path
stringlengths 13
17
| screenshot_names
sequencelengths 1
873
| code
stringlengths 0
40.4k
| cell_type
stringclasses 1
value |
|---|---|---|---|
2041757/cell_13 | [
"text_plain_output_1.png"
] | from keras.callbacks import ModelCheckpoint
from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense
from keras.layers import Dense, Dropout
from keras.models import Sequential
from keras.utils import np_utils
from sklearn.model_selection import train_test_split
import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
from sklearn.model_selection import train_test_split
seed = 5
np.random.seed(seed)
data_train = pd.read_csv('../input/train.csv')
X = data_train.drop('label', axis=1)
y = data_train['label']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=seed)
X_train_matrix = X_train_matrix.astype('float32') / 255
X_test_matrix = X_test_matrix.astype('float32') / 255
from keras.utils import np_utils
y_train_matrix = np_utils.to_categorical(y_train_matrix, 10)
y_test_matrix = np_utils.to_categorical(y_test_matrix, 10)
from keras.models import Sequential
from keras.layers import Dense, Dropout
model = Sequential()
model.add(Dense(784, input_shape=(len(X_train_matrix[0]),)))
model.add(Dense(392, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(196, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(10, activation='softmax'))
model.summary()
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
score = model.evaluate(X_test_matrix, y_test_matrix, verbose=0)
accuracy = score[1] * 100
history = model.fit(X_train_matrix, y_train_matrix, batch_size=100, epochs=6, verbose=1)
from keras.callbacks import ModelCheckpoint
cb_checkpoint = ModelCheckpoint(filepath='best-model.hdf5', verbose=1, save_best_only=True)
history = model.fit(X_train_matrix, y_train_matrix, batch_size=400, epochs=6, validation_split=0.2, shuffle=True, callbacks=[cb_checkpoint], verbose=1)
model.load_weights('best-model.hdf5')
score = model.evaluate(X_test_matrix, y_test_matrix, verbose=0)
accuracy = score[1] * 100
X_train = np.array(list(map(lambda x: np.reshape(x, (28, 28, 1)), X_train_matrix)))
X_test = np.array(list(map(lambda x: np.reshape(x, (28, 28, 1)), X_test_matrix)))
from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense
model = Sequential()
model.add(Conv2D(filters=32, kernel_size=5, padding='same', activation='relu', input_shape=(28, 28, 1)))
model.add(MaxPooling2D(pool_size=2))
model.add(Conv2D(filters=64, kernel_size=5, padding='same', activation='relu'))
model.add(MaxPooling2D(pool_size=2))
model.add(Flatten())
model.add(Dense(1024, activation='relu'))
model.add(Dropout(0.3))
model.add(Dense(10, activation='softmax'))
model.summary()
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
score = model.evaluate(X_test, y_test_matrix, verbose=0)
accuracy = score[1] * 100
print('Random test accuracy is %1.1f%%' % accuracy) | code |
2041757/cell_9 | [
"text_plain_output_1.png"
] | from keras.callbacks import ModelCheckpoint
from keras.layers import Dense, Dropout
from keras.models import Sequential
from keras.utils import np_utils
X_train_matrix = X_train_matrix.astype('float32') / 255
X_test_matrix = X_test_matrix.astype('float32') / 255
from keras.utils import np_utils
y_train_matrix = np_utils.to_categorical(y_train_matrix, 10)
y_test_matrix = np_utils.to_categorical(y_test_matrix, 10)
from keras.models import Sequential
from keras.layers import Dense, Dropout
model = Sequential()
model.add(Dense(784, input_shape=(len(X_train_matrix[0]),)))
model.add(Dense(392, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(196, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(10, activation='softmax'))
model.summary()
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
score = model.evaluate(X_test_matrix, y_test_matrix, verbose=0)
accuracy = score[1] * 100
history = model.fit(X_train_matrix, y_train_matrix, batch_size=100, epochs=6, verbose=1)
from keras.callbacks import ModelCheckpoint
cb_checkpoint = ModelCheckpoint(filepath='best-model.hdf5', verbose=1, save_best_only=True)
history = model.fit(X_train_matrix, y_train_matrix, batch_size=400, epochs=6, validation_split=0.2, shuffle=True, callbacks=[cb_checkpoint], verbose=1) | code |
2041757/cell_4 | [
"text_plain_output_1.png"
] | print('before:')
print(X_train_matrix[0])
X_train_matrix = X_train_matrix.astype('float32') / 255
X_test_matrix = X_test_matrix.astype('float32') / 255
print('after:')
print(X_train_matrix[0]) | code |
2041757/cell_6 | [
"text_plain_output_1.png"
] | from keras.layers import Dense, Dropout
from keras.models import Sequential
X_train_matrix = X_train_matrix.astype('float32') / 255
X_test_matrix = X_test_matrix.astype('float32') / 255
from keras.models import Sequential
from keras.layers import Dense, Dropout
model = Sequential()
model.add(Dense(784, input_shape=(len(X_train_matrix[0]),)))
model.add(Dense(392, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(196, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(10, activation='softmax'))
model.summary() | code |
2041757/cell_2 | [
"text_plain_output_1.png"
] | from sklearn.model_selection import train_test_split
import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
from sklearn.model_selection import train_test_split
seed = 5
np.random.seed(seed)
data_train = pd.read_csv('../input/train.csv')
X = data_train.drop('label', axis=1)
y = data_train['label']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=seed)
print('There are %d train entries and %d test entries' % (len(X_train), len(y_test))) | code |
2041757/cell_19 | [
"text_plain_output_1.png"
] | from keras.callbacks import ModelCheckpoint
from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense
from keras.layers import Dense, Dropout
from keras.models import Sequential
from keras.utils import np_utils
from sklearn.model_selection import train_test_split
import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
from sklearn.model_selection import train_test_split
seed = 5
np.random.seed(seed)
data_train = pd.read_csv('../input/train.csv')
X = data_train.drop('label', axis=1)
y = data_train['label']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=seed)
X_train_matrix = X_train_matrix.astype('float32') / 255
X_test_matrix = X_test_matrix.astype('float32') / 255
from keras.utils import np_utils
y_train_matrix = np_utils.to_categorical(y_train_matrix, 10)
y_test_matrix = np_utils.to_categorical(y_test_matrix, 10)
from keras.models import Sequential
from keras.layers import Dense, Dropout
model = Sequential()
model.add(Dense(784, input_shape=(len(X_train_matrix[0]),)))
model.add(Dense(392, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(196, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(10, activation='softmax'))
model.summary()
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
score = model.evaluate(X_test_matrix, y_test_matrix, verbose=0)
accuracy = score[1] * 100
history = model.fit(X_train_matrix, y_train_matrix, batch_size=100, epochs=6, verbose=1)
from keras.callbacks import ModelCheckpoint
cb_checkpoint = ModelCheckpoint(filepath='best-model.hdf5', verbose=1, save_best_only=True)
history = model.fit(X_train_matrix, y_train_matrix, batch_size=400, epochs=6, validation_split=0.2, shuffle=True, callbacks=[cb_checkpoint], verbose=1)
model.load_weights('best-model.hdf5')
score = model.evaluate(X_test_matrix, y_test_matrix, verbose=0)
accuracy = score[1] * 100
X_train = np.array(list(map(lambda x: np.reshape(x, (28, 28, 1)), X_train_matrix)))
X_test = np.array(list(map(lambda x: np.reshape(x, (28, 28, 1)), X_test_matrix)))
from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense
model = Sequential()
model.add(Conv2D(filters=32, kernel_size=5, padding='same', activation='relu', input_shape=(28, 28, 1)))
model.add(MaxPooling2D(pool_size=2))
model.add(Conv2D(filters=64, kernel_size=5, padding='same', activation='relu'))
model.add(MaxPooling2D(pool_size=2))
model.add(Flatten())
model.add(Dense(1024, activation='relu'))
model.add(Dropout(0.3))
model.add(Dense(10, activation='softmax'))
model.summary()
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
score = model.evaluate(X_test, y_test_matrix, verbose=0)
accuracy = score[1] * 100
cb_checkpoint = ModelCheckpoint(filepath='best-cnn-model.hdf5', verbose=1, save_best_only=True)
history = model.fit(X_train, y_train_matrix, batch_size=50, epochs=6, validation_split=0.2, shuffle=True, callbacks=[cb_checkpoint], verbose=1)
model.load_weights('best-cnn-model.hdf5')
score = model.evaluate(X_test, y_test_matrix, verbose=0)
accuracy = score[1] * 100
data_test = pd.read_csv('../input/test.csv')
test = np.array(list(map(lambda x: np.reshape(x, (28, 28, 1)), data_test.as_matrix())))
test.shape
pred = model.predict(test, batch_size=32, verbose=1)
pred[0]
predicted_labels = [np.argmax(r, axis=0) for r in pred]
predicted_labels | code |
2041757/cell_1 | [
"text_plain_output_1.png"
] | from subprocess import check_output
import numpy as np
import pandas as pd
from subprocess import check_output
print(check_output(['ls', '../input']).decode('utf8')) | code |
2041757/cell_7 | [
"text_plain_output_1.png"
] | from keras.layers import Dense, Dropout
from keras.models import Sequential
from keras.utils import np_utils
X_train_matrix = X_train_matrix.astype('float32') / 255
X_test_matrix = X_test_matrix.astype('float32') / 255
from keras.utils import np_utils
y_train_matrix = np_utils.to_categorical(y_train_matrix, 10)
y_test_matrix = np_utils.to_categorical(y_test_matrix, 10)
from keras.models import Sequential
from keras.layers import Dense, Dropout
model = Sequential()
model.add(Dense(784, input_shape=(len(X_train_matrix[0]),)))
model.add(Dense(392, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(196, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(10, activation='softmax'))
model.summary()
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
score = model.evaluate(X_test_matrix, y_test_matrix, verbose=0)
accuracy = score[1] * 100
print('Random test accuracy is %1.1f%%' % accuracy) | code |
2041757/cell_18 | [
"text_plain_output_1.png"
] | from keras.callbacks import ModelCheckpoint
from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense
from keras.layers import Dense, Dropout
from keras.models import Sequential
from keras.utils import np_utils
from sklearn.model_selection import train_test_split
import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
from sklearn.model_selection import train_test_split
seed = 5
np.random.seed(seed)
data_train = pd.read_csv('../input/train.csv')
X = data_train.drop('label', axis=1)
y = data_train['label']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=seed)
X_train_matrix = X_train_matrix.astype('float32') / 255
X_test_matrix = X_test_matrix.astype('float32') / 255
from keras.utils import np_utils
y_train_matrix = np_utils.to_categorical(y_train_matrix, 10)
y_test_matrix = np_utils.to_categorical(y_test_matrix, 10)
from keras.models import Sequential
from keras.layers import Dense, Dropout
model = Sequential()
model.add(Dense(784, input_shape=(len(X_train_matrix[0]),)))
model.add(Dense(392, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(196, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(10, activation='softmax'))
model.summary()
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
score = model.evaluate(X_test_matrix, y_test_matrix, verbose=0)
accuracy = score[1] * 100
history = model.fit(X_train_matrix, y_train_matrix, batch_size=100, epochs=6, verbose=1)
from keras.callbacks import ModelCheckpoint
cb_checkpoint = ModelCheckpoint(filepath='best-model.hdf5', verbose=1, save_best_only=True)
history = model.fit(X_train_matrix, y_train_matrix, batch_size=400, epochs=6, validation_split=0.2, shuffle=True, callbacks=[cb_checkpoint], verbose=1)
model.load_weights('best-model.hdf5')
score = model.evaluate(X_test_matrix, y_test_matrix, verbose=0)
accuracy = score[1] * 100
X_train = np.array(list(map(lambda x: np.reshape(x, (28, 28, 1)), X_train_matrix)))
X_test = np.array(list(map(lambda x: np.reshape(x, (28, 28, 1)), X_test_matrix)))
from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense
model = Sequential()
model.add(Conv2D(filters=32, kernel_size=5, padding='same', activation='relu', input_shape=(28, 28, 1)))
