gokuls
/

bert_12_layer_model_v1_complete_training_new

Generated from Trainer

Model card Files Files and versions

Metrics Training metrics Community

gokuls commited on Jun 2, 2023

Commit

1a84e71

·

1 Parent(s): 1cfcecc

update model card README.md

Files changed (1) hide show

README.md +15 -20

README.md CHANGED Viewed

@@ -15,8 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [](https://huggingface.co/) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 7.2562
-- Accuracy: 0.0506
 ## Model description
@@ -35,9 +35,9 @@ More information needed
 ### Training hyperparameters
 The following hyperparameters were used during training:
-- learning_rate: 0.001
-- train_batch_size: 64
-- eval_batch_size: 64
 - seed: 10
 - distributed_type: multi-GPU
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
@@ -49,21 +49,16 @@ The following hyperparameters were used during training:
 | Training Loss | Epoch | Step   | Validation Loss | Accuracy |
 |:-------------:|:-----:|:------:|:---------------:|:--------:|
-| 7.2924        | 0.11  | 10000  | 7.2917          | 0.0466   |
-| 7.2793        | 0.22  | 20000  | 7.2749          | 0.0428   |
-| 7.274         | 0.33  | 30000  | 7.2669          | 0.0466   |
-| 7.2748        | 0.44  | 40000  | 7.2710          | 0.0466   |
-| 7.2698        | 0.55  | 50000  | 7.2719          | 0.0466   |
-| 7.2604        | 0.66  | 60000  | 7.2552          | 0.0506   |
-| 7.2682        | 0.76  | 70000  | 7.2715          | 0.0465   |
-| 7.2657        | 0.87  | 80000  | 7.2745          | 0.0465   |
-| 7.2714        | 0.98  | 90000  | 7.2747          | 0.0465   |
-| 7.2725        | 1.09  | 100000 | 7.2679          | 0.0465   |
-| 7.2613        | 1.2   | 110000 | 7.2554          | 0.0507   |
-| 7.2669        | 1.31  | 120000 | 7.2683          | 0.0465   |
-| 7.2648        | 1.42  | 130000 | 7.2672          | 0.0429   |
-| 7.2559        | 1.53  | 140000 | 7.2718          | 0.0465   |
-| 7.2546        | 1.64  | 150000 | 7.2562          | 0.0506   |
 ### Framework versions

 This model is a fine-tuned version of [](https://huggingface.co/) on the None dataset.
 It achieves the following results on the evaluation set:
+- Loss: 4.5161
+- Accuracy: 0.3159
 ## Model description
 ### Training hyperparameters
 The following hyperparameters were used during training:
+- learning_rate: 1e-05
+- train_batch_size: 48
+- eval_batch_size: 48
 - seed: 10
 - distributed_type: multi-GPU
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 | Training Loss | Epoch | Step   | Validation Loss | Accuracy |
 |:-------------:|:-----:|:------:|:---------------:|:--------:|
+| 6.5749        | 0.08  | 10000  | 6.5382          | 0.1269   |
+| 6.3332        | 0.16  | 20000  | 6.3097          | 0.1411   |
+| 6.2343        | 0.25  | 30000  | 6.2182          | 0.1450   |
+| 6.1807        | 0.33  | 40000  | 6.1587          | 0.1478   |
+| 6.0548        | 0.41  | 50000  | 6.0215          | 0.1529   |
+| 5.8862        | 0.49  | 60000  | 5.8339          | 0.1672   |
+| 5.7493        | 0.57  | 70000  | 5.6845          | 0.1878   |
+| 5.596         | 0.66  | 80000  | 5.5063          | 0.2078   |
+| 5.071         | 0.74  | 90000  | 4.8887          | 0.2781   |
+| 4.6869        | 0.82  | 100000 | 4.5161          | 0.3159   |
 ### Framework versions