LovenOO
/

distilBERT_with_preprocessing_grid_search

@@ -20,11 +20,11 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7125
-- Precision: 0.8239
-- Recall: 0.8327
-- F1: 0.8269
-- Accuracy: 0.8581
 ## Model description
@@ -43,9 +43,9 @@ More information needed
 ### Training hyperparameters
 The following hyperparameters were used during training:
-- learning_rate: 2e-05
-- train_batch_size: 32
-- eval_batch_size: 32
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
@@ -55,16 +55,16 @@ The following hyperparameters were used during training:
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| No log        | 1.0   | 255  | 0.6963          | 0.7398    | 0.7965 | 0.7599 | 0.8292   |
-| 0.934         | 2.0   | 510  | 0.5516          | 0.7593    | 0.8244 | 0.7846 | 0.8390   |
-| 0.934         | 3.0   | 765  | 0.5481          | 0.7913    | 0.8211 | 0.8040 | 0.8513   |
-| 0.3454        | 4.0   | 1020 | 0.5285          | 0.8089    | 0.8388 | 0.8216 | 0.8616   |
-| 0.3454        | 5.0   | 1275 | 0.6033          | 0.8223    | 0.8361 | 0.8279 | 0.8635   |
-| 0.1887        | 6.0   | 1530 | 0.6543          | 0.8231    | 0.8291 | 0.8250 | 0.8621   |
-| 0.1887        | 7.0   | 1785 | 0.6787          | 0.8321    | 0.8373 | 0.8331 | 0.8655   |
-| 0.1329        | 8.0   | 2040 | 0.6760          | 0.8144    | 0.8348 | 0.8225 | 0.8571   |
-| 0.1329        | 9.0   | 2295 | 0.7171          | 0.8279    | 0.8348 | 0.8300 | 0.8616   |
-| 0.0948        | 10.0  | 2550 | 0.7125          | 0.8239    | 0.8327 | 0.8269 | 0.8581   |
 ### Framework versions

 This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.8590
+- Precision: 0.8444
+- Recall: 0.8474
+- F1: 0.8454
+- Accuracy: 0.8709
 ## Model description
 ### Training hyperparameters
 The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 16
+- eval_batch_size: 16
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
+| 0.8901        | 1.0   | 510  | 0.5727          | 0.7730    | 0.8217 | 0.7887 | 0.8439   |
+| 0.445         | 2.0   | 1020 | 0.5276          | 0.7930    | 0.8453 | 0.8123 | 0.8444   |
+| 0.2825        | 3.0   | 1530 | 0.7059          | 0.8374    | 0.8205 | 0.8256 | 0.8606   |
+| 0.2037        | 4.0   | 2040 | 0.7658          | 0.8562    | 0.8265 | 0.8399 | 0.8660   |
+| 0.1618        | 5.0   | 2550 | 0.7571          | 0.8332    | 0.8438 | 0.8377 | 0.8640   |
+| 0.1141        | 6.0   | 3060 | 0.8227          | 0.8499    | 0.8409 | 0.8444 | 0.8694   |
+| 0.0934        | 7.0   | 3570 | 0.7924          | 0.8377    | 0.8415 | 0.8378 | 0.8665   |
+| 0.0881        | 8.0   | 4080 | 0.8132          | 0.8365    | 0.8434 | 0.8387 | 0.8699   |
+| 0.065         | 9.0   | 4590 | 0.8545          | 0.8402    | 0.8430 | 0.8403 | 0.8670   |
+| 0.0562        | 10.0  | 5100 | 0.8590          | 0.8444    | 0.8474 | 0.8454 | 0.8709   |
 ### Framework versions