自编码器降维python
时间: 2023-10-13 15:20:36 浏览: 59
以下是使用Python实现自编码器降维的示例代码:
首先,导入必要的库:
```python
import numpy as np
import matplotlib.pyplot as plt
from keras.layers import Input, Dense
from keras.models import Model
from sklearn.datasets import load_digits
from sklearn.model_selection import train_test_split
```
接着,加载数据集并将其分成训练集和测试集:
```python
# 加载手写数字数据集
digits = load_digits()
X_train, X_test, y_train, y_test = train_test_split(digits.data, digits.target, test_size=0.2, random_state=42)
```
然后,定义自编码器模型:
```python
# 定义输入层
input_img = Input(shape=(64,))
# 定义编码层
encoded = Dense(32, activation='relu')(input_img)
# 定义解码层
decoded = Dense(64, activation='sigmoid')(encoded)
# 定义自编码器模型
autoencoder = Model(input_img, decoded)
```
接下来,编译自编码器模型并训练:
```python
# 编译自编码器模型
autoencoder.compile(optimizer='adam', loss='binary_crossentropy')
# 训练自编码器模型
autoencoder.fit(X_train, X_train, epochs=50, batch_size=256, shuffle=True, validation_data=(X_test, X_test))
```
训练完成后,我们可以使用训练好的编码器部分来将输入数据降维:
```python
# 定义编码器模型
encoder = Model(input_img, encoded)
# 使用编码器将输入数据降维
X_train_encoded = encoder.predict(X_train)
X_test_encoded = encoder.predict(X_test)
```
最后,我们可以可视化降维后的数据:
```python
# 可视化降维后的数据
plt.scatter(X_test_encoded[:, 0], X_test_encoded[:, 1], c=y_test)
plt.colorbar()
plt.show()
```
完整代码如下:
```python
import numpy as np
import matplotlib.pyplot as plt
from keras.layers import Input, Dense
from keras.models import Model
from sklearn.datasets import load_digits
from sklearn.model_selection import train_test_split
# 加载手写数字数据集
digits = load_digits()
X_train, X_test, y_train, y_test = train_test_split(digits.data, digits.target, test_size=0.2, random_state=42)
# 定义输入层
input_img = Input(shape=(64,))
# 定义编码层
encoded = Dense(32, activation='relu')(input_img)
# 定义解码层
decoded = Dense(64, activation='sigmoid')(encoded)
# 定义自编码器模型
autoencoder = Model(input_img, decoded)
# 编译自编码器模型
autoencoder.compile(optimizer='adam', loss='binary_crossentropy')
# 训练自编码器模型
autoencoder.fit(X_train, X_train, epochs=50, batch_size=256, shuffle=True, validation_data=(X_test, X_test))
# 定义编码器模型
encoder = Model(input_img, encoded)
# 使用编码器将输入数据降维
X_train_encoded = encoder.predict(X_train)
X_test_encoded = encoder.predict(X_test)
# 可视化降维后的数据
plt.scatter(X_test_encoded[:, 0], X_test_encoded[:, 1], c=y_test)
plt.colorbar()
plt.show()
```