|
--- |
|
license: apache-2.0 |
|
base_model: distilbert/distilbert-base-uncased |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: distilbert_base_uncased_ledgar |
|
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_ledgar |
|
|
|
This model is a fine-tuned version of [distilbert/distilbert-base-uncased](https://huggingface.co/distilbert/distilbert-base-uncased) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.6496 |
|
- Accuracy: 0.8311 |
|
- F1 Macro: 0.7116 |
|
- F1 Micro: 0.8311 |
|
|
|
## 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: 32 |
|
- eval_batch_size: 32 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- num_devices: 2 |
|
- total_train_batch_size: 64 |
|
- total_eval_batch_size: 64 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 3.0 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Micro | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:--------:| |
|
| 3.8165 | 0.11 | 100 | 3.5952 | 0.3489 | 0.0995 | 0.3489 | |
|
| 2.8293 | 0.21 | 200 | 2.6737 | 0.5385 | 0.2375 | 0.5385 | |
|
| 2.2564 | 0.32 | 300 | 2.0960 | 0.6212 | 0.3339 | 0.6212 | |
|
| 1.8259 | 0.43 | 400 | 1.7118 | 0.6792 | 0.4269 | 0.6792 | |
|
| 1.5846 | 0.53 | 500 | 1.4543 | 0.7232 | 0.4987 | 0.7232 | |
|
| 1.3927 | 0.64 | 600 | 1.2635 | 0.758 | 0.5628 | 0.758 | |
|
| 1.2065 | 0.75 | 700 | 1.1217 | 0.7719 | 0.5782 | 0.7719 | |
|
| 1.16 | 0.85 | 800 | 1.0303 | 0.7832 | 0.5984 | 0.7832 | |
|
| 1.0168 | 0.96 | 900 | 0.9443 | 0.7887 | 0.6119 | 0.7887 | |
|
| 0.9006 | 1.07 | 1000 | 0.8958 | 0.7934 | 0.6142 | 0.7934 | |
|
| 0.8956 | 1.17 | 1100 | 0.8517 | 0.8002 | 0.6294 | 0.8002 | |
|
| 0.9159 | 1.28 | 1200 | 0.8184 | 0.8033 | 0.6412 | 0.8033 | |
|
| 0.8237 | 1.39 | 1300 | 0.7814 | 0.8077 | 0.6529 | 0.8077 | |
|
| 0.7341 | 1.49 | 1400 | 0.7654 | 0.8099 | 0.6600 | 0.8099 | |
|
| 0.7475 | 1.6 | 1500 | 0.7458 | 0.8135 | 0.6650 | 0.8135 | |
|
| 0.7699 | 1.71 | 1600 | 0.7288 | 0.8183 | 0.6810 | 0.8183 | |
|
| 0.7472 | 1.81 | 1700 | 0.7125 | 0.8179 | 0.6820 | 0.8179 | |
|
| 0.689 | 1.92 | 1800 | 0.6965 | 0.8201 | 0.6822 | 0.8201 | |
|
| 0.6807 | 2.03 | 1900 | 0.6904 | 0.8192 | 0.6799 | 0.8192 | |
|
| 0.6514 | 2.13 | 2000 | 0.6836 | 0.8239 | 0.6923 | 0.8239 | |
|
| 0.6662 | 2.24 | 2100 | 0.6750 | 0.8267 | 0.7019 | 0.8267 | |
|
| 0.6247 | 2.35 | 2200 | 0.6703 | 0.8284 | 0.7028 | 0.8284 | |
|
| 0.6443 | 2.45 | 2300 | 0.6662 | 0.8265 | 0.7001 | 0.8265 | |
|
| 0.632 | 2.56 | 2400 | 0.6571 | 0.8295 | 0.7078 | 0.8295 | |
|
| 0.5922 | 2.67 | 2500 | 0.6539 | 0.8298 | 0.7084 | 0.8298 | |
|
| 0.6423 | 2.77 | 2600 | 0.6519 | 0.8311 | 0.7139 | 0.8311 | |
|
| 0.6156 | 2.88 | 2700 | 0.6500 | 0.8311 | 0.7123 | 0.8311 | |
|
| 0.6097 | 2.99 | 2800 | 0.6496 | 0.8311 | 0.7116 | 0.8311 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.39.0.dev0 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
|
|
