keonju
/

korean_disease_ner

@@ -19,11 +19,11 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [klue/bert-base](https://huggingface.co/klue/bert-base) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0855
-- Precision: 0.9424
-- Recall: 0.9475
-- F1: 0.9449
-- Accuracy: 0.9801
 ## Model description
@@ -43,8 +43,8 @@ More information needed
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 30
-- eval_batch_size: 30
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
@@ -52,23 +52,23 @@ The following hyperparameters were used during training:
 ### Training results
-| Training Loss | Epoch | Step   | Validation Loss | Precision | Recall | F1     | Accuracy |
-|:-------------:|:-----:|:------:|:---------------:|:---------:|:------:|:------:|:--------:|
-| 0.0663        | 1.0   | 15954  | 0.0599          | 0.9417    | 0.9246 | 0.9331 | 0.9763   |
-| 0.0471        | 2.0   | 31908  | 0.0514          | 0.9408    | 0.9442 | 0.9425 | 0.9795   |
-| 0.0384        | 3.0   | 47862  | 0.0511          | 0.9419    | 0.9471 | 0.9445 | 0.9802   |
-| 0.0292        | 4.0   | 63816  | 0.0558          | 0.9456    | 0.9449 | 0.9453 | 0.9804   |
-| 0.0253        | 5.0   | 79770  | 0.0572          | 0.9421    | 0.9507 | 0.9464 | 0.9807   |
-| 0.0225        | 6.0   | 95724  | 0.0649          | 0.9474    | 0.9435 | 0.9454 | 0.9805   |
-| 0.0209        | 7.0   | 111678 | 0.0695          | 0.9409    | 0.9504 | 0.9456 | 0.9805   |
-| 0.019         | 8.0   | 127632 | 0.0742          | 0.9431    | 0.9469 | 0.9450 | 0.9802   |
-| 0.0178        | 9.0   | 143586 | 0.0799          | 0.9425    | 0.9477 | 0.9451 | 0.9802   |
-| 0.016         | 10.0  | 159540 | 0.0855          | 0.9424    | 0.9475 | 0.9449 | 0.9801   |
 ### Framework versions
-- Transformers 4.26.0
 - Pytorch 1.13.1+cu116
-- Datasets 2.9.0
 - Tokenizers 0.13.2

 This model is a fine-tuned version of [klue/bert-base](https://huggingface.co/klue/bert-base) on the None dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.0792
+- Precision: 0.9478
+- Recall: 0.9553
+- F1: 0.9515
+- Accuracy: 0.9879
 ## Model description
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
+- train_batch_size: 50
+- eval_batch_size: 50
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 ### Training results
+| Training Loss | Epoch | Step  | Validation Loss | Precision | Recall | F1     | Accuracy |
+|:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
+| 0.1           | 1.0   | 1850  | 0.0461          | 0.9329    | 0.9401 | 0.9365 | 0.9850   |
+| 0.0346        | 2.0   | 3700  | 0.0433          | 0.9367    | 0.9500 | 0.9433 | 0.9864   |
+| 0.0211        | 3.0   | 5550  | 0.0482          | 0.9438    | 0.9493 | 0.9465 | 0.9871   |
+| 0.013         | 4.0   | 7400  | 0.0532          | 0.9449    | 0.9501 | 0.9475 | 0.9869   |
+| 0.0091        | 5.0   | 9250  | 0.0584          | 0.9430    | 0.9549 | 0.9489 | 0.9872   |
+| 0.0063        | 6.0   | 11100 | 0.0675          | 0.9497    | 0.9503 | 0.9500 | 0.9874   |
+| 0.0044        | 7.0   | 12950 | 0.0660          | 0.9467    | 0.9543 | 0.9505 | 0.9876   |
+| 0.0032        | 8.0   | 14800 | 0.0752          | 0.9429    | 0.9563 | 0.9495 | 0.9873   |
+| 0.0025        | 9.0   | 16650 | 0.0766          | 0.9463    | 0.9561 | 0.9512 | 0.9878   |
+| 0.0019        | 10.0  | 18500 | 0.0792          | 0.9478    | 0.9553 | 0.9515 | 0.9879   |
 ### Framework versions
+- Transformers 4.26.1
 - Pytorch 1.13.1+cu116
+- Datasets 2.10.0
 - Tokenizers 0.13.2