agi-css
/

distilroberta-base-mrl

Text Classification

Generated from Trainer

Inference Endpoints

Model card Files Files and versions Metrics Training metrics Community

agi-css commited on May 3, 2022

Commit

4da0ba4

•

1 Parent(s): b17f772

update model card README.md

Files changed (1) hide show

README.md +15 -13

README.md CHANGED Viewed

@@ -15,11 +15,11 @@ should probably proofread and complete it, then remove this comment. -->
 # distilroberta-base-mrl
-This model is a fine-tuned version of [distilroberta-base](https://huggingface.co/distilroberta-base) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5367
-- Accuracy: 0.82
-- F1: 0.8200
 ## Model description
@@ -38,25 +38,27 @@ More information needed
 ### Training hyperparameters
 The following hyperparameters were used during training:
-- learning_rate: 6.36861104886319e-05
 - train_batch_size: 400
 - eval_batch_size: 400
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 7
 ### Training results
 | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
-| No log        | 1.0   | 48   | 0.4522          | 0.7879   | 0.7867 |
-| No log        | 2.0   | 96   | 0.4238          | 0.8075   | 0.8073 |
-| No log        | 3.0   | 144  | 0.4283          | 0.8121   | 0.8120 |
-| No log        | 4.0   | 192  | 0.4556          | 0.8129   | 0.8128 |
-| No log        | 5.0   | 240  | 0.4743          | 0.8196   | 0.8195 |
-| No log        | 6.0   | 288  | 0.5183          | 0.8175   | 0.8175 |
-| No log        | 7.0   | 336  | 0.5367          | 0.82     | 0.8200 |
 ### Framework versions

 # distilroberta-base-mrl
+This model is a fine-tuned version of [distilroberta-base](https://huggingface.co/distilroberta-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.0170
+- Accuracy: 0.9967
+- F1: 0.9967
 ## Model description
 ### Training hyperparameters
 The following hyperparameters were used during training:
+- learning_rate: 2.1821851463909416e-05
 - train_batch_size: 400
 - eval_batch_size: 400
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
+- num_epochs: 9
 ### Training results
 | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
+| No log        | 1.0   | 48   | 0.0265          | 0.9946   | 0.9946 |
+| No log        | 2.0   | 96   | 0.0180          | 0.9962   | 0.9962 |
+| No log        | 3.0   | 144  | 0.0163          | 0.9962   | 0.9962 |
+| No log        | 4.0   | 192  | 0.0194          | 0.9946   | 0.9946 |
+| No log        | 5.0   | 240  | 0.0193          | 0.9942   | 0.9942 |
+| No log        | 6.0   | 288  | 0.0172          | 0.9967   | 0.9967 |
+| No log        | 7.0   | 336  | 0.0206          | 0.9954   | 0.9954 |
+| No log        | 8.0   | 384  | 0.0183          | 0.9962   | 0.9962 |
+| No log        | 9.0   | 432  | 0.0170          | 0.9967   | 0.9967 |
 ### Framework versions