dangkhoa99
/

falcon-7b-finetuned-QA-MRC-4-bit

Text Generation

text-generation-inference

endpoints-template

Inference Endpoints

Model card Files Files and versions Community

dangkhoa99 commited on Aug 19, 2023

Commit

3909063

•

1 Parent(s): 37eef1d

Update README.md

Files changed (1) hide show

README.md +11 -11

README.md CHANGED Viewed

@@ -93,18 +93,18 @@ The following `bitsandbytes` quantization config was used during training:
 Evaluated on the SQuAD 2.0 dev set with the [Metrics](https://huggingface.co/docs/datasets/v2.14.4/en/loading#metrics)
 ```python
-'exact': 71.42314990512334
-'f1': 76.60680328349153
-'total': 10540
-'HasAns_exact': 63.03317535545024
-'HasAns_f1': 73.39065528113719
-'HasAns_total': 5275
-'NoAns_exact': 79.82905982905983
-'NoAns_f1': 79.82905982905983
-'NoAns_total': 5265
-'best_exact': 71.42314990512334
 'best_exact_thresh': 0.0
-'best_f1': 76.60680328349183
 'best_f1_thresh': 0.0
 ```

 Evaluated on the SQuAD 2.0 dev set with the [Metrics](https://huggingface.co/docs/datasets/v2.14.4/en/loading#metrics)
 ```python
+'exact': 71.48993514697212
+'f1': 76.65914166347146
+'total': 11873
+'HasAns_exact': 62.78677462887989
+'HasAns_f1': 73.14001163468224
+'HasAns_total': 5928
+'NoAns_exact': 80.1682085786375
+'NoAns_f1': 80.1682085786375
+'NoAns_total': 5945
+'best_exact': 71.48993514697212
 'best_exact_thresh': 0.0
+'best_f1': 76.65914166347147
 'best_f1_thresh': 0.0
 ```