model.add(MaxPooling2D(pool_size=2))
model.add(Conv2D(filters=64, kernel_size=5, padding='same', activation='relu'))
model.add(MaxPooling2D(pool_size=2))
model.add(Flatten())
model.add(Dense(1024, activation='relu'))
model.add(Dropout(0.3))
model.add(Dense(10, activation='softmax'))
model.summary()
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
score = model.evaluate(X_test, y_test_matrix, verbose=0)
accuracy = score[1] * 100
cb_checkpoint = ModelCheckpoint(filepath='best-cnn-model.hdf5', verbose=1, save_best_only=True)
history = model.fit(X_train, y_train_matrix, batch_size=50, epochs=6, validation_split=0.2, shuffle=True, callbacks=[cb_checkpoint], verbose=1)
model.load_weights('best-cnn-model.hdf5')
score = model.evaluate(X_test, y_test_matrix, verbose=0)
accuracy = score[1] * 100
data_test = pd.read_csv('../input/test.csv')
test = np.array(list(map(lambda x: np.reshape(x, (28, 28, 1)), data_test.as_matrix())))
test.shape
pred = model.predict(test, batch_size=32, verbose=1)
pred[0] | code |
2041757/cell_8 | [
"text_plain_output_2.png",
"text_plain_output_1.png"
] | from keras.layers import Dense, Dropout
from keras.models import Sequential
from keras.utils import np_utils
X_train_matrix = X_train_matrix.astype('float32') / 255
X_test_matrix = X_test_matrix.astype('float32') / 255
from keras.utils import np_utils
y_train_matrix = np_utils.to_categorical(y_train_matrix, 10)
y_test_matrix = np_utils.to_categorical(y_test_matrix, 10)
from keras.models import Sequential
from keras.layers import Dense, Dropout
model = Sequential()
model.add(Dense(784, input_shape=(len(X_train_matrix[0]),)))
model.add(Dense(392, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(196, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(10, activation='softmax'))
model.summary()
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
score = model.evaluate(X_test_matrix, y_test_matrix, verbose=0)
accuracy = score[1] * 100
history = model.fit(X_train_matrix, y_train_matrix, batch_size=100, epochs=6, verbose=1) | code |
2041757/cell_15 | [
"text_plain_output_1.png"
] | from keras.callbacks import ModelCheckpoint
from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense
from keras.layers import Dense, Dropout
from keras.models import Sequential
from keras.utils import np_utils
from sklearn.model_selection import train_test_split
import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
from sklearn.model_selection import train_test_split
seed = 5
np.random.seed(seed)
data_train = pd.read_csv('../input/train.csv')
X = data_train.drop('label', axis=1)
y = data_train['label']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=seed)
X_train_matrix = X_train_matrix.astype('float32') / 255
X_test_matrix = X_test_matrix.astype('float32') / 255
from keras.utils import np_utils
y_train_matrix = np_utils.to_categorical(y_train_matrix, 10)
y_test_matrix = np_utils.to_categorical(y_test_matrix, 10)
from keras.models import Sequential
from keras.layers import Dense, Dropout
model = Sequential()
model.add(Dense(784, input_shape=(len(X_train_matrix[0]),)))
model.add(Dense(392, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(196, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(10, activation='softmax'))
model.summary()
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
score = model.evaluate(X_test_matrix, y_test_matrix, verbose=0)
accuracy = score[1] * 100
history = model.fit(X_train_matrix, y_train_matrix, batch_size=100, epochs=6, verbose=1)
from keras.callbacks import ModelCheckpoint
cb_checkpoint = ModelCheckpoint(filepath='best-model.hdf5', verbose=1, save_best_only=True)
history = model.fit(X_train_matrix, y_train_matrix, batch_size=400, epochs=6, validation_split=0.2, shuffle=True, callbacks=[cb_checkpoint], verbose=1)
model.load_weights('best-model.hdf5')
score = model.evaluate(X_test_matrix, y_test_matrix, verbose=0)
accuracy = score[1] * 100
X_train = np.array(list(map(lambda x: np.reshape(x, (28, 28, 1)), X_train_matrix)))
X_test = np.array(list(map(lambda x: np.reshape(x, (28, 28, 1)), X_test_matrix)))
from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense
model = Sequential()
model.add(Conv2D(filters=32, kernel_size=5, padding='same', activation='relu', input_shape=(28, 28, 1)))
model.add(MaxPooling2D(pool_size=2))
model.add(Conv2D(filters=64, kernel_size=5, padding='same', activation='relu'))
model.add(MaxPooling2D(pool_size=2))
model.add(Flatten())
model.add(Dense(1024, activation='relu'))
model.add(Dropout(0.3))
model.add(Dense(10, activation='softmax'))
model.summary()
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
score = model.evaluate(X_test, y_test_matrix, verbose=0)
accuracy = score[1] * 100
cb_checkpoint = ModelCheckpoint(filepath='best-cnn-model.hdf5', verbose=1, save_best_only=True)
history = model.fit(X_train, y_train_matrix, batch_size=50, epochs=6, validation_split=0.2, shuffle=True, callbacks=[cb_checkpoint], verbose=1)
model.load_weights('best-cnn-model.hdf5')
score = model.evaluate(X_test, y_test_matrix, verbose=0)
accuracy = score[1] * 100
print("Best Model's test accuracy is %1.1f%%" % accuracy) | code |
2041757/cell_3 | [
"text_plain_output_1.png"
] | import matplotlib.pyplot as plt
import matplotlib.cm as cm
X_train_matrix = X_train.as_matrix()
y_train_matrix = y_train.as_matrix()
X_test_matrix = X_test.as_matrix()
y_test_matrix = y_test.as_matrix()
fig = plt.figure(figsize=(20, 20))
for i in range(10):
axis = fig.add_subplot(1, 10, i + 1, xticks=[], yticks=[])
axis.imshow(np.reshape(X_train_matrix[i], (28, 28)), cmap='gray')
axis.set_title(str(y_train_matrix[i])) | code |
2041757/cell_17 | [
"text_plain_output_1.png"
] | from sklearn.model_selection import train_test_split
import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
from sklearn.model_selection import train_test_split
seed = 5
np.random.seed(seed)
data_train = pd.read_csv('../input/train.csv')
X = data_train.drop('label', axis=1)
y = data_train['label']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=seed)
X_train_matrix = X_train_matrix.astype('float32') / 255
X_test_matrix = X_test_matrix.astype('float32') / 255
X_train = np.array(list(map(lambda x: np.reshape(x, (28, 28, 1)), X_train_matrix)))
X_test = np.array(list(map(lambda x: np.reshape(x, (28, 28, 1)), X_test_matrix)))
data_test = pd.read_csv('../input/test.csv')
test = np.array(list(map(lambda x: np.reshape(x, (28, 28, 1)), data_test.as_matrix())))
test.shape | code |
2041757/cell_14 | [
"text_plain_output_2.png",
"application_vnd.jupyter.stderr_output_1.png"
] | from keras.callbacks import ModelCheckpoint
from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense
from keras.layers import Dense, Dropout
from keras.models import Sequential
from keras.utils import np_utils
from sklearn.model_selection import train_test_split
import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
from sklearn.model_selection import train_test_split
seed = 5
np.random.seed(seed)
data_train = pd.read_csv('../input/train.csv')
X = data_train.drop('label', axis=1)
y = data_train['label']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=seed)
X_train_matrix = X_train_matrix.astype('float32') / 255
X_test_matrix = X_test_matrix.astype('float32') / 255
from keras.utils import np_utils
y_train_matrix = np_utils.to_categorical(y_train_matrix, 10)
y_test_matrix = np_utils.to_categorical(y_test_matrix, 10)
from keras.models import Sequential
from keras.layers import Dense, Dropout
model = Sequential()
model.add(Dense(784, input_shape=(len(X_train_matrix[0]),)))
model.add(Dense(392, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(196, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(10, activation='softmax'))
model.summary()
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
score = model.evaluate(X_test_matrix, y_test_matrix, verbose=0)
accuracy = score[1] * 100
history = model.fit(X_train_matrix, y_train_matrix, batch_size=100, epochs=6, verbose=1)
from keras.callbacks import ModelCheckpoint
cb_checkpoint = ModelCheckpoint(filepath='best-model.hdf5', verbose=1, save_best_only=True)
history = model.fit(X_train_matrix, y_train_matrix, batch_size=400, epochs=6, validation_split=0.2, shuffle=True, callbacks=[cb_checkpoint], verbose=1)
model.load_weights('best-model.hdf5')
score = model.evaluate(X_test_matrix, y_test_matrix, verbose=0)
accuracy = score[1] * 100
X_train = np.array(list(map(lambda x: np.reshape(x, (28, 28, 1)), X_train_matrix)))
X_test = np.array(list(map(lambda x: np.reshape(x, (28, 28, 1)), X_test_matrix)))
from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense
model = Sequential()
model.add(Conv2D(filters=32, kernel_size=5, padding='same', activation='relu', input_shape=(28, 28, 1)))
model.add(MaxPooling2D(pool_size=2))
model.add(Conv2D(filters=64, kernel_size=5, padding='same', activation='relu'))
model.add(MaxPooling2D(pool_size=2))
model.add(Flatten())
model.add(Dense(1024, activation='relu'))
model.add(Dropout(0.3))
model.add(Dense(10, activation='softmax'))
model.summary()
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
score = model.evaluate(X_test, y_test_matrix, verbose=0)
accuracy = score[1] * 100
cb_checkpoint = ModelCheckpoint(filepath='best-cnn-model.hdf5', verbose=1, save_best_only=True)
history = model.fit(X_train, y_train_matrix, batch_size=50, epochs=6, validation_split=0.2, shuffle=True, callbacks=[cb_checkpoint], verbose=1) | code |
2041757/cell_10 | [
"text_plain_output_1.png"
] | from keras.callbacks import ModelCheckpoint
from keras.layers import Dense, Dropout
from keras.models import Sequential
from keras.utils import np_utils
X_train_matrix = X_train_matrix.astype('float32') / 255
X_test_matrix = X_test_matrix.astype('float32') / 255
from keras.utils import np_utils
y_train_matrix = np_utils.to_categorical(y_train_matrix, 10)
y_test_matrix = np_utils.to_categorical(y_test_matrix, 10)
from keras.models import Sequential
from keras.layers import Dense, Dropout
model = Sequential()
model.add(Dense(784, input_shape=(len(X_train_matrix[0]),)))
model.add(Dense(392, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(196, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(10, activation='softmax'))
model.summary()
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
score = model.evaluate(X_test_matrix, y_test_matrix, verbose=0)
accuracy = score[1] * 100
history = model.fit(X_train_matrix, y_train_matrix, batch_size=100, epochs=6, verbose=1)
from keras.callbacks import ModelCheckpoint
cb_checkpoint = ModelCheckpoint(filepath='best-model.hdf5', verbose=1, save_best_only=True)
history = model.fit(X_train_matrix, y_train_matrix, batch_size=400, epochs=6, validation_split=0.2, shuffle=True, callbacks=[cb_checkpoint], verbose=1)
model.load_weights('best-model.hdf5')
score = model.evaluate(X_test_matrix, y_test_matrix, verbose=0)
accuracy = score[1] * 100
print("Best Model's test accuracy is %1.1f%%" % accuracy) | code |
2041757/cell_12 | [
"image_output_1.png"
] | from keras.callbacks import ModelCheckpoint
from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense
from keras.layers import Dense, Dropout
from keras.models import Sequential
from keras.utils import np_utils
X_train_matrix = X_train_matrix.astype('float32') / 255
X_test_matrix = X_test_matrix.astype('float32') / 255
from keras.utils import np_utils
y_train_matrix = np_utils.to_categorical(y_train_matrix, 10)
y_test_matrix = np_utils.to_categorical(y_test_matrix, 10)
from keras.models import Sequential
from keras.layers import Dense, Dropout
model = Sequential()
model.add(Dense(784, input_shape=(len(X_train_matrix[0]),)))
model.add(Dense(392, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(196, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(10, activation='softmax'))
model.summary()
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
score = model.evaluate(X_test_matrix, y_test_matrix, verbose=0)
accuracy = score[1] * 100
history = model.fit(X_train_matrix, y_train_matrix, batch_size=100, epochs=6, verbose=1)
from keras.callbacks import ModelCheckpoint
cb_checkpoint = ModelCheckpoint(filepath='best-model.hdf5', verbose=1, save_best_only=True)
history = model.fit(X_train_matrix, y_train_matrix, batch_size=400, epochs=6, validation_split=0.2, shuffle=True, callbacks=[cb_checkpoint], verbose=1)
