By using this site, you agree to our updated Privacy Policy and our Terms of Use. Manage your Cookies Settings.
434,905 Members | 2,083 Online
Bytes IT Community
+ Ask a Question
Need help? Post your question and get tips & solutions from a community of 434,905 IT Pros & Developers. It's quick & easy.

Why I get the X_train_std is not defined

P: 7
May I know why I get the error message -

NameError: name 'X_train_std' is not defined






Expand|Select|Wrap|Line Numbers
  1.  
  2. from sklearn.linear_model import LogisticRegression
  3.  
  4. lr = LogisticRegression(C=1000.0, random_state=0)
  5. lr.fit(X_train_std, y_train)
  6.  
  7. plot_decision_regions(X_combined_std,
  8.                       y_combined, classifier=lr,
  9.                       test_idx=range(105,150))
  10.  
  11. plt.xlabel('petal length [standardized]')
  12. plt.ylabel('petal width [standardized]')
  13. plt.legend(loc='upper left')
  14. plt.tight_layout()
  15.  
  16. plt.show()
  17.  
  18. lr.predict_proba(X_test_std[0,:])
  19.  
  20. weights, params = [], []
  21. for c in np.arange(-5, 5):
  22.     lr = LogisticRegression(C=10**c, random_state=0)
  23.     lr.fit(X_train_std, y_train)
  24.     weights.append(lr.coef_[1])
  25.     params.append(10**c)
  26. weights = np.array(weights)
  27. plt.plot(params, weights[:, 0],
  28.          label='petal length')
  29. plt.plot(params, weights[:, 1], linestyle='--',
  30.          label='petal width')
  31. plt.ylabel('weight coefficient')
  32. plt.xlabel('C')
  33. plt.legend(loc='upper left')
  34. plt.xscale('log')
  35. plt.show()
  36.  
Attached Images
File Type: jpg Logistic regression.jpg (57.8 KB, 21 views)
Attached Files
File Type: pdf Logistic regression.pdf (291.6 KB, 32 views)
Mar 18 '19 #1
Share this Question
Share on Google+
2 Replies


Expert 100+
P: 621
Where do you define it?
Mar 19 '19 #2

P: 7




Look carefully -





from sklearn.model_selection import train_test_split






Mar 20 '19 #3

Post your reply

Sign in to post your reply or Sign up for a free account.