Model save
Browse files
README.md
ADDED
|
@@ -0,0 +1,93 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: HKUSTAudio/Llasa-1B
|
| 3 |
+
library_name: peft
|
| 4 |
+
license: cc-by-nc-4.0
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
model-index:
|
| 8 |
+
- name: A10-Llasa-1B_220K
|
| 9 |
+
results: []
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 13 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 14 |
+
|
| 15 |
+
# A10-Llasa-1B_220K
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [HKUSTAudio/Llasa-1B](https://huggingface.co/HKUSTAudio/Llasa-1B) on an unknown dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 7.3216
|
| 20 |
+
|
| 21 |
+
## Model description
|
| 22 |
+
|
| 23 |
+
More information needed
|
| 24 |
+
|
| 25 |
+
## Intended uses & limitations
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Training and evaluation data
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training procedure
|
| 34 |
+
|
| 35 |
+
### Training hyperparameters
|
| 36 |
+
|
| 37 |
+
The following hyperparameters were used during training:
|
| 38 |
+
- learning_rate: 5e-05
|
| 39 |
+
- train_batch_size: 3
|
| 40 |
+
- eval_batch_size: 3
|
| 41 |
+
- seed: 42
|
| 42 |
+
- distributed_type: multi-GPU
|
| 43 |
+
- num_devices: 2
|
| 44 |
+
- gradient_accumulation_steps: 4
|
| 45 |
+
- total_train_batch_size: 24
|
| 46 |
+
- total_eval_batch_size: 6
|
| 47 |
+
- optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-08
|
| 48 |
+
- lr_scheduler_type: cosine
|
| 49 |
+
- lr_scheduler_warmup_ratio: 0.03
|
| 50 |
+
- num_epochs: 1
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
+
|:-------------:|:------:|:-----:|:---------------:|
|
| 56 |
+
| No log | 0 | 0 | 7.8117 |
|
| 57 |
+
| 7.3694 | 0.0089 | 1000 | 7.5275 |
|
| 58 |
+
| 7.3694 | 0.0089 | 1000 | 7.5158 |
|
| 59 |
+
| 7.0835 | 0.0715 | 2000 | 7.4015 |
|
| 60 |
+
| 7.0793 | 0.1072 | 3000 | 7.3785 |
|
| 61 |
+
| 7.0461 | 0.1430 | 4000 | 7.3690 |
|
| 62 |
+
| 7.0339 | 0.1787 | 5000 | 7.3580 |
|
| 63 |
+
| 6.9696 | 0.2144 | 6000 | 7.3513 |
|
| 64 |
+
| 7.033 | 0.2502 | 7000 | 7.3444 |
|
| 65 |
+
| 6.9768 | 0.2859 | 8000 | 7.3387 |
|
| 66 |
+
| 7.1218 | 0.3216 | 9000 | 7.3378 |
|
| 67 |
+
| 7.041 | 0.3574 | 10000 | 7.3314 |
|
| 68 |
+
| 6.9799 | 0.3931 | 11000 | 7.3350 |
|
| 69 |
+
| 7.0261 | 0.4289 | 12000 | 7.3297 |
|
| 70 |
+
| 6.888 | 0.4646 | 13000 | 7.3288 |
|
| 71 |
+
| 6.9483 | 0.5003 | 14000 | 7.3285 |
|
| 72 |
+
| 6.989 | 0.5361 | 15000 | 7.3269 |
|
| 73 |
+
| 7.0167 | 0.5718 | 16000 | 7.3243 |
|
| 74 |
+
| 6.9611 | 0.6076 | 17000 | 7.3273 |
|
| 75 |
+
| 6.9077 | 0.6433 | 18000 | 7.3256 |
|
| 76 |
+
| 7.0845 | 0.6790 | 19000 | 7.3235 |
|
| 77 |
+
| 6.8593 | 0.7148 | 20000 | 7.3207 |
|
| 78 |
+
| 6.8621 | 0.7505 | 21000 | 7.3216 |
|
| 79 |
+
| 7.1707 | 0.7863 | 22000 | 7.3225 |
|
| 80 |
+
| 6.9153 | 0.8220 | 23000 | 7.3209 |
|
| 81 |
+
| 6.9139 | 0.8577 | 24000 | 7.3217 |
|
| 82 |
+
| 6.9 | 0.8935 | 25000 | 7.3214 |
|
| 83 |
+
| 6.7397 | 0.9292 | 26000 | 7.3207 |
|
| 84 |
+
| 6.9967 | 0.9649 | 27000 | 7.3216 |
|
| 85 |
+
|
| 86 |
+
|
| 87 |
+
### Framework versions
|
| 88 |
+
|
| 89 |
+
- PEFT 0.15.2
|
| 90 |
+
- Transformers 4.45.2
|
| 91 |
+
- Pytorch 2.4.1+cu121
|
| 92 |
+
- Datasets 3.0.1
|
| 93 |
+
- Tokenizers 0.20.1
|