End of training

README.md

@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [fahadqazi/testts1234](https://huggingface.co/fahadqazi/testts1234) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3365
+- Loss: 0.3375
 
 ## Model description
 
@@ -35,26 +35,24 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 5e-05
-- train_batch_size: 8
-- eval_batch_size: 64
+- train_batch_size: 64
+- eval_batch_size: 100
 - seed: 42
-- gradient_accumulation_steps: 8
-- total_train_batch_size: 64
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 256
 - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
-- lr_scheduler_type: linear
+- lr_scheduler_type: cosine
 - lr_scheduler_warmup_steps: 200
-- training_steps: 5000
+- num_epochs: 3
 - mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch   | Step | Validation Loss |
-|:-------------:|:-------:|:----:|:---------------:|
-| 3.1545        | 3.8474  | 1000 | 0.3448          |
-| 3.1205        | 7.6933  | 2000 | 0.3425          |
-| 3.0913        | 11.5392 | 3000 | 0.3408          |
-| 3.0625        | 15.3852 | 4000 | 0.3379          |
-| 3.0456        | 19.2311 | 5000 | 0.3365          |
+| Training Loss | Epoch  | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| No log        | 0.7692 | 50   | 0.3405          |
+| No log        | 1.5385 | 100  | 0.3410          |
+| No log        | 2.3077 | 150  | 0.3375          |
 
 
 ### Framework versions