LovenOO
/

distilBERT_mergeddata_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: 0.2087
-- Precision: 0.9680
-- Recall: 0.9683
-- F1: 0.9680
-- Accuracy: 0.9683
 ## 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: 16
-- eval_batch_size: 16
 - 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   | 450  | 0.1592          | 0.9588    | 0.9588 | 0.9584 | 0.9589   |
-| 0.5532        | 2.0   | 900  | 0.1639          | 0.9595    | 0.9595 | 0.9588 | 0.9594   |
-| 0.1164        | 3.0   | 1350 | 0.1651          | 0.9610    | 0.9604 | 0.9598 | 0.9606   |
-| 0.0632        | 4.0   | 1800 | 0.1638          | 0.9681    | 0.9684 | 0.9680 | 0.9683   |
-| 0.0419        | 5.0   | 2250 | 0.1786          | 0.9657    | 0.9662 | 0.9657 | 0.9661   |
-| 0.0233        | 6.0   | 2700 | 0.2082          | 0.9630    | 0.9635 | 0.9630 | 0.9633   |
-| 0.012         | 7.0   | 3150 | 0.1964          | 0.9652    | 0.9656 | 0.9652 | 0.9656   |
-| 0.0091        | 8.0   | 3600 | 0.1966          | 0.9678    | 0.9682 | 0.9680 | 0.9683   |
-| 0.0077        | 9.0   | 4050 | 0.2058          | 0.9674    | 0.9678 | 0.9675 | 0.9678   |
-| 0.0027        | 10.0  | 4500 | 0.2087          | 0.9680    | 0.9683 | 0.9680 | 0.9683   |
 ### 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: 0.1986
+- Precision: 0.9664
+- Recall: 0.9668
+- F1: 0.9665
+- Accuracy: 0.9667
 ## Model description
 ### Training hyperparameters
 The following hyperparameters were used during training:
+- learning_rate: 5e-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
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
+| No log        | 1.0   | 225  | 0.1748          | 0.9549    | 0.9514 | 0.9520 | 0.9528   |
+| No log        | 2.0   | 450  | 0.1584          | 0.9567    | 0.9563 | 0.9562 | 0.9567   |
+| 0.291         | 3.0   | 675  | 0.1553          | 0.9622    | 0.9627 | 0.9622 | 0.9622   |
+| 0.291         | 4.0   | 900  | 0.1571          | 0.9647    | 0.9651 | 0.9646 | 0.965    |
+| 0.0501        | 5.0   | 1125 | 0.1747          | 0.9667    | 0.9671 | 0.9666 | 0.9667   |
+| 0.0501        | 6.0   | 1350 | 0.1887          | 0.9650    | 0.9658 | 0.9653 | 0.9656   |
+| 0.0111        | 7.0   | 1575 | 0.1862          | 0.9668    | 0.9666 | 0.9665 | 0.9667   |
+| 0.0111        | 8.0   | 1800 | 0.1985          | 0.9647    | 0.9649 | 0.9647 | 0.965    |
+| 0.0044        | 9.0   | 2025 | 0.1954          | 0.9658    | 0.9662 | 0.9659 | 0.9661   |
+| 0.0044        | 10.0  | 2250 | 0.1986          | 0.9664    | 0.9668 | 0.9665 | 0.9667   |
 ### Framework versions