Downloads · 30 days
4
9% of all-time downloads
mBART-en_tags-hi-IITB-HVG
hrithikgupt/mBART-en_tags-hi-IITB-HVG
mBART-en_tags-hi-IITB-HVG is a translation model from hrithikgupt. Use it when you need text moved from one language to another. It is set up for transformers. The card lists the license as mit.
should probably proofread and complete it, then remove this comment. --
All-time downloads
47
Public
Repo size
7.4 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin2.4 GB · 99%
At a glance
- Task
- Translation
- Library
- transformers
- License
- mit
- Model type
- mbart
- Access
- Public
- Created
- Mar 24, 2023
- Updated
- Mar 24, 2023
- SHA
- b87a1c8e
Try a prompt
Model card
From the Hugging Face model README
<!-- 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. -->
mBART-en_tags-hi-IITB-HVG
This model is a fine-tuned version of hrithikgupt/mBART-en-hi-IITB on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.5400
- Bleu: 48.1119
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
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu |
|---|---|---|---|---|
| 0.7218 | 1.0 | 905 | 0.6396 | 46.5394 |
| 0.3797 | 2.0 | 1810 | 0.6241 | 48.9921 |
| 0.2444 | 3.0 | 2715 | 0.7032 | 49.2763 |
| 0.1593 | 4.0 | 3620 | 0.7818 | 48.3701 |
| 0.1088 | 5.0 | 4525 | 0.9005 | 48.0507 |
| 0.0755 | 6.0 | 5430 | 1.0067 | 48.9502 |
| 0.0562 | 7.0 | 6335 | 1.0722 | 48.3758 |
| 0.0459 | 8.0 | 7240 | 1.1511 | 47.6653 |
| 0.0386 | 9.0 | 8145 | 1.1360 | 48.0151 |
| 0.0307 | 10.0 | 9050 | 1.2381 | 48.3971 |
| 0.0245 | 11.0 | 9955 | 1.2196 | 48.5044 |
| 0.0204 | 12.0 | 10860 | 1.2590 | 48.6369 |
| 0.0183 | 13.0 | 11765 | 1.2894 | 49.1163 |
| 0.015 | 14.0 | 12670 | 1.3340 | 48.8843 |
| 0.014 | 15.0 | 13575 | 1.3383 | 49.1622 |
| 0.0108 | 16.0 | 14480 | 1.3863 | 48.4685 |
| 0.0086 | 17.0 | 15385 | 1.4247 | 47.8833 |
| 0.0085 | 18.0 | 16290 | 1.4159 | 47.7058 |
| 0.0065 | 19.0 | 17195 | 1.4502 | 48.1717 |
| 0.0062 | 20.0 | 18100 | 1.4589 | 47.7182 |
| 0.005 | 21.0 | 19005 | 1.4715 | 48.6974 |
| 0.0041 | 22.0 | 19910 | 1.4711 | 48.1568 |
| 0.0033 | 23.0 | 20815 | 1.5125 | 48.5741 |
| 0.0035 | 24.0 | 21720 | 1.4954 | 47.9083 |
| 0.0027 | 25.0 | 22625 | 1.5201 | 48.5192 |
| 0.0024 | 26.0 | 23530 | 1.5249 | 48.5003 |
| 0.0018 | 27.0 | 24435 | 1.5240 | 48.4175 |
| 0.0021 | 28.0 | 25340 | 1.5340 | 47.9497 |
| 0.0018 | 29.0 | 26245 | 1.5367 | 47.9215 |
| 0.0015 | 30.0 | 27150 | 1.5400 | 48.1119 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1
- Datasets 2.9.0
- Tokenizers 0.13.2
- Task
- Translation
- Library
- transformers
- Type
- mbart
- License
- mit
- Created
- Mar 24, 2023
- Updated
- Mar 24, 2023