ML_Sklearn_DecisionTree-master
所属分类：人工智能/神经网络/深度学习
开发工具：Python
文件大小：196KB
下载次数：3
上传日期：2021-03-19 15:01:51
上 传 者：
被迫学习的猫
说明： 合成孔径雷达（SAR）是一种微波设备，能够在复杂
天气状况下产生高分辨率遥感影像。SAR具有分辨率
高、穿透性好的特点，可以全天候、全天时地获取图像信
息。SAR图像变化检测是通过对同一地点不同时刻的
SAR图像进行差异分析，得到所需的地物变化信息。
(Change detection for SAR images can be achieved by clustering the differences image. Since SAR image can be interfered by speckle noise, thereby the change detection effect is affected. A method based on fuzzy clustering with spatial neighborhood information is proposed in this paper to improve clustering accuracy.)
文件列表:
0_data.csv (346, 2019-07-05)
1_decisionTreeClassifier.py (1225, 2019-07-05)
2_decisionTreeClassifier.py (1537, 2019-07-05)
3_decTreeClass_loadIris.py (1126, 2019-07-05)
4_decTreeClass_drawTree.py (1889, 2019-07-05)
4_drawTree.dot (1305, 2019-07-05)
4_drawTree.png (141240, 2019-07-05)
5_decTreeClass_face.png (26139, 2019-07-05)
5_decTreeClass_face.py (1447, 2019-07-05)
6_decTreeRegressor.png (42675, 2019-07-05)
6_decTreeRegressor.py (818, 2019-07-05)
# Decision Tree Classifier & Regressor
## …° Decision Tree Classifier
```python
import pandas as pd
import numpy as np
# =================================
# load csv & create dataframe
df = pd.read_csv('0_data.csv')
# print(df)
# =================================
# convert nominal data => ordinal data
from sklearn.preprocessing import LabelEncoder
labelKantor = LabelEncoder()
df['kantorLE'] = labelKantor.fit_transform(df['kantor'])
labelJabatan = LabelEncoder()
df['jabatanLE'] = labelJabatan.fit_transform(df['jabatan'])
labelTitel = LabelEncoder()
df['titelLE'] = labelTitel.fit_transform(df['titel'])
df = df.drop(
['kantor', 'jabatan', 'titel'],
axis = 'columns'
)
# print(df)
# ===============================
# kantorLE : 0 Facebook, 1 Google, 2 Tesla
# jabatanLE : 0 GM, 1 Manager, 2 Staf
# titelLE : 0 S1, 1 S2
# ===============================
# split: train 80% & test 20%
from sklearn.model_selection import train_test_split
x_train, x_tes, y_train, y_tes = train_test_split(
df[['kantorLE', 'jabatanLE', 'titelLE']],
df['gaji>50'],
test_size = .2,
random_state = 1
)
print(x_train)
# print(len(x_tes))
print(y_train)
# print(len(y_tes))
# ===============================
# decision tree algo
from sklearn import tree
model = tree.DecisionTreeClassifier()
# train
model.fit(x_train, y_train)
# accuracy
acc = model.score(x_train, y_train)
print(acc * 100, '%')
acc2 = model.score(x_tes, y_tes)
print(acc2 * 100, '%')
# predict kantor, jabatan, titel
print(model.predict([[1, 1, 0]]))
print(model.predict([[1, 1, 1]]))
# print(model.predict([[1, 3, 0]]))
# draw the decision tree raph
# import decision tree graph as .dot file
tree.export_graphviz(
model.fit(x_train, y_train),
out_file='4_drawTree.dot',
feature_names=['kantorLE', 'jabatanLE', 'titelLE'],
class_names=['gaji<50', 'gaji>50jt']
)
# go to https://dreampuf.github.io/GraphvizOnline
# to convert the .dot to .png/.svg/image!
```
#
## …± Decision Tree Regressor
```python
import numpy as np
import pandas as pd
x = np.sort(np.random.randn(1000))
y = np.sin(-x)
import matplotlib.pyplot as plt
# plt.scatter(x, y)
# plt.show()
# Fit regression model
from sklearn.tree import DecisionTreeRegressor
regr_1 = DecisionTreeRegressor(max_depth=2)
regr_2 = DecisionTreeRegressor(max_depth=5)
regr_1.fit(x.reshape(-1,1), y)
regr_2.fit(x.reshape(-1,1), y)
# Predict
y_1 = regr_1.predict(x.reshape(-1,1))
y_2 = regr_2.predict(x.reshape(-1,1))
# Plot the results
plt.figure()
plt.scatter(x, y, s=20, edgecolor="black", c="darkorange", label="data")
plt.plot(x, y_1, color="cornflowerblue", label="max_depth=2", linewidth=2)
plt.plot(x, y_2, color="yellowgreen", label="max_depth=5", linewidth=2)
plt.xlabel("data")
plt.ylabel("target")
plt.title("Decision Tree Regression")
plt.legend()
plt.show()
```
#
#### Lintang Wisesa :love_letter: _lintangwisesa@ymail.com_
[Facebook](https://www.facebook.com/lintangbagus) |
[Twitter](https://twitter.com/Lintang_Wisesa) |
[Google+](https://plus.google.com/u/0/+LintangWisesa1) |
[Youtube](https://www.youtube.com/user/lintangbagus) |
:octocat: [GitHub](https://github.com/LintangWisesa) |
[Hackster](https://www.hackster.io/lintangwisesa)
近期下载者：
相关文件：
收藏者：