update model card README.md
Browse files
README.md
CHANGED
|
@@ -3,8 +3,6 @@ license: apache-2.0
|
|
| 3 |
base_model: bert-base-uncased
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
-
metrics:
|
| 7 |
-
- accuracy
|
| 8 |
model-index:
|
| 9 |
- name: bert_multiple_choice
|
| 10 |
results: []
|
|
@@ -16,9 +14,6 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
# bert_multiple_choice
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the None dataset.
|
| 19 |
-
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.1972
|
| 21 |
-
- Accuracy: 0.96
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
@@ -43,43 +38,10 @@ The following hyperparameters were used during training:
|
|
| 43 |
- seed: 42
|
| 44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
- lr_scheduler_type: linear
|
| 46 |
-
- num_epochs:
|
| 47 |
|
| 48 |
### Training results
|
| 49 |
|
| 50 |
-
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 51 |
-
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
|
| 52 |
-
| 1.4939 | 1.0 | 1725 | 1.3275 | 0.43 |
|
| 53 |
-
| 1.3289 | 2.0 | 3450 | 1.1917 | 0.54 |
|
| 54 |
-
| 1.2169 | 3.0 | 5175 | 1.0759 | 0.625 |
|
| 55 |
-
| 1.1062 | 4.0 | 6900 | 0.9664 | 0.64 |
|
| 56 |
-
| 0.9912 | 5.0 | 8625 | 0.8481 | 0.685 |
|
| 57 |
-
| 0.8702 | 6.0 | 10350 | 0.7332 | 0.75 |
|
| 58 |
-
| 0.7478 | 7.0 | 12075 | 0.6544 | 0.79 |
|
| 59 |
-
| 0.6543 | 8.0 | 13800 | 0.5688 | 0.825 |
|
| 60 |
-
| 0.5572 | 9.0 | 15525 | 0.4836 | 0.835 |
|
| 61 |
-
| 0.4887 | 10.0 | 17250 | 0.4446 | 0.86 |
|
| 62 |
-
| 0.4228 | 11.0 | 18975 | 0.3740 | 0.9 |
|
| 63 |
-
| 0.3783 | 12.0 | 20700 | 0.3263 | 0.925 |
|
| 64 |
-
| 0.3312 | 13.0 | 22425 | 0.3073 | 0.91 |
|
| 65 |
-
| 0.2953 | 14.0 | 24150 | 0.2381 | 0.935 |
|
| 66 |
-
| 0.2684 | 15.0 | 25875 | 0.2300 | 0.94 |
|
| 67 |
-
| 0.2426 | 16.0 | 27600 | 0.2146 | 0.94 |
|
| 68 |
-
| 0.2227 | 17.0 | 29325 | 0.2301 | 0.935 |
|
| 69 |
-
| 0.204 | 18.0 | 31050 | 0.1867 | 0.945 |
|
| 70 |
-
| 0.1932 | 19.0 | 32775 | 0.1930 | 0.95 |
|
| 71 |
-
| 0.1782 | 20.0 | 34500 | 0.2092 | 0.94 |
|
| 72 |
-
| 0.1624 | 21.0 | 36225 | 0.1557 | 0.965 |
|
| 73 |
-
| 0.1498 | 22.0 | 37950 | 0.1886 | 0.95 |
|
| 74 |
-
| 0.1368 | 23.0 | 39675 | 0.1571 | 0.955 |
|
| 75 |
-
| 0.1306 | 24.0 | 41400 | 0.1920 | 0.95 |
|
| 76 |
-
| 0.1191 | 25.0 | 43125 | 0.1967 | 0.95 |
|
| 77 |
-
| 0.1171 | 26.0 | 44850 | 0.1428 | 0.95 |
|
| 78 |
-
| 0.1059 | 27.0 | 46575 | 0.1940 | 0.965 |
|
| 79 |
-
| 0.1019 | 28.0 | 48300 | 0.2397 | 0.965 |
|
| 80 |
-
| 0.0955 | 29.0 | 50025 | 0.1751 | 0.97 |
|
| 81 |
-
| 0.0888 | 30.0 | 51750 | 0.1517 | 0.96 |
|
| 82 |
-
| 0.0855 | 31.0 | 53475 | 0.1972 | 0.96 |
|
| 83 |
|
| 84 |
|
| 85 |
### Framework versions
|
|
|
|
| 3 |
base_model: bert-base-uncased
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
|
|
|
|
|
|
| 6 |
model-index:
|
| 7 |
- name: bert_multiple_choice
|
| 8 |
results: []
|
|
|
|
| 14 |
# bert_multiple_choice
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the None dataset.
|
|
|
|
|
|
|
|
|
|
| 17 |
|
| 18 |
## Model description
|
| 19 |
|
|
|
|
| 38 |
- seed: 42
|
| 39 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 40 |
- lr_scheduler_type: linear
|
| 41 |
+
- num_epochs: 3
|
| 42 |
|
| 43 |
### Training results
|
| 44 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 45 |
|
| 46 |
|
| 47 |
### Framework versions
|