LovenOO
/

BERT_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 [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9072
-- Precision: 0.8332
-- Recall: 0.8192
-- F1: 0.8259
-- Accuracy: 0.8660
 ## 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 |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| 1.0741        | 1.0   | 510  | 0.6328          | 0.7761    | 0.8093 | 0.7820 | 0.8292   |
-| 0.465         | 2.0   | 1020 | 0.5751          | 0.8326    | 0.8237 | 0.8265 | 0.8625   |
-| 0.2979        | 3.0   | 1530 | 0.5442          | 0.8285    | 0.8482 | 0.8370 | 0.8719   |
-| 0.2312        | 4.0   | 2040 | 0.6811          | 0.8434    | 0.8298 | 0.8350 | 0.8665   |
-| 0.1609        | 5.0   | 2550 | 0.6873          | 0.8216    | 0.8338 | 0.8271 | 0.8635   |
-| 0.14          | 6.0   | 3060 | 0.8476          | 0.8386    | 0.8175 | 0.8265 | 0.8640   |
-| 0.1135        | 7.0   | 3570 | 0.8456          | 0.8302    | 0.8202 | 0.8249 | 0.8630   |
-| 0.0973        | 8.0   | 4080 | 0.8595          | 0.8307    | 0.8186 | 0.8243 | 0.8625   |
-| 0.0758        | 9.0   | 4590 | 0.8828          | 0.8306    | 0.8201 | 0.8251 | 0.8655   |
-| 0.0669        | 10.0  | 5100 | 0.9072          | 0.8332    | 0.8192 | 0.8259 | 0.8660   |
 ### Framework versions

 This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the None dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.8911
+- Precision: 0.8371
+- Recall: 0.8239
+- F1: 0.8296
+- Accuracy: 0.8665
 ## 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   | 255  | 0.6320          | 0.7746    | 0.8197 | 0.7918 | 0.8360   |
+| 0.7073        | 2.0   | 510  | 0.6156          | 0.7967    | 0.8232 | 0.8055 | 0.8473   |
+| 0.7073        | 3.0   | 765  | 0.6028          | 0.8104    | 0.8381 | 0.8201 | 0.8552   |
+| 0.2389        | 4.0   | 1020 | 0.6896          | 0.8296    | 0.8296 | 0.8290 | 0.8655   |
+| 0.2389        | 5.0   | 1275 | 0.7462          | 0.8279    | 0.8353 | 0.8310 | 0.8694   |
+| 0.1264        | 6.0   | 1530 | 0.9275          | 0.8488    | 0.8112 | 0.8271 | 0.8684   |
+| 0.1264        | 7.0   | 1785 | 0.8244          | 0.8393    | 0.8313 | 0.8347 | 0.8729   |
+| 0.0851        | 8.0   | 2040 | 0.8776          | 0.8281    | 0.8226 | 0.8249 | 0.8655   |
+| 0.0851        | 9.0   | 2295 | 0.8838          | 0.8440    | 0.8278 | 0.8346 | 0.8675   |
+| 0.0546        | 10.0  | 2550 | 0.8911          | 0.8371    | 0.8239 | 0.8296 | 0.8665   |
 ### Framework versions