Python Keras深度学习

Python Keras的AI实例

以下是基于最新Python Keras的AI实例分类整理，涵盖深度学习常见任务和应用场景。所有示例均使用TensorFlow 2.x和Keras API实现，适合快速实践和二次开发。

---

计算机视觉

图像分类（CIFAR-10）

from keras.datasets import cifar10
from keras.models import Sequential
from keras.layers import Conv2D, MaxPooling2D, Dense, Flatten(x_train, y_train), (x_test, y_test) = cifar10.load_data()
model = Sequential([Conv2D(32, (3,3), activation='relu', input_shape=(32,32,3)),MaxPooling2D((2,2)),Flatten(),Dense(10, activation='softmax')
])
model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])
model.fit(x_train, y_train, epochs=10)

目标检测（YOLOv3简化版）

from keras.layers import Input, Conv2D
from keras.models import Modelinputs = Input(shape=(416,416,3))
x = Conv2D(64, (3,3), padding='same')(inputs)
outputs = Conv2D(255, (1,1))(x)
model = Model(inputs, outputs)

图像生成（DCGAN）

from keras.layers import Dense, Reshape, Conv2DTranspose
from keras.models import Sequentialgenerator = Sequential([Dense(7*7*256, input_dim=100),Reshape((7,7,256)),Conv2DTranspose(128, (5,5), strides=1, padding='same'),Conv2DTranspose(64, (5,5), strides=2, padding='same'),Conv2DTranspose(1, (5,5), strides=2, padding='same', activation='tanh')
])

---

自然语言处理

文本分类（IMDB情感分析）

from keras.preprocessing.text import Tokenizer
from keras.layers import Embedding, LSTMtokenizer = Tokenizer(num_words=10000)
tokenizer.fit_on_texts(texts)
sequences = tokenizer.texts_to_sequences(texts)model = Sequential([Embedding(10000, 128),LSTM(64),Dense(1, activation='sigmoid')
])

机器翻译（Seq2Seq）

from keras.layers import GRU, TimeDistributedencoder_inputs = Input(shape=(None, num_encoder_tokens))
encoder = GRU(256, return_state=True)
_, state_h = encoder(encoder_inputs)decoder_inputs = Input(shape=(None, num_decoder_tokens))
decoder_gru = GRU(256, return_sequences=True)
decoder_outputs = decoder_gru(decoder_inputs, initial_state=state_h)
outputs = TimeDistributed(Dense(num_decoder_tokens, activation='softmax'))(decoder_outputs)