File size: 1,984 Bytes
4550dd2 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: muril-base-cased-indic_glue
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. -->
# muril-base-cased-indic_glue
This model is a fine-tuned version of [google/muril-base-cased](https://huggingface.co/google/muril-base-cased) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6013
- Precision: 0.8504
- Recall: 0.8595
- F1: 0.8549
- Accuracy: 0.9386
## 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: 5e-05
- train_batch_size: 32
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
### Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
| 1.6533 | 0.31 | 200 | 1.3771 | 0.7055 | 0.7392 | 0.7220 | 0.8935 |
| 1.2034 | 0.62 | 400 | 1.0141 | 0.8253 | 0.8181 | 0.8217 | 0.9245 |
| 0.9116 | 0.94 | 600 | 0.7996 | 0.8442 | 0.8286 | 0.8363 | 0.9328 |
| 0.7435 | 1.25 | 800 | 0.6917 | 0.8291 | 0.8565 | 0.8426 | 0.9325 |
| 0.645 | 1.56 | 1000 | 0.6281 | 0.8494 | 0.8530 | 0.8512 | 0.9370 |
| 0.6003 | 1.88 | 1200 | 0.6013 | 0.8504 | 0.8595 | 0.8549 | 0.9386 |
### Framework versions
- Transformers 4.29.2
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
|