Update README.md
#2
by CompactAI - opened
README.md
CHANGED
|
@@ -22,6 +22,8 @@ It is based on a BERT architecture, but trained from scratch on the SQuAD datase
|
|
| 22 |
## Score
|
| 23 |
The score is **~20%**. (`{'exact_match': 10.387890255439924, 'f1': 19.81830726643602}`)
|
| 24 |
|
|
|
|
|
|
|
| 25 |
## Training code
|
| 26 |
You can find the full training code as `train.py` in this repo :-)
|
| 27 |
|
|
|
|
| 22 |
## Score
|
| 23 |
The score is **~20%**. (`{'exact_match': 10.387890255439924, 'f1': 19.81830726643602}`)
|
| 24 |
|
| 25 |
+
## Benchmark
|
| 26 |
+
The score is **~20%**, compared to the astounding 87% on BERT SQuAD
|
| 27 |
## Training code
|
| 28 |
You can find the full training code as `train.py` in this repo :-)
|
| 29 |
|