muhammadravi251001
/

fine-tuned-DatasetQAS-IDK-MRC-with-xlm-roberta-large-with-ITTL-with-freeze-LR-1e-05

Question Answering

Generated from Trainer

Inference Endpoints

Model card Files Files and versions Metrics Training metrics Community

muhammadravi251001 commited on May 9, 2023

Commit

f63e1e8

·

1 Parent(s): 4e12b60

update model card README.md

Files changed (1) hide show

README.md +16 -16

README.md CHANGED Viewed

@@ -16,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [xlm-roberta-large](https://huggingface.co/xlm-roberta-large) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.8949
-- Exact Match: 74.0838
-- F1: 80.5854
 ## Model description
@@ -51,19 +51,19 @@ The following hyperparameters were used during training:
 | Training Loss | Epoch | Step | Validation Loss | Exact Match | F1      |
 |:-------------:|:-----:|:----:|:---------------:|:-----------:|:-------:|
-| 6.3129        | 0.49  | 36   | 2.2604          | 50.0        | 50.0    |
-| 3.4672        | 0.98  | 72   | 1.7812          | 50.1309     | 52.2946 |
-| 1.8387        | 1.48  | 108  | 1.2330          | 57.9843     | 64.8075 |
-| 1.8387        | 1.97  | 144  | 1.0021          | 67.5393     | 74.9046 |
-| 1.2584        | 2.46  | 180  | 0.9278          | 65.7068     | 73.3457 |
-| 0.9797        | 2.95  | 216  | 0.8511          | 67.9319     | 74.9566 |
-| 0.8097        | 3.45  | 252  | 0.8526          | 70.6806     | 77.5681 |
-| 0.8097        | 3.94  | 288  | 0.8839          | 68.1937     | 75.4475 |
-| 0.7202        | 4.44  | 324  | 0.7894          | 74.0838     | 80.4578 |
-| 0.621         | 4.92  | 360  | 0.7881          | 75.5236     | 81.6748 |
-| 0.621         | 5.42  | 396  | 0.8178          | 74.8691     | 81.1730 |
-| 0.568         | 5.91  | 432  | 0.8236          | 74.8691     | 80.9766 |
-| 0.528         | 6.41  | 468  | 0.8949          | 74.0838     | 80.5854 |
 ### Framework versions

 This model is a fine-tuned version of [xlm-roberta-large](https://huggingface.co/xlm-roberta-large) on the None dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.8236
+- Exact Match: 75.9162
+- F1: 81.7215
 ## Model description
 | Training Loss | Epoch | Step | Validation Loss | Exact Match | F1      |
 |:-------------:|:-----:|:----:|:---------------:|:-----------:|:-------:|
+| 6.4323        | 0.49  | 36   | 2.3534          | 50.0        | 50.0    |
+| 3.6036        | 0.98  | 72   | 1.7984          | 47.5131     | 48.0031 |
+| 1.8711        | 1.48  | 108  | 1.1297          | 59.1623     | 67.4774 |
+| 1.8711        | 1.97  | 144  | 1.0215          | 62.9581     | 70.7402 |
+| 1.1663        | 2.46  | 180  | 0.8500          | 69.1099     | 76.2124 |
+| 0.9042        | 2.95  | 216  | 0.8389          | 68.8482     | 76.0043 |
+| 0.7502        | 3.45  | 252  | 0.8404          | 70.9424     | 78.2197 |
+| 0.7502        | 3.94  | 288  | 0.9341          | 68.5864     | 75.3916 |
+| 0.6715        | 4.44  | 324  | 0.7647          | 74.2147     | 80.2681 |
+| 0.576         | 4.92  | 360  | 0.7881          | 75.6545     | 81.9120 |
+| 0.576         | 5.42  | 396  | 0.8022          | 74.7382     | 80.7782 |
+| 0.5213        | 5.91  | 432  | 0.8218          | 74.2147     | 80.6803 |
+| 0.4811        | 6.41  | 468  | 0.8236          | 75.9162     | 81.7215 |
 ### Framework versions