|
--- |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- gokuls/wiki_book_corpus_complete_processed_bert_dataset |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: HBERTv1_emb_compress_48_L10_H768_A12 |
|
results: |
|
- task: |
|
name: Masked Language Modeling |
|
type: fill-mask |
|
dataset: |
|
name: gokuls/wiki_book_corpus_complete_processed_bert_dataset |
|
type: gokuls/wiki_book_corpus_complete_processed_bert_dataset |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.3705453911691882 |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# HBERTv1_emb_compress_48_L10_H768_A12 |
|
|
|
This model is a fine-tuned version of [](https://huggingface.co/) on the gokuls/wiki_book_corpus_complete_processed_bert_dataset dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 4.1748 |
|
- Accuracy: 0.3705 |
|
|
|
## 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: 1e-05 |
|
- train_batch_size: 48 |
|
- eval_batch_size: 48 |
|
- seed: 10 |
|
- distributed_type: multi-GPU |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 10000 |
|
- num_epochs: 5 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:------:|:---------------:|:--------:| |
|
| 7.1074 | 0.08 | 10000 | 7.0838 | 0.0828 | |
|
| 6.6784 | 0.16 | 20000 | 6.6795 | 0.1075 | |
|
| 6.535 | 0.25 | 30000 | 6.5322 | 0.1192 | |
|
| 6.4482 | 0.33 | 40000 | 6.4390 | 0.1267 | |
|
| 6.3716 | 0.41 | 50000 | 6.3711 | 0.1324 | |
|
| 6.3233 | 0.49 | 60000 | 6.3219 | 0.1351 | |
|
| 6.2821 | 0.57 | 70000 | 6.2781 | 0.1383 | |
|
| 6.251 | 0.66 | 80000 | 6.2431 | 0.1408 | |
|
| 6.2159 | 0.74 | 90000 | 6.2111 | 0.1425 | |
|
| 6.1838 | 0.82 | 100000 | 6.1774 | 0.1444 | |
|
| 6.1338 | 0.9 | 110000 | 6.1349 | 0.1464 | |
|
| 6.1022 | 0.98 | 120000 | 6.0939 | 0.1481 | |
|
| 6.0194 | 1.07 | 130000 | 6.0080 | 0.1517 | |
|
| 5.9309 | 1.15 | 140000 | 5.9199 | 0.1642 | |
|
| 5.8593 | 1.23 | 150000 | 5.8326 | 0.1769 | |
|
| 5.7093 | 1.31 | 160000 | 5.6659 | 0.2040 | |
|
| 5.5018 | 1.39 | 170000 | 5.4433 | 0.2339 | |
|
| 5.3036 | 1.47 | 180000 | 5.2292 | 0.2576 | |
|
| 5.0629 | 1.56 | 190000 | 4.9895 | 0.2834 | |
|
| 4.8311 | 1.64 | 200000 | 4.7638 | 0.3085 | |
|
| 4.6239 | 1.72 | 210000 | 4.5799 | 0.3278 | |
|
| 4.4305 | 1.8 | 220000 | 4.3821 | 0.3471 | |
|
| 4.2209 | 1.88 | 230000 | 4.1749 | 0.3704 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.33.2 |
|
- Pytorch 1.14.0a0+410ce96 |
|
- Datasets 2.14.5 |
|
- Tokenizers 0.13.3 |
|
|