system HF staff commited on
Commit
53c9c50
1 Parent(s): 3bae579

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -4
README.md CHANGED
@@ -1,10 +1,10 @@
1
  ## TextAttack Model Card
2
  This `bert-base-uncased` model was fine-tuned for sequence classification using TextAttack
3
  and the yelp_polarity dataset loaded using the `nlp` library. The model was fine-tuned
4
- for 5 epochs with a batch size of 8, a learning
5
- rate of 5e-05, and a maximum sequence length of 512.
6
  Since this was a classification task, the model was trained with a cross-entropy loss function.
7
- The best score the model achieved on this task was 0.5000526315789474, as measured by the
8
- eval set accuracy, found after 1 epoch.
9
 
10
  For more information, check out [TextAttack on Github](https://github.com/QData/TextAttack).
 
1
  ## TextAttack Model Card
2
  This `bert-base-uncased` model was fine-tuned for sequence classification using TextAttack
3
  and the yelp_polarity dataset loaded using the `nlp` library. The model was fine-tuned
4
+ for 5 epochs with a batch size of 16, a learning
5
+ rate of 5e-05, and a maximum sequence length of 256.
6
  Since this was a classification task, the model was trained with a cross-entropy loss function.
7
+ The best score the model achieved on this task was 0.9699473684210527, as measured by the
8
+ eval set accuracy, found after 4 epochs.
9
 
10
  For more information, check out [TextAttack on Github](https://github.com/QData/TextAttack).