model.load_weights('best-model.hdf5')
score = model.evaluate(X_test_matrix, y_test_matrix, verbose=0)
accuracy = score[1] * 100
from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense
model = Sequential()
model.add(Conv2D(filters=32, kernel_size=5, padding='same', activation='relu', input_shape=(28, 28, 1)))
model.add(MaxPooling2D(pool_size=2))
model.add(Conv2D(filters=64, kernel_size=5, padding='same', activation='relu'))
model.add(MaxPooling2D(pool_size=2))
model.add(Flatten())
model.add(Dense(1024, activation='relu'))
model.add(Dropout(0.3))
model.add(Dense(10, activation='softmax'))
model.summary() | code |
2041757/cell_5 | [
"text_plain_output_1.png"
] | from keras.utils import np_utils
from keras.utils import np_utils
print('Before:')
print(y_train_matrix[:8])
y_train_matrix = np_utils.to_categorical(y_train_matrix, 10)
y_test_matrix = np_utils.to_categorical(y_test_matrix, 10)
print('After:')
print(y_train_matrix[:8]) | code |
122247504/cell_4 | [
"text_plain_output_1.png"
] | from PIL import Image
import numpy as np # linear algebra
import os
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from PIL import Image
import tensorflow as tf
import keras
import os
os.listdir('/kaggle/input/')
import cv2
real = '/kaggle/input/real-and-fake-face-detection/real_and_fake_face/training_real/'
imagePaths = os.listdir(real)
for pic in imagePaths:
path = real + pic
pic = Image.open(path)
real_pictures = np.array(pic)
print(real_pictures.shape) | code |
122247504/cell_6 | [
"application_vnd.jupyter.stderr_output_1.png"
] | from PIL import Image
import numpy as np # linear algebra
import os
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from PIL import Image
import tensorflow as tf
import keras
import os
os.listdir('/kaggle/input/')
import cv2
real = '/kaggle/input/real-and-fake-face-detection/real_and_fake_face/training_real/'
imagePaths = os.listdir(real)
for pic in imagePaths:
path = real + pic
pic = Image.open(path)
real_pictures = np.array(pic)
realdf = pd.DataFrame(real_pictures) | code |
90127204/cell_21 | [
"image_output_1.png"
] | import matplotlib.pyplot as plt
import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O
import random
import seaborn as sns
import tensorflow as tf
train = pd.read_csv('/kaggle/input/digit-recognizer/train.csv')
test = pd.read_csv('/kaggle/input/digit-recognizer/test.csv')
y_train = train['label']
X_train = train.drop('label', axis=1)
X_train = X_train / 255
y_train.nunique()
fig, ax = plt.subplots(5, 5, figsize=(8, 8))
fig.suptitle('Digits images and labels', fontsize=16)
ax = ax.ravel()
for i in range(25):
sample_n = random.randint(0, X_train.shape[0])
ax[i].imshow(np.array(X_train.iloc[sample_n]).reshape(28, 28), cmap='inferno')
ax[i].get_xaxis().set_visible(False)
ax[i].get_yaxis().set_visible(False)
ax[i].set_title(y_train[sample_n], fontsize = 12)
plt.subplots_adjust(hspace=0.3)
plt.colorbar()
tf.random.set_seed(42)
model = tf.keras.Sequential([tf.keras.layers.Dense(units=784, activation='relu', input_shape=(784,)), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=392, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=151, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=50, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=10, activation='softmax')])
model.compile(optimizer='Adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])
es = tf.keras.callbacks.EarlyStopping(monitor='val_loss', min_delta=0, patience=5, verbose=0, mode='auto', baseline=None, restore_best_weights=False)
history = model.fit(X_train, y_train, epochs=100, validation_split=0.3, batch_size=50, callbacks=[es]) | code |
90127204/cell_9 | [
"text_html_output_1.png"
] | import pandas as pd # data processing, CSV file I/O
train = pd.read_csv('/kaggle/input/digit-recognizer/train.csv')
test = pd.read_csv('/kaggle/input/digit-recognizer/test.csv')
y_train = train['label']
X_train = train.drop('label', axis=1)
y_train.nunique() | code |
90127204/cell_25 | [
"text_plain_output_2.png",
"application_vnd.jupyter.stderr_output_1.png"
] | import matplotlib.pyplot as plt
import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O
import random
import seaborn as sns
import tensorflow as tf
train = pd.read_csv('/kaggle/input/digit-recognizer/train.csv')
test = pd.read_csv('/kaggle/input/digit-recognizer/test.csv')
y_train = train['label']
X_train = train.drop('label', axis=1)
X_train = X_train / 255
y_train.nunique()
fig, ax = plt.subplots(5, 5, figsize=(8, 8))
fig.suptitle('Digits images and labels', fontsize=16)
ax = ax.ravel()
for i in range(25):
sample_n = random.randint(0, X_train.shape[0])
ax[i].imshow(np.array(X_train.iloc[sample_n]).reshape(28, 28), cmap='inferno')
ax[i].get_xaxis().set_visible(False)
ax[i].get_yaxis().set_visible(False)
ax[i].set_title(y_train[sample_n], fontsize = 12)
plt.subplots_adjust(hspace=0.3)
plt.colorbar()
tf.random.set_seed(42)
model = tf.keras.Sequential([tf.keras.layers.Dense(units=784, activation='relu', input_shape=(784,)), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=392, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=151, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=50, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=10, activation='softmax')])
model.compile(optimizer='Adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])
es = tf.keras.callbacks.EarlyStopping(monitor='val_loss', min_delta=0, patience=5, verbose=0, mode='auto', baseline=None, restore_best_weights=False)
history = model.fit(X_train, y_train, epochs=100, validation_split=0.3, batch_size=50, callbacks=[es])
model.evaluate(X_train, y_train) | code |
90127204/cell_23 | [
"image_output_1.png"
] | import matplotlib.pyplot as plt
import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O
import random
import seaborn as sns
import tensorflow as tf
train = pd.read_csv('/kaggle/input/digit-recognizer/train.csv')
test = pd.read_csv('/kaggle/input/digit-recognizer/test.csv')
y_train = train['label']
X_train = train.drop('label', axis=1)
X_train = X_train / 255
y_train.nunique()
fig, ax = plt.subplots(5, 5, figsize=(8, 8))
fig.suptitle('Digits images and labels', fontsize=16)
ax = ax.ravel()
for i in range(25):
sample_n = random.randint(0, X_train.shape[0])
ax[i].imshow(np.array(X_train.iloc[sample_n]).reshape(28, 28), cmap='inferno')
ax[i].get_xaxis().set_visible(False)
ax[i].get_yaxis().set_visible(False)
ax[i].set_title(y_train[sample_n], fontsize = 12)
plt.subplots_adjust(hspace=0.3)
plt.colorbar()
fig, ax = plt.subplots(2, 5, figsize=(8, 4))
fig.suptitle('Average shape per digit', fontsize=16)
ax = ax.ravel()
for i in range(10):
ax[i].imshow(np.array(train[train['label'] == i].drop('label', axis=1).mean()).reshape(28, 28), cmap='inferno')
ax[i].get_xaxis().set_visible(False)
ax[i].get_yaxis().set_visible(False)
ax[i].set_title(i, fontsize = 12)
tf.random.set_seed(42)
model = tf.keras.Sequential([tf.keras.layers.Dense(units=784, activation='relu', input_shape=(784,)), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=392, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=151, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=50, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=10, activation='softmax')])
model.compile(optimizer='Adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])
es = tf.keras.callbacks.EarlyStopping(monitor='val_loss', min_delta=0, patience=5, verbose=0, mode='auto', baseline=None, restore_best_weights=False)
history = model.fit(X_train, y_train, epochs=100, validation_split=0.3, batch_size=50, callbacks=[es])
history.history.keys()
plt.title('Loss')
plt.plot(range(len(history.history['loss'])), history.history['loss'], marker='o', c='gray')
plt.plot(range(len(history.history['loss'])), history.history['val_loss'], marker='o')
plt.show() | code |
90127204/cell_30 | [
"text_plain_output_1.png"
] | import matplotlib.pyplot as plt
import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O
import random
import seaborn as sns
import tensorflow as tf
train = pd.read_csv('/kaggle/input/digit-recognizer/train.csv')
test = pd.read_csv('/kaggle/input/digit-recognizer/test.csv')
y_train = train['label']
X_train = train.drop('label', axis=1)
X_train = X_train / 255
y_train.nunique()
fig, ax = plt.subplots(5, 5, figsize=(8, 8))
fig.suptitle('Digits images and labels', fontsize=16)
ax = ax.ravel()
for i in range(25):
sample_n = random.randint(0, X_train.shape[0])
ax[i].imshow(np.array(X_train.iloc[sample_n]).reshape(28, 28), cmap='inferno')
ax[i].get_xaxis().set_visible(False)
ax[i].get_yaxis().set_visible(False)
ax[i].set_title(y_train[sample_n], fontsize = 12)
plt.subplots_adjust(hspace=0.3)
plt.colorbar()
fig, ax = plt.subplots(2, 5, figsize=(8, 4))
fig.suptitle('Average shape per digit', fontsize=16)
ax = ax.ravel()
for i in range(10):
ax[i].imshow(np.array(train[train['label'] == i].drop('label', axis=1).mean()).reshape(28, 28), cmap='inferno')
ax[i].get_xaxis().set_visible(False)
ax[i].get_yaxis().set_visible(False)
ax[i].set_title(i, fontsize = 12)
tf.random.set_seed(42)
model = tf.keras.Sequential([tf.keras.layers.Dense(units=784, activation='relu', input_shape=(784,)), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=392, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=151, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=50, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=10, activation='softmax')])
model.compile(optimizer='Adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])
es = tf.keras.callbacks.EarlyStopping(monitor='val_loss', min_delta=0, patience=5, verbose=0, mode='auto', baseline=None, restore_best_weights=False)
history = model.fit(X_train, y_train, epochs=100, validation_split=0.3, batch_size=50, callbacks=[es])
history.history.keys()
model.evaluate(X_train, y_train)
test = test / 255
test_result = model.predict(test)
test_result = np.array(pd.DataFrame(test_result).idxmax(axis=1))
fig, ax = plt.subplots(5, 5, figsize=(8, 8))
fig.suptitle('Digits images and labels', fontsize=16)
ax = ax.ravel()
for i in range(25):
sample_n = random.randint(0, test.shape[0])
ax[i].imshow(np.array(test.iloc[sample_n]).reshape(28, 28), cmap='inferno')
ax[i].get_xaxis().set_visible(False)
ax[i].get_yaxis().set_visible(False)
ax[i].set_title(test_result[sample_n], fontsize=12)
plt.subplots_adjust(hspace=0.3)
fig.show() | code |
90127204/cell_6 | [
"text_plain_output_2.png",
"text_plain_output_1.png"
] | import pandas as pd # data processing, CSV file I/O
train = pd.read_csv('/kaggle/input/digit-recognizer/train.csv')
test = pd.read_csv('/kaggle/input/digit-recognizer/test.csv')
y_train = train['label']
X_train = train.drop('label', axis=1)
X_train.head() | code |
90127204/cell_11 | [
"text_plain_output_1.png"
] | import matplotlib.pyplot as plt
import pandas as pd # data processing, CSV file I/O
import seaborn as sns
train = pd.read_csv('/kaggle/input/digit-recognizer/train.csv')
test = pd.read_csv('/kaggle/input/digit-recognizer/test.csv')
y_train = train['label']
X_train = train.drop('label', axis=1)
y_train.nunique()
sns.countplot(x=y_train)
plt.title('# Of samples')
plt.show() | code |
90127204/cell_1 | [
"text_plain_output_1.png"
] | import os
import numpy as np
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
import random
import tensorflow as tf
import os
for dirname, _, filenames in os.walk('/kaggle/input'):
for filename in filenames:
print(os.path.join(dirname, filename)) | code |
90127204/cell_7 | [
"image_output_1.png"
] | import pandas as pd # data processing, CSV file I/O
train = pd.read_csv('/kaggle/input/digit-recognizer/train.csv')
test = pd.read_csv('/kaggle/input/digit-recognizer/test.csv')
y_train = train['label']
X_train = train.drop('label', axis=1)
X_train.info() | code |
90127204/cell_18 | [
"image_output_1.png"
] | import tensorflow as tf
tf.random.set_seed(42)
model = tf.keras.Sequential([tf.keras.layers.Dense(units=784, activation='relu', input_shape=(784,)), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=392, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=151, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=50, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=10, activation='softmax')]) | code |
90127204/cell_15 | [
"text_plain_output_1.png"
