- 基于TSNE可视化
例子01：
import matplotlib
import matplotlib.pyplot as plt
import pandas as pd
import seaborn as sns
from sklearn.decomposition import PCA
from sklearn.manifold import TSNE
train_df = pd.read_csv("../data/d_train.csv")
valid_df = pd.read_csv("../data/d_test.csv")
X = train_df.loc[0:5000, 'pixel0':'pixel783']
y = train_df.label
print(train_df.shape)
print('pca...')
pca = PCA(n_components=30)
X_pca = pca.fit_transform(X)
print('tsne...')
tsne = TSNE()
X_tsne = tsne.fit_transform(X_pca[:5000])
matplotlib.rcParams['figure.figsize'] = (10.0, 10.0)
proj = pd.DataFrame(X_tsne)
proj.columns = ['comp_1', 'comp_2']
proj['labels'] = y
print('lmplot...')
sns.lmplot("comp_1", "comp_2", hue="labels", data=proj.sample(2000), fit_reg=False)
plt.title('Digit Distribution')
plt.show()可视化显示：
例子2：
import matplotlib.pyplot as plt
import pandas as pd
train_df = pd.read_csv(“../data/d_train.csv”)
valid_df = pd.read_csv(“../data/d_test.csv”)
X = train_df.loc[0:5, ‘pixel0’:’pixel783’]
X = X / 255.0
X = X.values.reshape(-1, 28, 28, 1)
fig = plt.figure()
ax = fig.add_subplot(221)
ax.imshow(X[0][:, :, 0])
ax = fig.add_subplot(222)
ax.imshow(X[1][:, :, 0])
ax = fig.add_subplot(223)
ax.imshow(X[2][:, :, 0])
ax = fig.add_subplot(224)
ax.imshow(X[3][:, :, 0])
plt.show()
参考：
https://www.kaggle.com/yassineghouzam/introduction-to-cnn-keras-0-997-top-6
对于曲线的一些附加教程：
http://blog.csdn.net/cheng9981/article/details/60583273