|
--- |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: distilbert-base-uncased-continued_training-medqa |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# distilbert-base-uncased-continued_training-medqa |
|
|
|
This model is a fine-tuned version of [Shaier/distilbert-base-uncased-continued_training-medqa](https://huggingface.co/Shaier/distilbert-base-uncased-continued_training-medqa) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.4063 |
|
|
|
## Model description |
|
|
|
More information needed |
|
|
|
## Intended uses & limitations |
|
|
|
More information needed |
|
|
|
## Training and evaluation data |
|
|
|
More information needed |
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 2e-05 |
|
- train_batch_size: 64 |
|
- eval_batch_size: 64 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 8 |
|
- total_train_batch_size: 512 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 100 |
|
- num_epochs: 50 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
| No log | 1.0 | 333 | 0.4659 | |
|
| No log | 2.0 | 666 | 0.4547 | |
|
| No log | 3.0 | 999 | 0.3882 | |
|
| No log | 4.0 | 1332 | 0.4310 | |
|
| No log | 5.0 | 1665 | 0.4194 | |
|
| No log | 6.0 | 1998 | 0.5209 | |
|
| No log | 7.0 | 2331 | 0.4812 | |
|
| 0.4829 | 8.0 | 2664 | 0.5321 | |
|
| 0.4829 | 9.0 | 2997 | 0.3646 | |
|
| 0.4829 | 10.0 | 3330 | 0.4339 | |
|
| 0.4829 | 11.0 | 3663 | 0.5188 | |
|
| 0.4829 | 12.0 | 3996 | 0.4148 | |
|
| 0.4829 | 13.0 | 4329 | 0.4615 | |
|
| 0.4829 | 14.0 | 4662 | 0.3825 | |
|
| 0.4829 | 15.0 | 4995 | 0.4617 | |
|
| 0.4773 | 16.0 | 5328 | 0.3400 | |
|
| 0.4773 | 17.0 | 5661 | 0.4740 | |
|
| 0.4773 | 18.0 | 5994 | 0.5057 | |
|
| 0.4773 | 19.0 | 6327 | 0.5477 | |
|
| 0.4773 | 20.0 | 6660 | 0.4426 | |
|
| 0.4773 | 21.0 | 6993 | 0.3574 | |
|
| 0.4773 | 22.0 | 7326 | 0.4031 | |
|
| 0.4773 | 23.0 | 7659 | 0.4491 | |
|
| 0.4715 | 24.0 | 7992 | 0.4340 | |
|
| 0.4715 | 25.0 | 8325 | 0.4602 | |
|
| 0.4715 | 26.0 | 8658 | 0.4659 | |
|
| 0.4715 | 27.0 | 8991 | 0.4321 | |
|
| 0.4715 | 28.0 | 9324 | 0.4335 | |
|
| 0.4715 | 29.0 | 9657 | 0.4458 | |
|
| 0.4715 | 30.0 | 9990 | 0.4285 | |
|
| 0.4715 | 31.0 | 10323 | 0.5002 | |
|
| 0.4671 | 32.0 | 10656 | 0.4706 | |
|
| 0.4671 | 33.0 | 10989 | 0.5368 | |
|
| 0.4671 | 34.0 | 11322 | 0.4028 | |
|
| 0.4671 | 35.0 | 11655 | 0.5171 | |
|
| 0.4671 | 36.0 | 11988 | 0.4506 | |
|
| 0.4671 | 37.0 | 12321 | 0.4163 | |
|
| 0.4671 | 38.0 | 12654 | 0.4905 | |
|
| 0.4671 | 39.0 | 12987 | 0.5168 | |
|
| 0.4646 | 40.0 | 13320 | 0.4412 | |
|
| 0.4646 | 41.0 | 13653 | 0.4773 | |
|
| 0.4646 | 42.0 | 13986 | 0.4835 | |
|
| 0.4646 | 43.0 | 14319 | 0.4716 | |
|
| 0.4646 | 44.0 | 14652 | 0.4431 | |
|
| 0.4646 | 45.0 | 14985 | 0.4187 | |
|
| 0.4646 | 46.0 | 15318 | 0.3389 | |
|
| 0.4646 | 47.0 | 15651 | 0.4699 | |
|
| 0.4628 | 48.0 | 15984 | 0.4880 | |
|
| 0.4628 | 49.0 | 16317 | 0.5058 | |
|
| 0.4628 | 50.0 | 16650 | 0.4275 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.18.0 |
|
- Pytorch 1.11.0 |
|
- Datasets 2.3.2 |
|
- Tokenizers 0.11.0 |
|
|
