End of training

README.md

@@ -3,24 +3,9 @@ tags:
 - generated_from_trainer
 datasets:
 - emotion
-metrics:
-- accuracy
 model-index:
 - name: sa_mobileBERT
-  results:
-  - task:
-      name: Text Classification
-      type: text-classification
-    dataset:
-      name: emotion
-      type: emotion
-      config: split
-      split: test
-      args: split
-    metrics:
-    - name: Accuracy
-      type: accuracy
-      value: 0.797
+  results: []
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -29,9 +14,6 @@ should probably proofread and complete it, then remove this comment. -->
 # sa_mobileBERT
 
 This model is a fine-tuned version of [](https://huggingface.co/) on the emotion dataset.
-It achieves the following results on the evaluation set:
-- Loss: 0.8327
-- Accuracy: 0.797
 
 ## Model description
 
@@ -51,59 +33,20 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 64
+- train_batch_size: 512
 - eval_batch_size: 64
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.5
 - lr_scheduler_warmup_steps: 500
-- num_epochs: 40
+- num_epochs: 1
 
 ### Training results
 
-| Training Loss | Epoch | Step  | Validation Loss | Accuracy |
-|:-------------:|:-----:|:-----:|:---------------:|:--------:|
-| No log        | 1.0   | 250   | 1.5599          | 0.3475   |
-| 1.627         | 2.0   | 500   | 1.5592          | 0.3475   |
-| 1.627         | 3.0   | 750   | 1.5544          | 0.3475   |
-| 1.4681        | 4.0   | 1000  | 1.2474          | 0.416    |
-| 1.4681        | 5.0   | 1250  | 1.2073          | 0.4455   |
-| 1.1155        | 6.0   | 1500  | 1.1868          | 0.461    |
-| 1.1155        | 7.0   | 1750  | 1.1605          | 0.4725   |
-| 1.0238        | 8.0   | 2000  | 1.1584          | 0.501    |
-| 1.0238        | 9.0   | 2250  | 1.0098          | 0.628    |
-| 0.8291        | 10.0  | 2500  | 0.9274          | 0.6835   |
-| 0.8291        | 11.0  | 2750  | 0.8888          | 0.699    |
-| 0.6307        | 12.0  | 3000  | 0.8986          | 0.7165   |
-| 0.6307        | 13.0  | 3250  | 0.8386          | 0.7295   |
-| 0.5668        | 14.0  | 3500  | 0.8552          | 0.7405   |
-| 0.5668        | 15.0  | 3750  | 0.8898          | 0.742    |
-| 0.5076        | 16.0  | 4000  | 0.8040          | 0.754    |
-| 0.5076        | 17.0  | 4250  | 0.7774          | 0.7715   |
-| 0.4339        | 18.0  | 4500  | 0.7777          | 0.79     |
-| 0.4339        | 19.0  | 4750  | 0.7534          | 0.781    |
-| 0.3963        | 20.0  | 5000  | 0.7293          | 0.7895   |
-| 0.3963        | 21.0  | 5250  | 0.7837          | 0.7955   |
-| 0.3704        | 22.0  | 5500  | 0.7520          | 0.8025   |
-| 0.3704        | 23.0  | 5750  | 0.7604          | 0.7945   |
-| 0.343         | 24.0  | 6000  | 0.7494          | 0.801    |
-| 0.343         | 25.0  | 6250  | 0.7794          | 0.79     |
-| 0.3175        | 26.0  | 6500  | 0.7747          | 0.8065   |
-| 0.3175        | 27.0  | 6750  | 0.7595          | 0.7965   |
-| 0.2975        | 28.0  | 7000  | 0.7423          | 0.8055   |
-| 0.2975        | 29.0  | 7250  | 0.7685          | 0.8      |
-| 0.2833        | 30.0  | 7500  | 0.7858          | 0.805    |
-| 0.2833        | 31.0  | 7750  | 0.7899          | 0.7925   |
-| 0.2743        | 32.0  | 8000  | 0.8048          | 0.7885   |
-| 0.2743        | 33.0  | 8250  | 0.7856          | 0.8075   |
-| 0.2581        | 34.0  | 8500  | 0.8239          | 0.801    |
-| 0.2581        | 35.0  | 8750  | 0.8195          | 0.802    |
-| 0.2502        | 36.0  | 9000  | 0.8283          | 0.8035   |
-| 0.2502        | 37.0  | 9250  | 0.8263          | 0.7995   |
-| 0.2438        | 38.0  | 9500  | 0.8356          | 0.797    |
-| 0.2438        | 39.0  | 9750  | 0.8265          | 0.7995   |
-| 0.238         | 40.0  | 10000 | 0.8327          | 0.797    |
+| Training Loss | Epoch | Step | Validation Loss | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|
+| No log        | 1.0   | 32   | 1.7881          | 0.3475   |
 
 
 ### Framework versions