update model card README.md

README.md

@@ -13,7 +13,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [allenai/scibert_scivocab_uncased](https://huggingface.co/allenai/scibert_scivocab_uncased) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1722
+- Loss: 0.3494
 
 ## Model description
 
@@ -32,22 +32,29 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 2e-05
-- train_batch_size: 64
-- eval_batch_size: 64
+- learning_rate: 3e-05
+- train_batch_size: 32
+- eval_batch_size: 32
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 3.0
+- num_epochs: 10
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| No log | 1.0 | 30 | 0.4661 |
-| No log | 2.0 | 60 | 0.2044 |
-| No log | 3.0 | 90 | 0.1722 |
+| No log | 1.0 | 60 | 0.4161 |
+| No log | 2.0 | 120 | 0.3882 |
+| No log | 3.0 | 180 | 0.3667 |
+| No log | 4.0 | 240 | 0.3318 |
+| No log | 5.0 | 300 | 0.3162 |
+| No log | 6.0 | 360 | 0.3145 |
+| No log | 7.0 | 420 | 0.3277 |
+| No log | 8.0 | 480 | 0.3375 |
+| 0.4754 | 9.0 | 540 | 0.3431 |
+| 0.4754 | 10.0 | 600 | 0.3494 |
 
 
 ### Framework versions