Das deutsche Python-Forum

Code: Alles auswählen

import tensorflow as tf
# etc

Code: Alles auswählen

import tensorflow as tf
from tensorflow import keras
from tensorflow.keras.models import Sequential

from tensorflow.keras.layers import Dense, Flatten, Conv2D, MaxPooling2D, Dropout
from tensorflow.keras import layers
from tensorflow.keras.utils import to_categorical
import tensorflow_datasets as tfds
import seaborn as sns

import os # noch rausfinden, was dieses Package bringt
import PIL

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
plt.style.use("fivethirtyeight")

import sklearn
from sklearn.metrics import classification_report  
from sklearn.metrics import confusion_matrix
import sklearn.metrics as metrics  

tf.keras.datasets.mnist.load_data(path="mnist.npz")
Data1 = tf.keras.datasets.mnist.load_data()
Data1

builder = tfds.builder("Data1")
info = builder.info

(x_train, y_train), (x_test, y_test) = Data1

print(type(x_train))
print(type(y_train))
print(type(x_test))
print(type(y_test))

print("x_train shape:", x_train.shape)
print("y_train shape:", y_train.shape)
print("x_test shape:", x_test.shape)
print("y_test shape:", y_test.shape)

y_train

df_y_train = pd.DataFrame(y_train)
df_y_train.head()

df_y_train[0].value_counts(sort=False)

df_y_train[0].value_counts(normalize=True)

y_test

df_y_test = pd.DataFrame(y_test)
df_y_test.head()

df_y_test[0].value_counts(sort=False)

df_y_test[0].value_counts(normalize=True)

index = 1
classification = ["0", "1", "2", "3", "4", "5", "6", "7", "8", "9"]
print("the image class is:", classification[y_train[index][1]])

index = 1
x_train[index]

img = plt.imshow(x_train[index]) 

print("the image label is:", y_train[index])

print(y_train)
f, ax = plt.subplots(figsize=(16, 16))
sns.heatmap(index, annot=True, fmt='.1f', square=True, cmap="YlGnBu")
plt.show()

y_train_one_hot = to_categorical(y_train)
y_test_one_hot = to_categorical(y_test)

print(y_train_one_hot)

print("the one hot label is:", y_train_one_hot[index])

x_train = x_train / 255               
x_test = x_test / 255

model = Sequential()
model.add(Conv2D(28, (5,5), activation = "relu", input_shape = (28, 28, 2)))
model.add(MaxPooling2D(pool_size = (2, 2)))
model.add(Conv2D(28, (5,5), activation = "relu"))
model.add(MaxPooling2D(pool_size = (2, 2)))
model.add(Flatten())
model.add(Dense(1000, activation = "relu"))
model.add(Dropout(0.5))
model.add(Dense(500, activation = "relu"))
model.add(Dropout(0.5))
model.add(Dense(250, activation = "relu"))
model.add(Dense(10, activation = "softmax"))

model.compile(loss=keras.losses.categorical_crossentropy,
              optimizer=keras.optimizers.Adadelta(),
              metrics=['accuracy'])
              
hist = model.fit(x_train, y_train_one_hot, 
                 batch_size = 256,
                 epochs = 10,
                 verbose=1,
                 validation_split = 0.2, shuffle=True)
                 
from sklearn.naive_bayes import GaussianNB
gnb=GaussianNB()

model = gnb.fit(x_train, y_train_one_hot)

history = model.fit(x_train, y_train_one_hot, batch_size=32, epochs=10, 
                verbose=1,
                validation_split=0.2, shuffle=True)

Code: Alles auswählen

print(x_train.shape)
x_train = np.expand_dims(x_train, -1)
x_test = np.expand_dims(x_test, -1)
print(x_train.shape)

Code: Alles auswählen

model = Sequential()
model.add(Conv2D(28, (5,5), activation = "relu", input_shape = (28, 28, 1)))

Code: Alles auswählen

model.compile(loss=keras.losses.sparse_categorical_crossentropy,
              optimizer=keras.optimizers.Adadelta(),
              metrics=['accuracy'])
              
history = model.fit(x_train, y_train,  # hier kann der normale Labelvektor rein
                 batch_size = 32, epochs = 50, verbose=1,
                 validation_split = 0.2, shuffle=True)