Update README.md
Browse files
README.md
CHANGED
@@ -99,12 +99,13 @@ The following hyperparameters were used during training:
|
|
99 |
+ eval_batch_size: 16
|
100 |
+ seed: 42
|
101 |
+ optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
|
|
102 |
+ num_epochs: 5
|
103 |
|
104 |
## Accuracy
|
105 |
|
106 |
-
The
|
107 |
-
- Accuracy (exact) is the exact match
|
108 |
- Accuracy (off-by-1) is the percentage of reviews where the number of stars the model predicts differs by a maximum of 1 from the number given by the human reviewer.
|
109 |
|
110 |
| Split | Accuracy (exact) | Accuracy (off-by-1) |
|
|
|
99 |
+ eval_batch_size: 16
|
100 |
+ seed: 42
|
101 |
+ optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
102 |
+
+ lr_scheduler_type: linear
|
103 |
+ num_epochs: 5
|
104 |
|
105 |
## Accuracy
|
106 |
|
107 |
+
The fine-tuned model was evaluated on the French test set of `amazon_reviews_multi`.
|
108 |
+
- Accuracy (exact) is the exact match of the number of stars.
|
109 |
- Accuracy (off-by-1) is the percentage of reviews where the number of stars the model predicts differs by a maximum of 1 from the number given by the human reviewer.
|
110 |
|
111 |
| Split | Accuracy (exact) | Accuracy (off-by-1) |
|