File size: 2,065 Bytes
7559d9f 9c6a3db 7559d9f |
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 72 |
---
license: apache-2.0
base_model: riotu-lab/ArabianGPT-03B
tags:
- generated_from_trainer
metrics:
- bleu
- rouge
model-index:
- name: res_nw_dj_03
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. -->
# res_nw_dj_03
This model is a fine-tuned version of [riotu-lab/ArabianGPT-03B](https://huggingface.co/riotu-lab/ArabianGPT-03B) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5335
- Bleu: 0.4225
- Rouge1: 0.6608
- Rouge2: 0.4163
- Rougel: 0.6590
## 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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 20.0
### Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Rouge1 | Rouge2 | Rougel |
|:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|:------:|:------:|
| 0.8829 | 1.0 | 2679 | 0.5616 | 0.4033 | 0.6239 | 0.3696 | 0.6222 |
| 0.4337 | 2.0 | 5358 | 0.5335 | 0.4225 | 0.6608 | 0.4163 | 0.6590 |
| 0.2951 | 3.0 | 8037 | 0.5473 | 0.4309 | 0.6703 | 0.4320 | 0.6688 |
| 0.2054 | 4.0 | 10716 | 0.5758 | 0.4277 | 0.6764 | 0.4400 | 0.6748 |
| 0.1557 | 5.0 | 13395 | 0.6073 | 0.4309 | 0.6728 | 0.4398 | 0.6716 |
| 0.1308 | 6.0 | 16074 | 0.6263 | 0.4315 | 0.6767 | 0.4425 | 0.6752 |
| 0.1183 | 7.0 | 18753 | 0.6491 | 0.4280 | 0.6731 | 0.4383 | 0.6715 |
### Framework versions
- Transformers 4.45.0.dev0
- Pytorch 2.3.1+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1
|