] | import matplotlib.pyplot as plt
import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O
import random
import seaborn as sns
train = pd.read_csv('/kaggle/input/digit-recognizer/train.csv')
test = pd.read_csv('/kaggle/input/digit-recognizer/test.csv')
y_train = train['label']
X_train = train.drop('label', axis=1)
X_train = X_train / 255
y_train.nunique()
fig, ax = plt.subplots(5, 5, figsize=(8, 8))
fig.suptitle('Digits images and labels', fontsize=16)
ax = ax.ravel()
for i in range(25):
sample_n = random.randint(0, X_train.shape[0])
ax[i].imshow(np.array(X_train.iloc[sample_n]).reshape(28, 28), cmap='inferno')
ax[i].get_xaxis().set_visible(False)
ax[i].get_yaxis().set_visible(False)
ax[i].set_title(y_train[sample_n], fontsize = 12)
plt.subplots_adjust(hspace=0.3)
plt.colorbar()
fig, ax = plt.subplots(2, 5, figsize=(8, 4))
fig.suptitle('Average shape per digit', fontsize=16)
ax = ax.ravel()
for i in range(10):
ax[i].imshow(np.array(train[train['label'] == i].drop('label', axis=1).mean()).reshape(28, 28), cmap='inferno')
ax[i].get_xaxis().set_visible(False)
ax[i].get_yaxis().set_visible(False)
ax[i].set_title(i, fontsize=12) | code |
90127204/cell_24 | [
"application_vnd.jupyter.stderr_output_1.png"
] | import matplotlib.pyplot as plt
import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O
import random
import seaborn as sns
import tensorflow as tf
train = pd.read_csv('/kaggle/input/digit-recognizer/train.csv')
test = pd.read_csv('/kaggle/input/digit-recognizer/test.csv')
y_train = train['label']
X_train = train.drop('label', axis=1)
X_train = X_train / 255
y_train.nunique()
fig, ax = plt.subplots(5, 5, figsize=(8, 8))
fig.suptitle('Digits images and labels', fontsize=16)
ax = ax.ravel()
for i in range(25):
sample_n = random.randint(0, X_train.shape[0])
ax[i].imshow(np.array(X_train.iloc[sample_n]).reshape(28, 28), cmap='inferno')
ax[i].get_xaxis().set_visible(False)
ax[i].get_yaxis().set_visible(False)
ax[i].set_title(y_train[sample_n], fontsize = 12)
plt.subplots_adjust(hspace=0.3)
plt.colorbar()
fig, ax = plt.subplots(2, 5, figsize=(8, 4))
fig.suptitle('Average shape per digit', fontsize=16)
ax = ax.ravel()
for i in range(10):
ax[i].imshow(np.array(train[train['label'] == i].drop('label', axis=1).mean()).reshape(28, 28), cmap='inferno')
ax[i].get_xaxis().set_visible(False)
ax[i].get_yaxis().set_visible(False)
ax[i].set_title(i, fontsize = 12)
tf.random.set_seed(42)
model = tf.keras.Sequential([tf.keras.layers.Dense(units=784, activation='relu', input_shape=(784,)), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=392, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=151, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=50, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=10, activation='softmax')])
model.compile(optimizer='Adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])
es = tf.keras.callbacks.EarlyStopping(monitor='val_loss', min_delta=0, patience=5, verbose=0, mode='auto', baseline=None, restore_best_weights=False)
history = model.fit(X_train, y_train, epochs=100, validation_split=0.3, batch_size=50, callbacks=[es])
history.history.keys()
plt.title('accuracy')
plt.plot(range(len(history.history['loss'])), history.history['accuracy'], c='gray', marker='o')
plt.plot(range(len(history.history['loss'])), history.history['val_accuracy'], marker='o')
plt.show() | code |
90127204/cell_14 | [
"text_plain_output_1.png"
] | import matplotlib.pyplot as plt
import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O
import random
import seaborn as sns
train = pd.read_csv('/kaggle/input/digit-recognizer/train.csv')
test = pd.read_csv('/kaggle/input/digit-recognizer/test.csv')
y_train = train['label']
X_train = train.drop('label', axis=1)
X_train = X_train / 255
y_train.nunique()
fig, ax = plt.subplots(5, 5, figsize=(8, 8))
fig.suptitle('Digits images and labels', fontsize=16)
ax = ax.ravel()
for i in range(25):
sample_n = random.randint(0, X_train.shape[0])
ax[i].imshow(np.array(X_train.iloc[sample_n]).reshape(28, 28), cmap='inferno')
ax[i].get_xaxis().set_visible(False)
ax[i].get_yaxis().set_visible(False)
ax[i].set_title(y_train[sample_n], fontsize = 12)
plt.subplots_adjust(hspace=0.3)
plt.imshow(np.array(X_train.mean()).reshape(28, 28), cmap='inferno')
plt.colorbar()
plt.title('average shape', {'fontsize': 16})
plt.show() | code |
90127204/cell_22 | [
"image_output_1.png"
] | import matplotlib.pyplot as plt
import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O
import random
import seaborn as sns
import tensorflow as tf
train = pd.read_csv('/kaggle/input/digit-recognizer/train.csv')
test = pd.read_csv('/kaggle/input/digit-recognizer/test.csv')
y_train = train['label']
X_train = train.drop('label', axis=1)
X_train = X_train / 255
y_train.nunique()
fig, ax = plt.subplots(5, 5, figsize=(8, 8))
fig.suptitle('Digits images and labels', fontsize=16)
ax = ax.ravel()
for i in range(25):
sample_n = random.randint(0, X_train.shape[0])
ax[i].imshow(np.array(X_train.iloc[sample_n]).reshape(28, 28), cmap='inferno')
ax[i].get_xaxis().set_visible(False)
ax[i].get_yaxis().set_visible(False)
ax[i].set_title(y_train[sample_n], fontsize = 12)
plt.subplots_adjust(hspace=0.3)
plt.colorbar()
tf.random.set_seed(42)
model = tf.keras.Sequential([tf.keras.layers.Dense(units=784, activation='relu', input_shape=(784,)), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=392, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=151, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=50, activation='relu'), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(units=10, activation='softmax')])
model.compile(optimizer='Adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])
es = tf.keras.callbacks.EarlyStopping(monitor='val_loss', min_delta=0, patience=5, verbose=0, mode='auto', baseline=None, restore_best_weights=False)
history = model.fit(X_train, y_train, epochs=100, validation_split=0.3, batch_size=50, callbacks=[es])
history.history.keys() | code |
90127204/cell_12 | [
"text_html_output_1.png"
] | import matplotlib.pyplot as plt
import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O
import random
import seaborn as sns
train = pd.read_csv('/kaggle/input/digit-recognizer/train.csv')
test = pd.read_csv('/kaggle/input/digit-recognizer/test.csv')
y_train = train['label']
X_train = train.drop('label', axis=1)
X_train = X_train / 255
y_train.nunique()
fig, ax = plt.subplots(5, 5, figsize=(8, 8))
fig.suptitle('Digits images and labels', fontsize=16)
ax = ax.ravel()
for i in range(25):
sample_n = random.randint(0, X_train.shape[0])
ax[i].imshow(np.array(X_train.iloc[sample_n]).reshape(28, 28), cmap='inferno')
ax[i].get_xaxis().set_visible(False)
ax[i].get_yaxis().set_visible(False)
ax[i].set_title(y_train[sample_n], fontsize=12)
plt.subplots_adjust(hspace=0.3) | code |
90127204/cell_5 | [
"image_output_1.png"
] | import pandas as pd # data processing, CSV file I/O
train = pd.read_csv('/kaggle/input/digit-recognizer/train.csv')
test = pd.read_csv('/kaggle/input/digit-recognizer/test.csv')
y_train = train['label']
X_train = train.drop('label', axis=1)
y_train.head() | code |
74067422/cell_21 | [
"text_html_output_1.png"
] | import pandas as pd
TitanicPath = '../input/titanic-data/train.csv'
import pandas as pd
Titanic = pd.read_csv(TitanicPath)
Columns = pd.DataFrame({'Nombre': Titanic.columns, 'Tipo': [str(type(c)) for c in Titanic.columns]})
Columns[:]
Titanic[Titanic.PassengerId == 148]
FirstClass = Titanic[Titanic.Pclass == 1].Name
print(''.join([f'{str(p)}, ' for p in FirstClass])) | code |
74067422/cell_13 | [
"text_plain_output_1.png"
] | import pandas as pd
TitanicPath = '../input/titanic-data/train.csv'
import pandas as pd
Titanic = pd.read_csv(TitanicPath)
Columns = pd.DataFrame({'Nombre': Titanic.columns, 'Tipo': [str(type(c)) for c in Titanic.columns]})
Columns[:] | code |
74067422/cell_25 | [
"text_plain_output_1.png"
] | import pandas as pd
TitanicPath = '../input/titanic-data/train.csv'
import pandas as pd
Titanic = pd.read_csv(TitanicPath)
Columns = pd.DataFrame({'Nombre': Titanic.columns, 'Tipo': [str(type(c)) for c in Titanic.columns]})
Columns[:]
Titanic[Titanic.PassengerId == 148]
FirstClass = Titanic[Titanic.Pclass == 1].Name
Titanic['NoUnderAge'] = Titanic.Age >= 18
Titanic[:]
print(Titanic.Fare.max())
Titanic[Titanic.Fare == Titanic.Fare.max()] | code |
74067422/cell_23 | [
"text_html_output_1.png"
] | import pandas as pd
TitanicPath = '../input/titanic-data/train.csv'
import pandas as pd
Titanic = pd.read_csv(TitanicPath)
Columns = pd.DataFrame({'Nombre': Titanic.columns, 'Tipo': [str(type(c)) for c in Titanic.columns]})
Columns[:]
Titanic[Titanic.PassengerId == 148]
FirstClass = Titanic[Titanic.Pclass == 1].Name
Titanic['NoUnderAge'] = Titanic.Age >= 18
Titanic[:] | code |
74067422/cell_30 | [
"text_html_output_1.png",
"text_plain_output_1.png"
] | import pandas as pd
TitanicPath = '../input/titanic-data/train.csv'
import pandas as pd
Titanic = pd.read_csv(TitanicPath)
Columns = pd.DataFrame({'Nombre': Titanic.columns, 'Tipo': [str(type(c)) for c in Titanic.columns]})
Columns[:]
Titanic[Titanic.PassengerId == 148]
FirstClass = Titanic[Titanic.Pclass == 1].Name
Titanic['NoUnderAge'] = Titanic.Age >= 18
Titanic[:]
Titanic[Titanic.Fare == Titanic.Fare.max()]
Titanic[Titanic.Fare == Titanic.Fare.min()]
f = open('titanic.json', 'w')
f.write(Titanic.to_json())
f.close()
TitanicSurvived = Titanic[Titanic.Survived == 1]
TitanicDied = Titanic[Titanic.Survived == 0]
TitanicSurvivedMalesCount = TitanicSurvived[TitanicSurvived.Sex == 'male'].Sex.count()
TitanicSurvivedFemalesCount = TitanicSurvived[TitanicSurvived.Sex == 'female'].Sex.count()
TitanicSurvivedNoUnderAgeCount = TitanicSurvived[TitanicSurvived.NoUnderAge == True].NoUnderAge.count()
TitanicSurvivedUnderAgeCount = TitanicSurvived[TitanicSurvived.NoUnderAge == False].NoUnderAge.count()
print(f'1 - La clase de pasajero media fue mas alta en los sobrevivientes {TitanicSurvived.Pclass.median()} que del resto {TitanicDied.Pclass.median()}')
print(f'2 - El precio del tiquete medio fue mas alto en los sobrevivientes {TitanicSurvived.Fare.median()} que del resto {TitanicDied.Fare.median()}')
print(f'3 - Sobrevivieron mas mujeres {TitanicSurvivedFemalesCount} que hombres {TitanicSurvivedMalesCount}')
print(f'4 - Sobrevivieron mas mayores de edad {TitanicSurvivedNoUnderAgeCount} que menores de edad {TitanicSurvivedUnderAgeCount}')
TitanicSurvived[:] | code |
74067422/cell_26 | [
"text_html_output_1.png"
] | import pandas as pd
TitanicPath = '../input/titanic-data/train.csv'
import pandas as pd
Titanic = pd.read_csv(TitanicPath)
Columns = pd.DataFrame({'Nombre': Titanic.columns, 'Tipo': [str(type(c)) for c in Titanic.columns]})
Columns[:]
Titanic[Titanic.PassengerId == 148]
FirstClass = Titanic[Titanic.Pclass == 1].Name
Titanic['NoUnderAge'] = Titanic.Age >= 18
Titanic[:]
Titanic[Titanic.Fare == Titanic.Fare.max()]
print(Titanic.Fare.min())
Titanic[Titanic.Fare == Titanic.Fare.min()] | code |
74067422/cell_11 | [
"text_plain_output_1.png"
] | import pandas as pd
TitanicPath = '../input/titanic-data/train.csv'
import pandas as pd
Titanic = pd.read_csv(TitanicPath)
print(f'Dimensiones: {Titanic.shape[0]}x{Titanic.shape[1]}')
print(f'Numero de Datos: {Titanic.shape[0]}')
print(f'Numero de Columnas: {Titanic.shape[1]}') | code |
74067422/cell_19 | [
"text_html_output_1.png"
] | import pandas as pd
TitanicPath = '../input/titanic-data/train.csv'
import pandas as pd
Titanic = pd.read_csv(TitanicPath)
Columns = pd.DataFrame({'Nombre': Titanic.columns, 'Tipo': [str(type(c)) for c in Titanic.columns]})
Columns[:]
Titanic[Titanic.PassengerId == 148] | code |
74067422/cell_7 | [
"text_plain_output_1.png",
"image_output_1.png"
] | def fibonacci(terms: int=1):
out: list[int] = []
for i in range(terms):
if i == 0:
out.append(0)
elif i == 1:
out.append(1)
else:
out.append(out[-2] + out[-1])
return out
def countWords(string: str='', sep: str=', '):
out: dict[str, int] = {}
for t in string.split(sep):
out[t] = string.count(t)
return out
def frequentWord(wordCount: dict={}):
out: tuple[str, int] = ('None', 0)
for key, val in wordCount.items():
if val > out[1]:
out = (key, val)
return out
Words = 'Perro, Gato, Conejo, Perro Lobo, Gato Leon, Gato Blanco'
WordsCount = countWords(Words)
