文章目錄

• • 1. Baseline KNN
  • 2. Try SVC

Digit Recognizer 練習地址

相關博文：[Hands On ML] 3. 分類（MNIST手寫數字預測）

1. Baseline KNN

• 讀取數據

import pandas as pd train = pd.read_csv('train.csv') X_test = pd.read_csv('test.csv')

• 特征、標簽分離

train.head() y_train = train['label'] X_train = train.drop(['label'], axis=1) X_train

• 網格搜索 KNN 模型最佳參數

from sklearn.neighbors import KNeighborsClassifier from sklearn.model_selection import GridSearchCV from sklearn.metrics import accuracy_score # help(KNeighborsClassifier) para_dict = [{'weights':["uniform", "distance"], 'n_neighbors':[3,4,5], 'leaf_size':[10,20]} ] knn_clf = KNeighborsClassifier() grid_search = GridSearchCV(knn_clf, para_dict, cv=3,scoring='accuracy',n_jobs=-1) grid_search.fit(X_train, y_train) 輸出 GridSearchCV(cv=3, estimator=KNeighborsClassifier(), n_jobs=-1,param_grid=[{'leaf_size': [10, 20], 'n_neighbors': [3, 4, 5],'weights': ['uniform', 'distance']}],scoring='accuracy')

• 最佳參數

grid_search.best_params_ # {'leaf_size': 10, 'n_neighbors': 4, 'weights': 'distance'}

• 最好得分

grid_search.best_score_ # 0.9677619047619048

• 生成 test 集預測結果

y_pred = grid_search.predict(X_test)

• 寫入結果文件

image_id = pd.Series(range(1,len(y_pred)+1)) output = pd.DataFrame({'ImageId':image_id, 'Label':y_pred}) output.to_csv("submission.csv", index=False) # 不要index列

• 預測結果

排行榜

以上 KNN 模型得分 0.97067，目前排名2467

2. Try SVC

• 讀取數據

import pandas as pd train = pd.read_csv('train.csv') X_test = pd.read_csv('test.csv') y_train = train['label'] X_train = train.drop(['label'], axis=1)

• 導入包

from sklearn.pipeline import Pipeline from sklearn.preprocessing import StandardScaler from sklearn.model_selection import train_test_split from sklearn.svm import SVC, LinearSVC from sklearn.model_selection import GridSearchCV from sklearn.metrics import classification_report from sklearn.metrics import accuracy_score

• 搜索最佳參數

pipeline = Pipeline([("scaler",StandardScaler()),('clf', SVC(decision_function_shape="ovr", gamma="auto")) ])from sklearn.model_selection import RandomizedSearchCV from scipy.stats import reciprocal, uniformparam_distributions = {"clf__gamma": reciprocal(0.001, 0.1), "clf__C": uniform(1, 10)} rnd_search_cv = RandomizedSearchCV(pipeline, param_distributions, n_iter=10, verbose=2, cv=3)rnd_search_cv.fit(X_train, y_train)

• 訓練花費12個小時 [Parallel(n_jobs=1)]: Done 30 out of 30 | elapsed: 744.1min finished

rnd_search_cv.best_estimator_

• 最佳評估器

Pipeline(steps=[('scaler', StandardScaler()),('clf',SVC(C=10.729327185542381, gamma=0.0022750096640207287))])

• 最好得分

rnd_search_cv.best_score_ # 0.9584285714285713

• 預測

y_pred = rnd_search_cv.best_estimator_.predict(X_test) image_id = pd.Series(range(1,len(y_pred)+1)) output = pd.DataFrame({'ImageId':image_id, 'Label':y_pred}) output.to_csv("submission_svc.csv", index=False)

SVC 支持向量機分類模型 得分 0.96464 沒有上面 KNN 模型高（KNN 得分 0.97067）

總結

以上是生活随笔為你收集整理的[Kaggle] Digit Recognizer 手写数字识别的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。