Merge pull request #194 from ccyf00/dev

knn

competitions/getting-started/digit-recognizer/ccyf00_knn.md

@@ -0,0 +1,46 @@
+#!/usr/bin/python
+# coding: utf-8
+'''
+Created on 2018-05-16
+Update  on 2018-05-16
+Author: ccyf00
+Github: https://github.com/ccyf00/kaggle-1
+'''
+
+
+import pandas as pd
+import numpy as np
+import matplotlib.pyplot as plt
+import seaborn as sns
+from sklearn.model_selection import train_test_split
+from sklearn.decomposition import PCA
+from sklearn.neighbors import KNeighborsClassifier
+train=pd.read_csv('datasets/getting-started/digit-recognizer/input/train.csv')
+test=pd.read_csv('datasets/getting-started/digit-recognizer/input/test.csv')
+Y_train=train["label"]
+X_train=train.drop(['label'],axis=1)
+del train
+
+
+pca = PCA(n_components=45)
+X_train_transformed=pca.fit_transform(X_train)
+X_test_transformed=pca.transform(test)
+X_train_pca, X_test_pca, Y_train_pca, Y_test_pca = train_test_split(
+    X_train_transformed, Y_train, test_size=0.1, random_state=13)
+components = [10,15,20,25,30,35,40,45]
+neighbors = [2,3,4,5,6,7]
+scores = np.zeros((components[len(components)-1]+1,neighbors[len(neighbors)-1]+1))
+for component in components:
+    for n in neighbors:
+        knn=KNeighborsClassifier(n_neighbors=n)
+        knn.fit(X_train_pca[:,:component],Y_train_pca)
+        score = knn.score(X_test_pca[:,:component],Y_test_pca)
+        scores[component][n]=score
+        print('Components=',component,'neighbors = ',n,'Score = ',score)
+knn=KNeighborsClassifier(n_neighbors=5)
+knn.fit(X_train_pca[:,:35],Y_train_pca)
+predictLabel=knn.predict(X_test_transformed[:,:35])
+Submission = pd.DataFrame({"ImageId":range(1,predictLabel.shape[0]+1),
+                           "Label":predictLabel})
+Submission.to_csv("datasets/getting-started/digit-recognizer/ouput/KnnMnistSubmission.csv", index=False)
+#分数：0.97385