print(WordsCount)
FrequentWord = frequentWord(WordsCount)
print(f'{FrequentWord[0]} se repite {FrequentWord[1]}', 'vez' if FrequentWord[1] == 1 else 'veces') | code |
74067422/cell_15 | [
"text_plain_output_1.png"
] | import pandas as pd
TitanicPath = '../input/titanic-data/train.csv'
import pandas as pd
Titanic = pd.read_csv(TitanicPath)
Columns = pd.DataFrame({'Nombre': Titanic.columns, 'Tipo': [str(type(c)) for c in Titanic.columns]})
Columns[:]
Titanic[:10] | code |
74067422/cell_17 | [
"text_html_output_1.png"
] | import pandas as pd
TitanicPath = '../input/titanic-data/train.csv'
import pandas as pd
Titanic = pd.read_csv(TitanicPath)
Columns = pd.DataFrame({'Nombre': Titanic.columns, 'Tipo': [str(type(c)) for c in Titanic.columns]})
Columns[:]
Titanic[-10:] | code |
74067422/cell_5 | [
"text_html_output_1.png",
"text_plain_output_1.png"
] | def fibonacci(terms: int=1):
out: list[int] = []
for i in range(terms):
if i == 0:
out.append(0)
elif i == 1:
out.append(1)
else:
out.append(out[-2] + out[-1])
return out
for i in range(15):
print(fibonacci(i)) | code |
325674/cell_4 | [
"text_plain_output_1.png",
"image_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
global_temperatures = pd.read_csv('../input/GlobalTemperatures.csv', infer_datetime_format=True, index_col='dt', parse_dates=['dt'])
global_temperatures[global_temperatures.index.year > 2000]['LandAverageTemperature'].plot(figsize=(13, 7)) | code |
325674/cell_6 | [
"text_plain_output_1.png",
"image_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
global_temperatures = pd.read_csv('../input/GlobalTemperatures.csv', infer_datetime_format=True, index_col='dt', parse_dates=['dt'])
global_temperatures.groupby(global_temperatures.index.year)['LandAverageTemperature'].mean().plot(figsize=(13, 7)) | code |
325674/cell_2 | [
"text_plain_output_1.png"
] | from subprocess import check_output
import numpy as np
import pandas as pd
from subprocess import check_output
print(check_output(['ls', '../input']).decode('utf8'))
from matplotlib import pyplot as plt | code |
325674/cell_3 | [
"text_plain_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
global_temperatures = pd.read_csv('../input/GlobalTemperatures.csv', infer_datetime_format=True, index_col='dt', parse_dates=['dt'])
print(global_temperatures.info()) | code |
73069103/cell_21 | [
"text_plain_output_1.png"
] | from torch.utils.data import Dataset, DataLoader, ConcatDataset
from torchvision.transforms import transforms
import cv2
import cv2
import matplotlib.pyplot as plt
import numpy as np
import numpy as np
import numpy as np # linear algebra
import os
import os
import pandas as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import torch
import torchvision
import numpy as np
import pandas as pd
import os
data_lebel = pd.read_csv('/kaggle/input/soft-computing-even-id-dataset/training-a.csv')
img = cv2.imread('/kaggle/input/soft-computing-even-id-dataset/training-a/a00000.png')
img.shape
IMAGE_SIZE = 40
transform = transforms.Compose([transforms.ToPILImage(), torchvision.transforms.ColorJitter(brightness=0.4, saturation=0.4, contrast=0.4, hue=0.4), transforms.RandomRotation(20, expand=True), transforms.Resize(IMAGE_SIZE), transforms.CenterCrop(IMAGE_SIZE), transforms.ToTensor(), transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])
class digit_Dataset(Dataset):
def __init__(self, csv_file, root_dir, transform=True):
self.annotations = pd.read_csv(csv_file)
self.root_dir = root_dir
self.transform = transform
def __len__(self):
return len(self.annotations)
def __getitem__(self, index):
img_path = os.path.join(self.root_dir, self.annotations.iloc[index, 0])
image = cv2.imread(img_path)
y_label = torch.tensor(int(self.annotations.iloc[index, 3]))
if self.transform:
image = self.transform(image)
return (image, y_label)
dataset = digit_Dataset(csv_file='/kaggle/input/soft-computing-even-id-dataset/training-a.csv', root_dir='/kaggle/input/soft-computing-even-id-dataset/training-a', transform=transform)
train_set, test_set = torch.utils.data.random_split(dataset, [15702, 4000])
train_loader = DataLoader(dataset=train_set, batch_size=16, shuffle=True)
test_loader = DataLoader(dataset=test_set, batch_size=16, shuffle=True)
data_iter = iter(train_loader)
images, labels = data_iter.next()
images.shape
dataiter = iter(train_loader)
images, labels = dataiter.next()
plt.subplots(figsize=(20, 32))
for i in range(10):
plt.subplot(10 / 2, 10, i + 1)
img = images[i].detach().numpy().transpose((1, 2, 0))
mean = np.array([0.5, 0.5, 0.5])
std = np.array([0.5, 0.5, 0.5])
img = std * img + mean
img = np.clip(img, 0, 1)
plt.title(labels[i])
plt.imshow(img)
plt.show() | code |
73069103/cell_4 | [
"text_plain_output_1.png"
] | import pandas as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
data_lebel = pd.read_csv('/kaggle/input/soft-computing-even-id-dataset/training-a.csv')
data_lebel.head() | code |
73069103/cell_23 | [
"text_plain_output_1.png"
] | from torch.utils.data import Dataset, DataLoader, ConcatDataset
from torchvision.transforms import transforms
import cv2
import cv2
import os
import os
import pandas as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import torch
import torch.nn as nn
import torch.nn.functional as f
import torchvision
import numpy as np
import pandas as pd
import os
data_lebel = pd.read_csv('/kaggle/input/soft-computing-even-id-dataset/training-a.csv')
img = cv2.imread('/kaggle/input/soft-computing-even-id-dataset/training-a/a00000.png')
IMAGE_SIZE = 40
transform = transforms.Compose([transforms.ToPILImage(), torchvision.transforms.ColorJitter(brightness=0.4, saturation=0.4, contrast=0.4, hue=0.4), transforms.RandomRotation(20, expand=True), transforms.Resize(IMAGE_SIZE), transforms.CenterCrop(IMAGE_SIZE), transforms.ToTensor(), transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])
class digit_Dataset(Dataset):
def __init__(self, csv_file, root_dir, transform=True):
self.annotations = pd.read_csv(csv_file)
self.root_dir = root_dir
self.transform = transform
def __len__(self):
return len(self.annotations)
def __getitem__(self, index):
img_path = os.path.join(self.root_dir, self.annotations.iloc[index, 0])
image = cv2.imread(img_path)
y_label = torch.tensor(int(self.annotations.iloc[index, 3]))
if self.transform:
image = self.transform(image)
return (image, y_label)
class Net(nn.Module):
def __init__(self):
super().__init__()
self.fc1 = nn.Linear(3 * 40 * 40, 200)
self.fc2 = nn.Linear(200, 200)
self.fc3 = nn.Linear(200, 200)
self.fc4 = nn.Linear(200, 200)
self.fc5 = nn.Linear(200, 200)
self.fc6 = nn.Linear(200, 10)
def forward(self, x):
x = f.relu(self.fc1(x))
x = f.relu(self.fc2(x))
x = f.relu(self.fc3(x))
x = f.relu(self.fc4(x))
x = f.relu(self.fc5(x))
x = self.fc6(x)
return x
net = Net()
net.cuda() | code |
73069103/cell_20 | [
"text_plain_output_1.png"
] | from torch.utils.data import Dataset, DataLoader, ConcatDataset
from torchvision.transforms import transforms
import cv2
import cv2
import os
import os
import pandas as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import torch
import torchvision
import numpy as np
import pandas as pd
import os
data_lebel = pd.read_csv('/kaggle/input/soft-computing-even-id-dataset/training-a.csv')
img = cv2.imread('/kaggle/input/soft-computing-even-id-dataset/training-a/a00000.png')
IMAGE_SIZE = 40
transform = transforms.Compose([transforms.ToPILImage(), torchvision.transforms.ColorJitter(brightness=0.4, saturation=0.4, contrast=0.4, hue=0.4), transforms.RandomRotation(20, expand=True), transforms.Resize(IMAGE_SIZE), transforms.CenterCrop(IMAGE_SIZE), transforms.ToTensor(), transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])
class digit_Dataset(Dataset):
def __init__(self, csv_file, root_dir, transform=True):
self.annotations = pd.read_csv(csv_file)
self.root_dir = root_dir
self.transform = transform
def __len__(self):
return len(self.annotations)
def __getitem__(self, index):
img_path = os.path.join(self.root_dir, self.annotations.iloc[index, 0])
image = cv2.imread(img_path)
y_label = torch.tensor(int(self.annotations.iloc[index, 3]))
if self.transform:
image = self.transform(image)
return (image, y_label)
dataset = digit_Dataset(csv_file='/kaggle/input/soft-computing-even-id-dataset/training-a.csv', root_dir='/kaggle/input/soft-computing-even-id-dataset/training-a', transform=transform)
train_set, test_set = torch.utils.data.random_split(dataset, [15702, 4000])
train_loader = DataLoader(dataset=train_set, batch_size=16, shuffle=True)
test_loader = DataLoader(dataset=test_set, batch_size=16, shuffle=True)
data_iter = iter(train_loader)
images, labels = data_iter.next()
images.shape | code |
73069103/cell_6 | [
"text_plain_output_1.png"
] | import pandas as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
data_lebel = pd.read_csv('/kaggle/input/soft-computing-even-id-dataset/training-a.csv')
data_lebel['digit'].value_counts() | code |
73069103/cell_29 | [
"text_plain_output_1.png"
] | from torch.utils.data import Dataset, DataLoader, ConcatDataset
from torchvision.transforms import transforms
import cv2
import cv2
import os
import os
import pandas as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import torch
import torch.nn as nn
import torch.nn.functional as f
import torch.optim as optim
import torchvision
import numpy as np
import pandas as pd
import os
data_lebel = pd.read_csv('/kaggle/input/soft-computing-even-id-dataset/training-a.csv')
img = cv2.imread('/kaggle/input/soft-computing-even-id-dataset/training-a/a00000.png')
IMAGE_SIZE = 40
transform = transforms.Compose([transforms.ToPILImage(), torchvision.transforms.ColorJitter(brightness=0.4, saturation=0.4, contrast=0.4, hue=0.4), transforms.RandomRotation(20, expand=True), transforms.Resize(IMAGE_SIZE), transforms.CenterCrop(IMAGE_SIZE), transforms.ToTensor(), transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])
class digit_Dataset(Dataset):
def __init__(self, csv_file, root_dir, transform=True):
self.annotations = pd.read_csv(csv_file)
self.root_dir = root_dir
self.transform = transform
def __len__(self):
return len(self.annotations)
def __getitem__(self, index):
img_path = os.path.join(self.root_dir, self.annotations.iloc[index, 0])
image = cv2.imread(img_path)
y_label = torch.tensor(int(self.annotations.iloc[index, 3]))
if self.transform:
image = self.transform(image)
return (image, y_label)
dataset = digit_Dataset(csv_file='/kaggle/input/soft-computing-even-id-dataset/training-a.csv', root_dir='/kaggle/input/soft-computing-even-id-dataset/training-a', transform=transform)
train_set, test_set = torch.utils.data.random_split(dataset, [15702, 4000])
train_loader = DataLoader(dataset=train_set, batch_size=16, shuffle=True)
test_loader = DataLoader(dataset=test_set, batch_size=16, shuffle=True)
class Net(nn.Module):
def __init__(self):
super().__init__()
self.fc1 = nn.Linear(3 * 40 * 40, 200)
self.fc2 = nn.Linear(200, 200)
self.fc3 = nn.Linear(200, 200)
self.fc4 = nn.Linear(200, 200)
self.fc5 = nn.Linear(200, 200)
self.fc6 = nn.Linear(200, 10)
def forward(self, x):
x = f.relu(self.fc1(x))
x = f.relu(self.fc2(x))
x = f.relu(self.fc3(x))
x = f.relu(self.fc4(x))
x = f.relu(self.fc5(x))
x = self.fc6(x)
return x
net = Net()
net.cuda()
optimizer = optim.SGD(net.parameters(), lr=0.01, momentum=0.9)
criterion = torch.nn.CrossEntropyLoss()
device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
Path = './net_final.pth'
traininglosses = []
trainingaccuracy = []
testinglosses = []
testaccuracy = []
totalsteps = []
epochs = 20
steps = 0
running_loss = 0
print_every = 5000
for epoch in range(epochs):
accuracy = 0
for inputs, labels in train_loader:
net.train()
steps += 1
inputs, labels = (inputs.to(device), labels.to(device))
optimizer.zero_grad()
logps = net.forward(inputs.view(-1, 3 * 40 * 40))
loss = criterion(logps, labels)
loss.backward()
optimizer.step()
pred = torch.argmax(logps, dim=1)
correct = pred.eq(labels)
running_loss += loss.item()
accuracy += torch.mean(correct.float())
if steps % print_every == 0:
after_train_accuracy = accuracy / print_every
test_loss = 0
accuracy = 0
net.eval()
with torch.no_grad():
for inputs, labels in test_loader:
inputs, labels = (inputs.to(device), labels.to(device))
logps = net.forward(inputs.view(-1, 3 * 40 * 40))
batch_loss = criterion(logps, labels)
