|
--- |
|
license: mit |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
- precision |
|
- recall |
|
- f1 |
|
model-index: |
|
- name: indic-bert-finetuned-ours-DS |
|
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. --> |
|
|
|
# indic-bert-finetuned-ours-DS |
|
|
|
This model is a fine-tuned version of [ai4bharat/indic-bert](https://huggingface.co/ai4bharat/indic-bert) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.0941 |
|
- Accuracy: 0.275 |
|
- Precision: 0.3056 |
|
- Recall: 0.3467 |
|
- F1: 0.1803 |
|
|
|
## 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-07 |
|
- train_batch_size: 16 |
|
- eval_batch_size: 32 |
|
- seed: 43 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 25 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:| |
|
| 1.0988 | 0.99 | 99 | 1.0984 | 0.3 | 0.3611 | 0.3661 | 0.2750 | |
|
| 1.0981 | 1.98 | 198 | 1.0980 | 0.29 | 0.2713 | 0.3568 | 0.1997 | |
|
| 1.0981 | 2.97 | 297 | 1.0977 | 0.315 | 0.3029 | 0.3736 | 0.2259 | |
|
| 1.0976 | 3.96 | 396 | 1.0974 | 0.3 | 0.2816 | 0.3601 | 0.2122 | |
|
| 1.0976 | 4.95 | 495 | 1.0971 | 0.295 | 0.2780 | 0.3601 | 0.2041 | |
|
| 1.097 | 5.94 | 594 | 1.0968 | 0.29 | 0.2680 | 0.3533 | 0.2012 | |
|
| 1.0962 | 6.93 | 693 | 1.0965 | 0.3 | 0.2816 | 0.3601 | 0.2122 | |
|
| 1.0963 | 7.92 | 792 | 1.0963 | 0.29 | 0.2761 | 0.3533 | 0.2012 | |
|
| 1.0969 | 8.91 | 891 | 1.0961 | 0.3 | 0.2895 | 0.3601 | 0.2122 | |
|
| 1.0958 | 9.9 | 990 | 1.0958 | 0.3 | 0.2895 | 0.3601 | 0.2122 | |
|
| 1.0959 | 10.89 | 1089 | 1.0956 | 0.3 | 0.2983 | 0.3601 | 0.2122 | |
|
| 1.0953 | 11.88 | 1188 | 1.0954 | 0.3 | 0.2983 | 0.3601 | 0.2122 | |
|
| 1.0955 | 12.87 | 1287 | 1.0952 | 0.295 | 0.3019 | 0.3567 | 0.2067 | |
|
| 1.0948 | 13.86 | 1386 | 1.0951 | 0.295 | 0.3083 | 0.3601 | 0.2040 | |
|
| 1.095 | 14.85 | 1485 | 1.0949 | 0.29 | 0.3013 | 0.3568 | 0.1983 | |
|
| 1.0951 | 15.84 | 1584 | 1.0948 | 0.29 | 0.3013 | 0.3568 | 0.1983 | |
|
| 1.0948 | 16.83 | 1683 | 1.0946 | 0.29 | 0.3143 | 0.3568 | 0.1982 | |
|
| 1.0942 | 17.82 | 1782 | 1.0945 | 0.29 | 0.3291 | 0.3568 | 0.1982 | |
|
| 1.0949 | 18.81 | 1881 | 1.0944 | 0.28 | 0.3145 | 0.3500 | 0.1863 | |
|
| 1.095 | 19.8 | 1980 | 1.0943 | 0.275 | 0.3056 | 0.3467 | 0.1803 | |
|
| 1.0945 | 20.79 | 2079 | 1.0943 | 0.275 | 0.3056 | 0.3467 | 0.1803 | |
|
| 1.0942 | 21.78 | 2178 | 1.0942 | 0.275 | 0.3056 | 0.3467 | 0.1803 | |
|
| 1.0938 | 22.77 | 2277 | 1.0942 | 0.275 | 0.3056 | 0.3467 | 0.1803 | |
|
| 1.0953 | 23.76 | 2376 | 1.0941 | 0.275 | 0.3056 | 0.3467 | 0.1803 | |
|
| 1.0943 | 24.75 | 2475 | 1.0941 | 0.275 | 0.3056 | 0.3467 | 0.1803 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.20.1 |
|
- Pytorch 1.10.1+cu111 |
|
- Datasets 2.3.2 |
|
- Tokenizers 0.12.1 |
|
|
