how to calculate the precision and F1?

Question

i am using TF_IDF for for feature selection and Naive Bayes Classifier. i want to calculate the total accuracy and precision.

accuracy = train_model(naive_bayes.MultinomialNB(), xtrain_count, train_y, xvalid_count)
  print("NB, Count Vectors: ", accuracy)

 # Naive Bayes on Word Level TF IDF Vectors
  accuracy = train_model(naive_bayes.MultinomialNB(), xtrain_tfidf, train_y, xvalid_tfidf)
    print("NB, WordLevel TF-IDF: ", accuracy)

    # Naive Bayes on Ngram Level TF IDF Vectors
   accuracy = train_model(naive_bayes.MultinomialNB(), xtrain_tfidf_ngram, train_y, 
   xvalid_tfidf_ngram)
  print("NB, N-Gram Vectors: ", accuracy)

 # Naive Bayes on Character Level TF IDF Vectors
  accuracy = train_model(naive_bayes.MultinomialNB(), xtrain_tfidf_ngram_chars, train_y, 
   xvalid_tfidf_ngram_chars)
  print("NB, CharLevel Vectors: ", accuracy)

Nice talking to you hope your problem is resolved – Divyessh Jul 18 '20 at 13:44 — Divyessh, Jul 18 '20 at 13:44

score 3 · Answer 1 · answered Jul 18 '20 at 13:07

3

use this:

from sklearn.metrics import classification_report

print(classification_report(true_value,predicted_value))

this will give you all that you want

answered Jul 18 '20 at 13:07

Divyessh

2,540
1
7
24

what is true value and predicted value? – Irfan Yaqub Jul 18 '20 at 13:08
what do you want to calculate accuracy and f1 for – Divyessh Jul 18 '20 at 13:10
1

what are you using natural language processing for?? – Divyessh Jul 18 '20 at 13:14
i am receiving the three different accuracies, word, character and N gram level, idk which one to use , i wanted one accuracy as a result. – Irfan Yaqub Jul 18 '20 at 13:15
please give a brief about your project @ MR.Robot – Divyessh Jul 18 '20 at 13:17
i want to calculate the accuracy by using TF_IDF as feature selection and Naive Baise as classifier, but as TF-IDF gives 3 kind of features. as you can see above, the model is being used at different categories. i want one accuracy as whole my model perdicted. – Irfan Yaqub Jul 18 '20 at 13:19
According to me if you are on a spam vs ham type of project then you should consider words or N-grams – Divyessh Jul 18 '20 at 13:20
if you are on vowel detection or some kind of that thing then you should go with the character one – Divyessh Jul 18 '20 at 13:20
In max of the cases the words accuracy is the best – Divyessh Jul 18 '20 at 13:21
Are you using tfidftransformer or anything else of tfidf – Divyessh Jul 18 '20 at 13:22
1

And tfidf dont give you the accuracy , it gives you the frequency, both are different – Divyessh Jul 18 '20 at 13:23
i have a title(where paper published)of research paper and abstract of research paper. so i have to perdict the where paper should publish. – Irfan Yaqub Jul 18 '20 at 13:23
Let us [continue this discussion in chat](https://chat.stackoverflow.com/rooms/218091/discussion-between-divyessh-maheshwari-and-mr-robot). – Divyessh Jul 18 '20 at 13:24

how to calculate the precision and F1?

1 Answers1