test_loss += batch_loss.item()
pred = torch.argmax(logps, dim=1)
correct = pred.eq(labels)
accuracy += torch.mean(correct.float())
traininglosses.append(running_loss / print_every)
trainingaccuracy.append(after_train_accuracy)
testinglosses.append(test_loss / len(test_loader))
testaccuracy.append(accuracy / len(test_loader))
totalsteps.append(steps)
running_loss = 0
accuracy = 0
net.train()
torch.save(net.state_dict(), Path)
correct = 0
total = 0
with torch.no_grad():
for data in test_loader:
X, y = data
X, y = (X.to(device), y.to(device))
output = net(X.view(-1, 3 * 40 * 40))
for idx, i in enumerate(output):
if torch.argmax(i) == y[idx]:
correct += 1
total += 1
print(round(correct / total, 3)) | code |
73069103/cell_26 | [
"application_vnd.jupyter.stderr_output_1.png",
"image_output_1.png"
] | from torch.utils.data import Dataset, DataLoader, ConcatDataset
from torchvision.transforms import transforms
import cv2
import cv2
import os
import os
import pandas as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import torch
import torch.nn as nn
import torch.nn.functional as f
import torch.optim as optim
import torchvision
import numpy as np
import pandas as pd
import os
data_lebel = pd.read_csv('/kaggle/input/soft-computing-even-id-dataset/training-a.csv')
img = cv2.imread('/kaggle/input/soft-computing-even-id-dataset/training-a/a00000.png')
IMAGE_SIZE = 40
transform = transforms.Compose([transforms.ToPILImage(), torchvision.transforms.ColorJitter(brightness=0.4, saturation=0.4, contrast=0.4, hue=0.4), transforms.RandomRotation(20, expand=True), transforms.Resize(IMAGE_SIZE), transforms.CenterCrop(IMAGE_SIZE), transforms.ToTensor(), transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])
class digit_Dataset(Dataset):
def __init__(self, csv_file, root_dir, transform=True):
self.annotations = pd.read_csv(csv_file)
self.root_dir = root_dir
self.transform = transform
def __len__(self):
return len(self.annotations)
def __getitem__(self, index):
img_path = os.path.join(self.root_dir, self.annotations.iloc[index, 0])
image = cv2.imread(img_path)
y_label = torch.tensor(int(self.annotations.iloc[index, 3]))
if self.transform:
image = self.transform(image)
return (image, y_label)
dataset = digit_Dataset(csv_file='/kaggle/input/soft-computing-even-id-dataset/training-a.csv', root_dir='/kaggle/input/soft-computing-even-id-dataset/training-a', transform=transform)
train_set, test_set = torch.utils.data.random_split(dataset, [15702, 4000])
train_loader = DataLoader(dataset=train_set, batch_size=16, shuffle=True)
test_loader = DataLoader(dataset=test_set, batch_size=16, shuffle=True)
class Net(nn.Module):
def __init__(self):
super().__init__()
self.fc1 = nn.Linear(3 * 40 * 40, 200)
self.fc2 = nn.Linear(200, 200)
self.fc3 = nn.Linear(200, 200)
self.fc4 = nn.Linear(200, 200)
self.fc5 = nn.Linear(200, 200)
self.fc6 = nn.Linear(200, 10)
def forward(self, x):
x = f.relu(self.fc1(x))
x = f.relu(self.fc2(x))
x = f.relu(self.fc3(x))
x = f.relu(self.fc4(x))
x = f.relu(self.fc5(x))
x = self.fc6(x)
return x
net = Net()
net.cuda()
optimizer = optim.SGD(net.parameters(), lr=0.01, momentum=0.9)
criterion = torch.nn.CrossEntropyLoss()
device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
device | code |
73069103/cell_19 | [
"text_plain_output_1.png"
] | from torch.utils.data import Dataset, DataLoader, ConcatDataset
from torchvision.transforms import transforms
import cv2
import cv2
import os
import os
import pandas as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import torch
import torchvision
import numpy as np
import pandas as pd
import os
data_lebel = pd.read_csv('/kaggle/input/soft-computing-even-id-dataset/training-a.csv')
img = cv2.imread('/kaggle/input/soft-computing-even-id-dataset/training-a/a00000.png')
IMAGE_SIZE = 40
transform = transforms.Compose([transforms.ToPILImage(), torchvision.transforms.ColorJitter(brightness=0.4, saturation=0.4, contrast=0.4, hue=0.4), transforms.RandomRotation(20, expand=True), transforms.Resize(IMAGE_SIZE), transforms.CenterCrop(IMAGE_SIZE), transforms.ToTensor(), transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])
class digit_Dataset(Dataset):
def __init__(self, csv_file, root_dir, transform=True):
self.annotations = pd.read_csv(csv_file)
self.root_dir = root_dir
self.transform = transform
def __len__(self):
return len(self.annotations)
def __getitem__(self, index):
img_path = os.path.join(self.root_dir, self.annotations.iloc[index, 0])
image = cv2.imread(img_path)
y_label = torch.tensor(int(self.annotations.iloc[index, 3]))
if self.transform:
image = self.transform(image)
return (image, y_label)
dataset = digit_Dataset(csv_file='/kaggle/input/soft-computing-even-id-dataset/training-a.csv', root_dir='/kaggle/input/soft-computing-even-id-dataset/training-a', transform=transform)
train_set, test_set = torch.utils.data.random_split(dataset, [15702, 4000])
train_loader = DataLoader(dataset=train_set, batch_size=16, shuffle=True)
test_loader = DataLoader(dataset=test_set, batch_size=16, shuffle=True)
type(train_loader) | code |
73069103/cell_18 | [
"text_plain_output_1.png"
] | from torch.utils.data import Dataset, DataLoader, ConcatDataset
from torchvision.transforms import transforms
import cv2
import cv2
import os
import os
import pandas as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import torch
import torchvision
import numpy as np
import pandas as pd
import os
data_lebel = pd.read_csv('/kaggle/input/soft-computing-even-id-dataset/training-a.csv')
img = cv2.imread('/kaggle/input/soft-computing-even-id-dataset/training-a/a00000.png')
IMAGE_SIZE = 40
transform = transforms.Compose([transforms.ToPILImage(), torchvision.transforms.ColorJitter(brightness=0.4, saturation=0.4, contrast=0.4, hue=0.4), transforms.RandomRotation(20, expand=True), transforms.Resize(IMAGE_SIZE), transforms.CenterCrop(IMAGE_SIZE), transforms.ToTensor(), transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])
class digit_Dataset(Dataset):
def __init__(self, csv_file, root_dir, transform=True):
self.annotations = pd.read_csv(csv_file)
self.root_dir = root_dir
self.transform = transform
def __len__(self):
return len(self.annotations)
def __getitem__(self, index):
img_path = os.path.join(self.root_dir, self.annotations.iloc[index, 0])
image = cv2.imread(img_path)
y_label = torch.tensor(int(self.annotations.iloc[index, 3]))
if self.transform:
image = self.transform(image)
return (image, y_label)
dataset = digit_Dataset(csv_file='/kaggle/input/soft-computing-even-id-dataset/training-a.csv', root_dir='/kaggle/input/soft-computing-even-id-dataset/training-a', transform=transform)
train_set, test_set = torch.utils.data.random_split(dataset, [15702, 4000])
train_loader = DataLoader(dataset=train_set, batch_size=16, shuffle=True)
test_loader = DataLoader(dataset=test_set, batch_size=16, shuffle=True)
len(train_loader) | code |
73069103/cell_28 | [
"text_plain_output_1.png"
] | from torch.utils.data import Dataset, DataLoader, ConcatDataset
from torchvision.transforms import transforms
import cv2
import cv2
import matplotlib.pyplot as plt
import numpy as np
import numpy as np
import numpy as np # linear algebra
import os
import os
import pandas as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import torch
import torch.nn as nn
import torch.nn.functional as f
import torch.optim as optim
import torchvision
import numpy as np
import pandas as pd
import os
data_lebel = pd.read_csv('/kaggle/input/soft-computing-even-id-dataset/training-a.csv')
img = cv2.imread('/kaggle/input/soft-computing-even-id-dataset/training-a/a00000.png')
img.shape
IMAGE_SIZE = 40
transform = transforms.Compose([transforms.ToPILImage(), torchvision.transforms.ColorJitter(brightness=0.4, saturation=0.4, contrast=0.4, hue=0.4), transforms.RandomRotation(20, expand=True), transforms.Resize(IMAGE_SIZE), transforms.CenterCrop(IMAGE_SIZE), transforms.ToTensor(), transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])
class digit_Dataset(Dataset):
def __init__(self, csv_file, root_dir, transform=True):
self.annotations = pd.read_csv(csv_file)
self.root_dir = root_dir
self.transform = transform
def __len__(self):
return len(self.annotations)
def __getitem__(self, index):
img_path = os.path.join(self.root_dir, self.annotations.iloc[index, 0])
image = cv2.imread(img_path)
y_label = torch.tensor(int(self.annotations.iloc[index, 3]))
if self.transform:
image = self.transform(image)
return (image, y_label)
dataset = digit_Dataset(csv_file='/kaggle/input/soft-computing-even-id-dataset/training-a.csv', root_dir='/kaggle/input/soft-computing-even-id-dataset/training-a', transform=transform)
train_set, test_set = torch.utils.data.random_split(dataset, [15702, 4000])
train_loader = DataLoader(dataset=train_set, batch_size=16, shuffle=True)
test_loader = DataLoader(dataset=test_set, batch_size=16, shuffle=True)
data_iter = iter(train_loader)
images, labels = data_iter.next()
images.shape
dataiter = iter(train_loader)
images, labels = dataiter.next()
for i in range(10):
img = images[i].detach().numpy().transpose((1, 2, 0))
mean = np.array([0.5, 0.5, 0.5])
std = np.array([0.5, 0.5, 0.5])
img = std * img + mean
img = np.clip(img, 0, 1)
class Net(nn.Module):
def __init__(self):
super().__init__()
self.fc1 = nn.Linear(3 * 40 * 40, 200)
self.fc2 = nn.Linear(200, 200)
self.fc3 = nn.Linear(200, 200)
self.fc4 = nn.Linear(200, 200)
self.fc5 = nn.Linear(200, 200)
self.fc6 = nn.Linear(200, 10)
def forward(self, x):
x = f.relu(self.fc1(x))
x = f.relu(self.fc2(x))
x = f.relu(self.fc3(x))
x = f.relu(self.fc4(x))
x = f.relu(self.fc5(x))
x = self.fc6(x)
return x
net = Net()
net.cuda()
optimizer = optim.SGD(net.parameters(), lr=0.01, momentum=0.9)
criterion = torch.nn.CrossEntropyLoss()
device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
Path = './net_final.pth'
traininglosses = []
trainingaccuracy = []
testinglosses = []
testaccuracy = []
totalsteps = []
epochs = 20
steps = 0
running_loss = 0
print_every = 5000
for epoch in range(epochs):
accuracy = 0
for inputs, labels in train_loader:
net.train()
steps += 1
inputs, labels = (inputs.to(device), labels.to(device))
optimizer.zero_grad()
logps = net.forward(inputs.view(-1, 3 * 40 * 40))
loss = criterion(logps, labels)
loss.backward()
optimizer.step()
pred = torch.argmax(logps, dim=1)
correct = pred.eq(labels)
running_loss += loss.item()
accuracy += torch.mean(correct.float())
if steps % print_every == 0:
after_train_accuracy = accuracy / print_every
test_loss = 0
accuracy = 0
net.eval()
with torch.no_grad():
for inputs, labels in test_loader:
inputs, labels = (inputs.to(device), labels.to(device))
logps = net.forward(inputs.view(-1, 3 * 40 * 40))
batch_loss = criterion(logps, labels)
test_loss += batch_loss.item()
pred = torch.argmax(logps, dim=1)
correct = pred.eq(labels)
accuracy += torch.mean(correct.float())
traininglosses.append(running_loss / print_every)
trainingaccuracy.append(after_train_accuracy)
testinglosses.append(test_loss / len(test_loader))
testaccuracy.append(accuracy / len(test_loader))
totalsteps.append(steps)
running_loss = 0
accuracy = 0
net.train()
torch.save(net.state_dict(), Path)
plt.figure(figsize=(50, 10))
plt.plot(totalsteps, traininglosses, label='Train Loss')
plt.plot(totalsteps, trainingaccuracy, label='Test Accuracy')
plt.plot(totalsteps, testinglosses, label='Test Loss')
plt.plot(totalsteps, testaccuracy, label='Test Accuracy')
plt.legend()
plt.grid()
plt.show() | code |
73069103/cell_16 | [
"text_plain_output_1.png"
] | from torchvision.transforms import transforms
import torchvision
IMAGE_SIZE = 40
transform = transforms.Compose([transforms.ToPILImage(), torchvision.transforms.ColorJitter(brightness=0.4, saturation=0.4, contrast=0.4, hue=0.4), transforms.RandomRotation(20, expand=True), transforms.Resize(IMAGE_SIZE), transforms.CenterCrop(IMAGE_SIZE), transforms.ToTensor(), transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])
dataset = digit_Dataset(csv_file='/kaggle/input/soft-computing-even-id-dataset/training-a.csv', root_dir='/kaggle/input/soft-computing-even-id-dataset/training-a', transform=transform)
len(dataset) | code |
73069103/cell_22 | [
"text_plain_output_1.png"
] | from torch.utils.data import Dataset, DataLoader, ConcatDataset
from torchvision.transforms import transforms
