update model card README.md

README.md

@@ -3,7 +3,7 @@ license: apache-2.0
 tags:
 - generated_from_trainer
 datasets:
-- cosmos_qa
+- race
 metrics:
 - accuracy
 model-index:
@@ -16,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # t5_base_race_cosmos_qa
 
-This model is a fine-tuned version of [t5-base](https://huggingface.co/t5-base) on the cosmos_qa dataset.
+This model is a fine-tuned version of [t5-base](https://huggingface.co/t5-base) on the race dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5038
-- Accuracy: 0.7176
+- Loss: 0.4414
+- Accuracy: 0.7424
 
 ## Model description
 
@@ -39,11 +39,9 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0001
-- train_batch_size: 2
+- train_batch_size: 8
 - eval_batch_size: 16
 - seed: 42
-- gradient_accumulation_steps: 4
-- total_train_batch_size: 8
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 100
@@ -51,16 +49,16 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Accuracy |
-|:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.4408        | 1.0   | 3157 | 0.4825          | 0.6315   |
-| 0.3166        | 2.0   | 6314 | 0.4108          | 0.7109   |
-| 0.2           | 3.0   | 9471 | 0.5038          | 0.7176   |
+| Training Loss | Epoch | Step  | Validation Loss | Accuracy |
+|:-------------:|:-----:|:-----:|:---------------:|:--------:|
+| 0.4355        | 1.0   | 10984 | 0.3910          | 0.7072   |
+| 0.3233        | 2.0   | 21968 | 0.3833          | 0.7321   |
+| 0.229         | 3.0   | 32952 | 0.4414          | 0.7424   |
 
 
 ### Framework versions
 
 - Transformers 4.11.3
-- Pytorch 1.9.1
+- Pytorch 1.9.0
 - Datasets 1.12.1
 - Tokenizers 0.10.3