ChangeNo': 'int32'}
for key, value in type_dict.items():
df[key] = df[key].astype(value)
new_order = ['TeamId', 'TeamName', 'PubRank', 'PubScore', 'PrivRank', 'PrivScore', 'ChangeDirection', 'ChangeNo', 'SubmissionDate', 'Entries', 'KernelTitle', 'KernelHref']
df = df[new_order]
return df
DF = load_data()
DF.sample(5)
RANK = 500
SOURCE = DF[DF['PrivRank'] <= RANK].copy()
x_data = SOURCE["Entries"]
y_data = SOURCE["PrivRank"]
color_data = SOURCE["ChangeDirection"]
size_data = SOURCE["Entries"]
fig = px.scatter(SOURCE, x=x_data, y=y_data, color=color_data,
size=size_data, opacity=0.75, hover_data=['TeamName'])
fig.show()
x_data = SOURCE['SubmissionDate']
y_data = SOURCE['Entries']
color_data = SOURCE['PrivRank']
size_data = SOURCE['Entries']
fig = px.scatter(SOURCE, x=x_data, y=y_data, color=color_data, size=size_data, opacity=0.75, hover_data=['TeamName'])
fig.show() | code |
1004763/cell_9 | [
"image_output_1.png"
] | import matplotlib.pyplot as plt
import pandas as pd
DATA_FILE = '../input/uber-raw-data-aug14.csv'
uber_data = pd.read_csv(DATA_FILE)
uber_weekdays = uber_data.pivot_table(index=['DayOfWeekNum', 'DayOfWeek'], values='Base', aggfunc='count')
uber_weekdays.plot(kind='bar', figsize=(8, 6), color='red')
plt.ylabel('Number of Journeys')
plt.title('Journeys by Day') | code |
1004763/cell_4 | [
"text_html_output_1.png"
] | import pandas as pd
DATA_FILE = '../input/uber-raw-data-aug14.csv'
uber_data = pd.read_csv(DATA_FILE)
uber_data.head() | code |
1004763/cell_6 | [
"text_plain_output_1.png"
] | import pandas as pd
DATA_FILE = '../input/uber-raw-data-aug14.csv'
uber_data = pd.read_csv(DATA_FILE)
uber_data['Base'].unique() | code |
1004763/cell_2 | [
"text_plain_output_1.png"
] | import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
from mpl_toolkits.basemap import Basemap
from matplotlib import cm
print('Done') | code |
1004763/cell_8 | [
"text_html_output_1.png"
] | import pandas as pd
DATA_FILE = '../input/uber-raw-data-aug14.csv'
uber_data = pd.read_csv(DATA_FILE)
uber_data.head() | code |
1004763/cell_5 | [
"text_plain_output_1.png"
] | import pandas as pd
DATA_FILE = '../input/uber-raw-data-aug14.csv'
uber_data = pd.read_csv(DATA_FILE)
uber_data.info() | code |
105191248/cell_21 | [
"text_plain_output_1.png"
] | import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
youtube = pd.read_csv('/kaggle/input/youtube-new/GBvideos.csv')
missing_values_count = youtube.isnull().sum()
missing_values_count
total_cells = np.product(youtube.shape)
total_missing = missing_values_count.sum()
total_missing / total_cells * 100
youtube.loc[pd.isna(youtube['description']), :].index
youtube = youtube.fillna('no description available for this video')
youtube.loc[91]
column_names = ['title', 'trending_date']
duplicates2 = youtube.duplicated(subset=column_names, keep=False)
youtube[duplicates2].sort_values(by='title')
youtube = youtube.drop_duplicates(['title', 'trending_date'], keep='first')
youtube[youtube['title'] == '13 Reasons Why: Season 2 | Official Trailer [HD] | Netflix']
youtube['trending_date'] = pd.to_datetime(youtube['trending_date'], format='%y.%d.%m')
youtube['publish_time'] = pd.to_datetime(youtube['publish_time'])
youtube.to_csv('youtube_edited.csv', index=False)
youtube.groupby('category_id').agg({'views': 'max', 'likes': 'max', 'dislikes': 'max'})
grouped = youtube.groupby('title')
grouped.filter(lambda x: len(x) > 30) | code |
105191248/cell_13 | [
"text_plain_output_1.png"
] | import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
youtube = pd.read_csv('/kaggle/input/youtube-new/GBvideos.csv')
missing_values_count = youtube.isnull().sum()
missing_values_count
total_cells = np.product(youtube.shape)
total_missing = missing_values_count.sum()
total_missing / total_cells * 100
youtube.loc[pd.isna(youtube['description']), :].index
youtube = youtube.fillna('no description available for this video')
youtube.loc[91]
youtube.info() | code |
105191248/cell_9 | [
"text_html_output_1.png"
] | import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
youtube = pd.read_csv('/kaggle/input/youtube-new/GBvideos.csv')
missing_values_count = youtube.isnull().sum()
missing_values_count
total_cells = np.product(youtube.shape)
total_missing = missing_values_count.sum()
total_missing / total_cells * 100
indices = list(np.where(youtube['description'].isnull())[0])
indices | code |
105191248/cell_4 | [
"text_html_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
youtube = pd.read_csv('/kaggle/input/youtube-new/GBvideos.csv')
youtube.info() | code |
105191248/cell_20 | [
"text_html_output_1.png"
] | import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
youtube = pd.read_csv('/kaggle/input/youtube-new/GBvideos.csv')
missing_values_count = youtube.isnull().sum()
missing_values_count
total_cells = np.product(youtube.shape)
total_missing = missing_values_count.sum()
total_missing / total_cells * 100
youtube.loc[pd.isna(youtube['description']), :].index
youtube = youtube.fillna('no description available for this video')
youtube.loc[91]
column_names = ['title', 'trending_date']
duplicates2 = youtube.duplicated(subset=column_names, keep=False)
youtube[duplicates2].sort_values(by='title')
youtube = youtube.drop_duplicates(['title', 'trending_date'], keep='first')
youtube[youtube['title'] == '13 Reasons Why: Season 2 | Official Trailer [HD] | Netflix']
youtube['trending_date'] = pd.to_datetime(youtube['trending_date'], format='%y.%d.%m')
youtube['publish_time'] = pd.to_datetime(youtube['publish_time'])
youtube.to_csv('youtube_edited.csv', index=False)
youtube.groupby('category_id').agg({'views': 'max', 'likes': 'max', 'dislikes': 'max'}) | code |
105191248/cell_6 | [
"text_html_output_1.png"
] | import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
youtube = pd.read_csv('/kaggle/input/youtube-new/GBvideos.csv')
missing_values_count = youtube.isnull().sum()
missing_values_count
total_cells = np.product(youtube.shape)
total_missing = missing_values_count.sum()
total_missing / total_cells * 100 | code |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.