import cv2
import cv2
import os
import os
import pandas as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import torch
import torch.nn as nn
import torch.nn.functional as f
import torchvision
import numpy as np
import pandas as pd
import os
data_lebel = pd.read_csv('/kaggle/input/soft-computing-even-id-dataset/training-a.csv')
img = cv2.imread('/kaggle/input/soft-computing-even-id-dataset/training-a/a00000.png')
IMAGE_SIZE = 40
transform = transforms.Compose([transforms.ToPILImage(), torchvision.transforms.ColorJitter(brightness=0.4, saturation=0.4, contrast=0.4, hue=0.4), transforms.RandomRotation(20, expand=True), transforms.Resize(IMAGE_SIZE), transforms.CenterCrop(IMAGE_SIZE), transforms.ToTensor(), transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])
class digit_Dataset(Dataset):
def __init__(self, csv_file, root_dir, transform=True):
self.annotations = pd.read_csv(csv_file)
self.root_dir = root_dir
self.transform = transform
def __len__(self):
return len(self.annotations)
def __getitem__(self, index):
img_path = os.path.join(self.root_dir, self.annotations.iloc[index, 0])
image = cv2.imread(img_path)
y_label = torch.tensor(int(self.annotations.iloc[index, 3]))
if self.transform:
image = self.transform(image)
return (image, y_label)
class Net(nn.Module):
def __init__(self):
super().__init__()
self.fc1 = nn.Linear(3 * 40 * 40, 200)
self.fc2 = nn.Linear(200, 200)
self.fc3 = nn.Linear(200, 200)
self.fc4 = nn.Linear(200, 200)
self.fc5 = nn.Linear(200, 200)
self.fc6 = nn.Linear(200, 10)
def forward(self, x):
x = f.relu(self.fc1(x))
x = f.relu(self.fc2(x))
x = f.relu(self.fc3(x))
x = f.relu(self.fc4(x))
x = f.relu(self.fc5(x))
x = self.fc6(x)
return x
net = Net()
print(net) | code |
73069103/cell_10 | [
"text_html_output_1.png"
] | import cv2
import cv2
img = cv2.imread('/kaggle/input/soft-computing-even-id-dataset/training-a/a00000.png')
img.shape | code |
73069103/cell_27 | [
"text_plain_output_1.png"
] | from torch.utils.data import Dataset, DataLoader, ConcatDataset
from torchvision.transforms import transforms
import cv2
import cv2
import os
import os
import pandas as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import torch
import torch.nn as nn
import torch.nn.functional as f
import torch.optim as optim
import torchvision
import numpy as np
import pandas as pd
import os
data_lebel = pd.read_csv('/kaggle/input/soft-computing-even-id-dataset/training-a.csv')
img = cv2.imread('/kaggle/input/soft-computing-even-id-dataset/training-a/a00000.png')
IMAGE_SIZE = 40
transform = transforms.Compose([transforms.ToPILImage(), torchvision.transforms.ColorJitter(brightness=0.4, saturation=0.4, contrast=0.4, hue=0.4), transforms.RandomRotation(20, expand=True), transforms.Resize(IMAGE_SIZE), transforms.CenterCrop(IMAGE_SIZE), transforms.ToTensor(), transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])
class digit_Dataset(Dataset):
def __init__(self, csv_file, root_dir, transform=True):
self.annotations = pd.read_csv(csv_file)
self.root_dir = root_dir
self.transform = transform
def __len__(self):
return len(self.annotations)
def __getitem__(self, index):
img_path = os.path.join(self.root_dir, self.annotations.iloc[index, 0])
image = cv2.imread(img_path)
y_label = torch.tensor(int(self.annotations.iloc[index, 3]))
if self.transform:
image = self.transform(image)
return (image, y_label)
dataset = digit_Dataset(csv_file='/kaggle/input/soft-computing-even-id-dataset/training-a.csv', root_dir='/kaggle/input/soft-computing-even-id-dataset/training-a', transform=transform)
train_set, test_set = torch.utils.data.random_split(dataset, [15702, 4000])
train_loader = DataLoader(dataset=train_set, batch_size=16, shuffle=True)
test_loader = DataLoader(dataset=test_set, batch_size=16, shuffle=True)
class Net(nn.Module):
def __init__(self):
super().__init__()
self.fc1 = nn.Linear(3 * 40 * 40, 200)
self.fc2 = nn.Linear(200, 200)
self.fc3 = nn.Linear(200, 200)
self.fc4 = nn.Linear(200, 200)
self.fc5 = nn.Linear(200, 200)
self.fc6 = nn.Linear(200, 10)
def forward(self, x):
x = f.relu(self.fc1(x))
x = f.relu(self.fc2(x))
x = f.relu(self.fc3(x))
x = f.relu(self.fc4(x))
x = f.relu(self.fc5(x))
x = self.fc6(x)
return x
net = Net()
net.cuda()
optimizer = optim.SGD(net.parameters(), lr=0.01, momentum=0.9)
criterion = torch.nn.CrossEntropyLoss()
device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
Path = './net_final.pth'
traininglosses = []
trainingaccuracy = []
testinglosses = []
testaccuracy = []
totalsteps = []
epochs = 20
steps = 0
running_loss = 0
print_every = 5000
for epoch in range(epochs):
accuracy = 0
for inputs, labels in train_loader:
net.train()
steps += 1
inputs, labels = (inputs.to(device), labels.to(device))
optimizer.zero_grad()
logps = net.forward(inputs.view(-1, 3 * 40 * 40))
loss = criterion(logps, labels)
loss.backward()
optimizer.step()
pred = torch.argmax(logps, dim=1)
correct = pred.eq(labels)
running_loss += loss.item()
accuracy += torch.mean(correct.float())
if steps % print_every == 0:
after_train_accuracy = accuracy / print_every
test_loss = 0
accuracy = 0
net.eval()
with torch.no_grad():
for inputs, labels in test_loader:
inputs, labels = (inputs.to(device), labels.to(device))
logps = net.forward(inputs.view(-1, 3 * 40 * 40))
batch_loss = criterion(logps, labels)
test_loss += batch_loss.item()
pred = torch.argmax(logps, dim=1)
correct = pred.eq(labels)
accuracy += torch.mean(correct.float())
traininglosses.append(running_loss / print_every)
trainingaccuracy.append(after_train_accuracy)
testinglosses.append(test_loss / len(test_loader))
testaccuracy.append(accuracy / len(test_loader))
totalsteps.append(steps)
print(f'Device {device} Epoch {epoch + 1}/{epochs} Step {steps} Train loss: {running_loss / print_every:f} Train accuracy: {after_train_accuracy:f} Test loss: {test_loss / len(test_loader):f} Test accuracy: {accuracy / len(test_loader):f}')
running_loss = 0
accuracy = 0
net.train()
print('Finish Train')
torch.save(net.state_dict(), Path) | code |
73069103/cell_5 | [
"image_output_1.png"
] | import pandas as pd
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
data_lebel = pd.read_csv('/kaggle/input/soft-computing-even-id-dataset/training-a.csv')
data_lebel['digit'].unique() | code |
18155655/cell_9 | [
"image_output_1.png"
] | from IPython.display import Image
Image('../input/hogweed-outscreenshot/screenshot_out_example.jpg') | code |
18155655/cell_2 | [
"image_output_1.png"
] | from IPython.display import Image
from IPython.display import Image
Image('../input/hogweednew/heracleum_lanantum_maxima_03.jpg') | code |
18155655/cell_1 | [
"text_plain_output_1.png"
] | import os
import os
print(os.listdir('../input/hogweed-outscreenshot')) | code |
18155655/cell_5 | [
"image_output_1.png"
] | from IPython.display import Image
Image('../input/hogweed-screenshot/screenshot_example.jpg') | code |
2016010/cell_9 | [
"application_vnd.jupyter.stderr_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train = pd.read_csv('../input/train.csv')
test = pd.read_csv('../input/test.csv')
sub = pd.read_csv('../input/sample_submission.csv')
test.shape
test.dtypes | code |
2016010/cell_4 | [
"text_html_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train = pd.read_csv('../input/train.csv')
test = pd.read_csv('../input/test.csv')
sub = pd.read_csv('../input/sample_submission.csv')
test.shape | code |
2016010/cell_6 | [
"text_plain_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train = pd.read_csv('../input/train.csv')
test = pd.read_csv('../input/test.csv')
sub = pd.read_csv('../input/sample_submission.csv')
test.shape
test.head() | code |
2016010/cell_1 | [
"text_plain_output_1.png"
] | from subprocess import check_output
import numpy as np
import pandas as pd
from subprocess import check_output
print(check_output(['ls', '../input']).decode('utf8')) | code |
2016010/cell_7 | [
"text_plain_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train = pd.read_csv('../input/train.csv')
test = pd.read_csv('../input/test.csv')
sub = pd.read_csv('../input/sample_submission.csv')
sub.head() | code |
2016010/cell_8 | [
"text_plain_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train = pd.read_csv('../input/train.csv')
test = pd.read_csv('../input/test.csv')
sub = pd.read_csv('../input/sample_submission.csv')
train.shape
train.dtypes | code |
2016010/cell_15 | [
"text_plain_output_1.png"
] | from sklearn import model_selection
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.linear_model import LogisticRegression
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import re
import time
train = pd.read_csv('../input/train.csv')
test = pd.read_csv('../input/test.csv')
sub = pd.read_csv('../input/sample_submission.csv')
train.shape
test.shape
train.dtypes
test.dtypes
import re
def clean_text(text):
text = re.sub("[^A-za-z0-9^,?!.\\/'+-=]", ' ', text)
text = re.sub("what's", 'what is ', text)
text = re.sub("\\'s", ' ', text)
text = re.sub("\\'ve", ' have ', text)
text = re.sub("can't", 'cannot ', text)
text = re.sub("n't", ' not ', text)
text = re.sub("i'm", 'i am ', text)
text = re.sub("\\'re", ' are ', text)
text = re.sub("\\'d", ' would ', text)
text = re.sub("\\'ll", ' will ', text)
text = re.sub("\\'scuse", ' excuse ', text)
text = re.sub(',', ' ', text)
text = re.sub('\\.', ' ', text)
text = re.sub('!', ' _exclamationmark_ ', text)
text = re.sub('\\?', ' _questionmark_ ', text)
return text
def build_data_set(ngram=3, stem=False, max_features=2000, min_df=2, remove_stopwords=True):
train = pd.read_csv('../input/train.csv')
test = pd.read_csv('../input/test.csv')
test.fillna('missing', inplace=True)
clean_train_comments = []
for i in range(train.shape[0]):
clean_train_comments.append(clean_text(train['comment_text'][i]))
for i in range(test.shape[0]):
clean_train_comments.append(clean_text(test['comment_text'][i]))
qs = pd.Series(clean_train_comments).astype(str)
if not stem:
vect = TfidfVectorizer(analyzer=u'word', stop_words='english', min_df=min_df, ngram_range=(1, ngram), max_features=max_features)
ifidf_vect = vect.fit_transform(qs)
X = ifidf_vect.toarray()
X_train = X[:train.shape[0]]
X_test = X[train.shape[0]:]
else:
vect_stem = StemmedTfidfVectorizer(analyzer=u'word', stop_words='english', min_df=min_df, ngram_range=(1, ngram), max_features=max_features)
ifidf_vect_stem = vect_stem.fit_transform(qs)
X = ifidf_vect_stem.toarray()
X_train = X[:train.shape[0]]
X_test = X[train.shape[0]:]
Y_train = train[['toxic', 'severe_toxic', 'obscene', 'threat', 'insult', 'identity_hate']]
assert Y_train.shape[0] == X_train.shape[0]
del train, test
return (X_train, X_test, Y_train)
labels = ['toxic', 'severe_toxic', 'obscene', 'threat', 'insult', 'identity_hate']
params = {'toxic': {'ngrams': 1, 'stem': True, 'max_features': 1000, 'C': 10}, 'threat': {'ngrams': 1, 'stem': False, 'max_features': 1000, 'C': 10}, 'severe_toxic': {'ngrams': 1, 'stem': True, 'max_features': 1000, 'C': 1.2}, 'obscene': {'ngrams': 1, 'stem': True, 'max_features': 1000, 'C': 10}, 'insult': {'ngrams': 1, 'stem': True, 'max_features': 1000, 'C': 1.2}, 'identity_hate': {'ngrams': 1, 'stem': True, 'max_features': 1000, 'C': 10}}
start_time = time.time()
for label in labels:
print('>>> processing ', label)
X_train, X_test, Y_train = build_data_set(ngram=params[label]['ngrams'], stem=params[label]['stem'], max_features=params[label]['max_features'], min_df=2, remove_stopwords=True)
Y_train_lab = Y_train[label]
seed = 7
scoring = 'accuracy'
models = []
models.append(('LR', LogisticRegression()))
results = []
names = []
for name, model in models:
kfold = model_selection.KFold(n_splits=10, random_state=seed)
cv_results = model_selection.cross_val_score(model, X_train, Y_train_lab, cv=kfold, scoring=scoring)
results.append(cv_results)
names.append(name)
msg = '%s: %f (%f)' % (name, cv_results.mean(), cv_results.std())
print(msg) | code |
2016010/cell_16 | [
"text_plain_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train = pd.read_csv('../input/train.csv')
test = pd.read_csv('../input/test.csv')
sub = pd.read_csv('../input/sample_submission.csv')
sub[label] = output
sub.to_csv('output_LR.csv', index=False) | code |
2016010/cell_3 | [
"text_html_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train = pd.read_csv('../input/train.csv')
