update model card README.md

README.md

@@ -16,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.8592
-- Accuracy: 0.7194
-- F1 Macro: 0.4819
+- Loss: 0.7271
+- Accuracy: 0.7698
+- F1 Macro: 0.6713
 
 ## Model description
 
@@ -38,11 +38,11 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 32
-- eval_batch_size: 64
+- train_batch_size: 8
+- eval_batch_size: 16
 - seed: 42
 - gradient_accumulation_steps: 2
-- total_train_batch_size: 64
+- total_train_batch_size: 16
 - optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-06
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.06
@@ -52,16 +52,14 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|
-| No log        | 0.98  | 26   | 1.3259          | 0.5175   | 0.1137   |
-| No log        | 2.0   | 53   | 1.1130          | 0.6228   | 0.2479   |
-| No log        | 2.98  | 79   | 1.0243          | 0.6667   | 0.3126   |
-| 1.2194        | 4.0   | 106  | 0.9297          | 0.7018   | 0.3506   |
-| 1.2194        | 4.98  | 132  | 0.9334          | 0.7018   | 0.3593   |
-| 1.2194        | 6.0   | 159  | 0.8904          | 0.7368   | 0.5001   |
-| 1.2194        | 6.98  | 185  | 0.8714          | 0.7281   | 0.4661   |
-| 0.6526        | 8.0   | 212  | 0.8810          | 0.7368   | 0.4847   |
-| 0.6526        | 8.98  | 238  | 0.8807          | 0.7456   | 0.5552   |
-| 0.6526        | 9.81  | 260  | 0.8945          | 0.7193   | 0.5422   |
+| 1.419         | 1.0   | 105  | 1.0253          | 0.6842   | 0.3034   |
+| 1.0312        | 2.0   | 211  | 0.9262          | 0.6842   | 0.3424   |
+| 0.8072        | 3.0   | 316  | 0.8110          | 0.7018   | 0.3958   |
+| 0.5688        | 4.0   | 422  | 0.7826          | 0.7632   | 0.6019   |
+| 0.4064        | 5.0   | 527  | 0.7750          | 0.7719   | 0.6794   |
+| 0.3165        | 6.0   | 633  | 0.8077          | 0.7544   | 0.6073   |
+| 0.2172        | 7.0   | 738  | 0.9722          | 0.7544   | 0.6403   |
+| 0.1455        | 8.0   | 844  | 0.9993          | 0.7719   | 0.6642   |
 
 
 ### Framework versions