update model card README.md

README.md

@@ -14,10 +14,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # vilt-b32-mlm-mami
 
-This model is a fine-tuned version of [dandelin/vilt-b32-mlm](https://huggingface.co/dandelin/vilt-b32-mlm) on the None dataset.
+This model is a fine-tuned version of [dandelin/vilt-b32-mlm](https://huggingface.co/dandelin/vilt-b32-mlm) on the MAMI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.8414
-- F1: 0.7822
+- Loss: 0.6324
+- F1: 0.7900
 
 ## Model description
 
@@ -41,7 +41,7 @@ The following hyperparameters were used during training:
 - eval_batch_size: 16
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: constant_with_warmup
+- lr_scheduler_type: reduce_lr_on_plateau
 - lr_scheduler_warmup_ratio: 0.1
 - num_epochs: 10.0
 
@@ -49,11 +49,13 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | F1     |
 |:-------------:|:-----:|:----:|:---------------:|:------:|
-| 0.5835        | 0.96  | 200  | 0.5061          | 0.7544 |
-| 0.4219        | 1.91  | 400  | 0.4394          | 0.7948 |
-| 0.2613        | 2.87  | 600  | 0.4886          | 0.7897 |
-| 0.1399        | 3.83  | 800  | 0.6446          | 0.7892 |
-| 0.0707        | 4.78  | 1000 | 0.8414          | 0.7822 |
+| 0.6261        | 0.48  | 100  | 0.5528          | 0.7149 |
+| 0.4739        | 0.96  | 200  | 0.4688          | 0.7795 |
+| 0.3717        | 1.44  | 300  | 0.4479          | 0.7948 |
+| 0.3705        | 1.91  | 400  | 0.4385          | 0.7924 |
+| 0.253         | 2.39  | 500  | 0.5210          | 0.7956 |
+| 0.2095        | 2.87  | 600  | 0.5216          | 0.7965 |
+| 0.1371        | 3.35  | 700  | 0.6324          | 0.7900 |
 
 
 ### Framework versions