test = pd.read_csv('../input/test.csv')
sub = pd.read_csv('../input/sample_submission.csv')
train.shape | code |
2016010/cell_5 | [
"text_html_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
train = pd.read_csv('../input/train.csv')
test = pd.read_csv('../input/test.csv')
sub = pd.read_csv('../input/sample_submission.csv')
train.shape
train.head() | code |
1008715/cell_13 | [
"text_plain_output_1.png"
] | import numpy as np
import pandas as pd
import pandas as pd
from sklearn.linear_model import LogisticRegression
from sklearn.svm import SVC, LinearSVC
from sklearn.ensemble import RandomForestClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn.naive_bayes import GaussianNB
train = pd.read_csv('../input/train.csv')
test = pd.read_csv('../input/test.csv')
full = pd.concat([train, test], join='outer')
full['family'] = full.Parch + full.SibSp
import numpy as np
full['singlton'] = np.where(full.family == 1, 1, 0)
full['small'] = np.where(np.logical_and(full.family > 1, full.family < 5), 1, 0)
full['large'] = np.where(full.family > 4, 1, 0)
full['Fare'].fillna(full.Fare.mean(), inplace=True)
full.info() | code |
1008715/cell_4 | [
"text_html_output_1.png"
] | import pandas as pd
import pandas as pd
from sklearn.linear_model import LogisticRegression
from sklearn.svm import SVC, LinearSVC
from sklearn.ensemble import RandomForestClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn.naive_bayes import GaussianNB
train = pd.read_csv('../input/train.csv')
test = pd.read_csv('../input/test.csv')
full = pd.concat([train, test], join='outer')
full.info() | code |
1008715/cell_23 | [
"text_plain_output_1.png"
] | import numpy as np
import pandas as pd
import pandas as pd
from sklearn.linear_model import LogisticRegression
from sklearn.svm import SVC, LinearSVC
from sklearn.ensemble import RandomForestClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn.naive_bayes import GaussianNB
train = pd.read_csv('../input/train.csv')
test = pd.read_csv('../input/test.csv')
full = pd.concat([train, test], join='outer')
full['family'] = full.Parch + full.SibSp
import numpy as np
full['singlton'] = np.where(full.family == 1, 1, 0)
full['small'] = np.where(np.logical_and(full.family > 1, full.family < 5), 1, 0)
full['large'] = np.where(full.family > 4, 1, 0)
full['Fare'].fillna(full.Fare.mean(), inplace=True)
full.Age.isnull().sum()
rand = np.random.randint(full.Age.mean() - full.Age.std(), full.Age.mean() + full.Age.std(), full.Age.isnull().sum())
full.loc[full.Age.isnull(), 'Age'] = rand
full.drop(['Cabin', 'Name', 'Ticket', 'Sex'], axis=1, inplace=True)
full.head() | code |
1008715/cell_11 | [
"text_html_output_1.png"
] | import numpy as np
import pandas as pd
import pandas as pd
from sklearn.linear_model import LogisticRegression
from sklearn.svm import SVC, LinearSVC
from sklearn.ensemble import RandomForestClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn.naive_bayes import GaussianNB
train = pd.read_csv('../input/train.csv')
test = pd.read_csv('../input/test.csv')
full = pd.concat([train, test], join='outer')
full['family'] = full.Parch + full.SibSp
import numpy as np
full['singlton'] = np.where(full.family == 1, 1, 0)
full['small'] = np.where(np.logical_and(full.family > 1, full.family < 5), 1, 0)
full['large'] = np.where(full.family > 4, 1, 0)
full.info() | code |
1008715/cell_7 | [
"text_html_output_1.png"
] | import pandas as pd
import seaborn as sns
import pandas as pd
from sklearn.linear_model import LogisticRegression
from sklearn.svm import SVC, LinearSVC
from sklearn.ensemble import RandomForestClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn.naive_bayes import GaussianNB
train = pd.read_csv('../input/train.csv')
test = pd.read_csv('../input/test.csv')
full = pd.concat([train, test], join='outer')
full['family'] = full.Parch + full.SibSp
sns.countplot(x='family', hue='Survived', data=full) | code |
1008715/cell_18 | [
"text_plain_output_1.png"
] | import numpy as np
import pandas as pd
import seaborn as sns
import pandas as pd
from sklearn.linear_model import LogisticRegression
from sklearn.svm import SVC, LinearSVC
from sklearn.ensemble import RandomForestClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn.naive_bayes import GaussianNB
train = pd.read_csv('../input/train.csv')
test = pd.read_csv('../input/test.csv')
full = pd.concat([train, test], join='outer')
full['family'] = full.Parch + full.SibSp
import numpy as np
full['singlton'] = np.where(full.family == 1, 1, 0)
full['small'] = np.where(np.logical_and(full.family > 1, full.family < 5), 1, 0)
full['large'] = np.where(full.family > 4, 1, 0)
full['Fare'].fillna(full.Fare.mean(), inplace=True)
full.Age.isnull().sum()
rand = np.random.randint(full.Age.mean() - full.Age.std(), full.Age.mean() + full.Age.std(), full.Age.isnull().sum())
full.loc[full.Age.isnull(), 'Age'] = rand
sns.factorplot(x='Age', hue='Survived', row='Sex', data=full, kind='count', ci=None, aspect=5) | code |
1008715/cell_3 | [
"text_plain_output_1.png",
"image_output_1.png"
] | import pandas as pd
import pandas as pd
from sklearn.linear_model import LogisticRegression
from sklearn.svm import SVC, LinearSVC
from sklearn.ensemble import RandomForestClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn.naive_bayes import GaussianNB
train = pd.read_csv('../input/train.csv')
test = pd.read_csv('../input/test.csv')
full = pd.concat([train, test], join='outer')
full.head() | code |
1008715/cell_17 | [
"text_plain_output_1.png"
] | import numpy as np
import pandas as pd
import pandas as pd
from sklearn.linear_model import LogisticRegression
from sklearn.svm import SVC, LinearSVC
from sklearn.ensemble import RandomForestClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn.naive_bayes import GaussianNB
train = pd.read_csv('../input/train.csv')
test = pd.read_csv('../input/test.csv')
full = pd.concat([train, test], join='outer')
full['family'] = full.Parch + full.SibSp
import numpy as np
full['singlton'] = np.where(full.family == 1, 1, 0)
full['small'] = np.where(np.logical_and(full.family > 1, full.family < 5), 1, 0)
full['large'] = np.where(full.family > 4, 1, 0)
full['Fare'].fillna(full.Fare.mean(), inplace=True)
full.Age.isnull().sum()
rand = np.random.randint(full.Age.mean() - full.Age.std(), full.Age.mean() + full.Age.std(), full.Age.isnull().sum())
full.loc[full.Age.isnull(), 'Age'] = rand
full.info() | code |
1008715/cell_14 | [
"text_plain_output_1.png",
"image_output_1.png"
] | import numpy as np
import pandas as pd
import pandas as pd
from sklearn.linear_model import LogisticRegression
from sklearn.svm import SVC, LinearSVC
from sklearn.ensemble import RandomForestClassifier
from sklearn.neighbors import KNeighborsClassifier
from sklearn.naive_bayes import GaussianNB
train = pd.read_csv('../input/train.csv')
test = pd.read_csv('../input/test.csv')
full = pd.concat([train, test], join='outer')
full['family'] = full.Parch + full.SibSp
import numpy as np
full['singlton'] = np.where(full.family == 1, 1, 0)
full['small'] = np.where(np.logical_and(full.family > 1, full.family < 5), 1, 0)
full['large'] = np.where(full.family > 4, 1, 0)
full['Fare'].fillna(full.Fare.mean(), inplace=True)
full.Age.isnull().sum() | code |
129000049/cell_42 | [
"text_plain_output_1.png"
] | from sklearn.linear_model import LinearRegression
from sklearn.linear_model import LinearRegression
from sklearn.linear_model import LogisticRegression
import numpy as np
import pandas as pd
wdf = pd.read_csv('/kaggle/input/historical-weather-data-for-indian-cities/jaipur.csv', parse_dates=['date_time'], index_col='date_time')
train_X.shape
train_y.shape
model = LinearRegression()
model.fit(train_X, train_y)
prediction = model.predict(test_X)
mae = np.mean(np.absolute(prediction - test_y))
variance_score = model.score(test_X, test_y)
prediction = np.round(prediction, 2)
results = pd.DataFrame({'Actual': test_y, 'Prediction': prediction, 'Difference': test_y - prediction})
model = LogisticRegression()
model.fit(train_X, train_y)
prediction = model.predict(test_X)
mae = np.mean(np.absolute(prediction - test_y))
print('Mean Absolute Error:', mae) | code |
129000049/cell_13 | [
"text_plain_output_1.png"
] | import pandas as pd
wdf = pd.read_csv('/kaggle/input/historical-weather-data-for-indian-cities/jaipur.csv', parse_dates=['date_time'], index_col='date_time')
wdf.columns
wdf.shape
wdf.isnull().any()
wdf_num = wdf.loc[:, ['mintempC', 'tempC', 'HeatIndexC', 'pressure']]
wdf_num.head() | code |
129000049/cell_9 | [
"text_plain_output_1.png"
] | import pandas as pd
wdf = pd.read_csv('/kaggle/input/historical-weather-data-for-indian-cities/jaipur.csv', parse_dates=['date_time'], index_col='date_time')
wdf.columns
wdf.shape
wdf.describe() | code |
129000049/cell_25 | [
"text_plain_output_1.png"
] | train_y.shape
train_y.head() | code |
129000049/cell_4 | [
"image_output_3.png",
"image_output_2.png",
"image_output_1.png"
] | import pandas as pd
wdf = pd.read_csv('/kaggle/input/historical-weather-data-for-indian-cities/jaipur.csv', parse_dates=['date_time'], index_col='date_time')
wdf.head(5) | code |
129000049/cell_34 | [
"text_plain_output_1.png"
] | from sklearn.linear_model import LinearRegression
from sklearn.linear_model import LinearRegression
import numpy as np
import pandas as pd
wdf = pd.read_csv('/kaggle/input/historical-weather-data-for-indian-cities/jaipur.csv', parse_dates=['date_time'], index_col='date_time')
train_X.shape
train_y.shape
model = LinearRegression()
model.fit(train_X, train_y)
prediction = model.predict(test_X)
mae = np.mean(np.absolute(prediction - test_y))
print('Mean Absolute Error:', mae)
variance_score = model.score(test_X, test_y)
print('Variance score:', variance_score)
prediction = np.round(prediction, 2)
results = pd.DataFrame({'Actual': test_y, 'Prediction': prediction, 'Difference': test_y - prediction})
print(results) | code |
129000049/cell_23 | [
"text_plain_output_1.png"
] | train_y.shape | code |
129000049/cell_44 | [
"text_plain_output_1.png"
] | from sklearn.linear_model import LinearRegression
from sklearn.linear_model import LinearRegression
from sklearn.linear_model import LogisticRegression
import numpy as np
import pandas as pd
wdf = pd.read_csv('/kaggle/input/historical-weather-data-for-indian-cities/jaipur.csv', parse_dates=['date_time'], index_col='date_time')
train_X.shape
train_y.shape
model = LinearRegression()
model.fit(train_X, train_y)
prediction = model.predict(test_X)
mae = np.mean(np.absolute(prediction - test_y))
variance_score = model.score(test_X, test_y)
prediction = np.round(prediction, 2)
results = pd.DataFrame({'Actual': test_y, 'Prediction': prediction, 'Difference': test_y - prediction})
model = LogisticRegression()
model.fit(train_X, train_y)
prediction = model.predict(test_X)
variance_score = model.score(test_X, test_y)
print('Variance score:', variance_score) | code |
129000049/cell_6 | [
"text_plain_output_1.png"
] | import pandas as pd
wdf = pd.read_csv('/kaggle/input/historical-weather-data-for-indian-cities/jaipur.csv', parse_dates=['date_time'], index_col='date_time')
wdf.columns | code |
129000049/cell_29 | [
"text_html_output_1.png"
] | import matplotlib.pyplot as plt
import matplotlib.pyplot as plt
import pandas as pd
wdf = pd.read_csv('/kaggle/input/historical-weather-data-for-indian-cities/jaipur.csv', parse_dates=['date_time'], index_col='date_time')
wdf.columns
wdf.shape
wdf.isnull().any()
wdf_num = wdf.loc[:, ['mintempC', 'tempC', 'HeatIndexC', 'pressure']]
wdf_num.shape
wdf_num.columns
weth = wdf_num['2019':'2020']
plt.hexbin(weth.mintempC, weth.tempC, gridsize=20)
plt.xlabel('Minimum Temperature')
plt.ylabel('Temperature')
plt.show()
plt.hexbin(weth.HeatIndexC, weth.tempC, gridsize=20)
plt.xlabel('Heat Index')
plt.ylabel('Temperature')
plt.show()
plt.hexbin(weth.pressure, weth.tempC, gridsize=20)
plt.xlabel('Pressure')
plt.ylabel('Temperature')
plt.show() | code |
129000049/cell_11 | [
"text_html_output_1.png"
] | import pandas as pd
wdf = pd.read_csv('/kaggle/input/historical-weather-data-for-indian-cities/jaipur.csv', parse_dates=['date_time'], index_col='date_time')
wdf.columns
wdf.shape
wdf.isnull().any() | code |
End of preview. Expand
in Dataset Viewer.
README.md exists but content is empty.
- Downloads last month
- 43