finished 8-epoch LoRA fine-tune
Browse files- README.md +73 -0
- adapter_model.safetensors +1 -1
README.md
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
license: mit
|
| 4 |
+
base_model: openai/whisper-large-v3-turbo
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
metrics:
|
| 8 |
+
- wer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: whisper-hf-lora
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# whisper-hf-lora
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [openai/whisper-large-v3-turbo](https://huggingface.co/openai/whisper-large-v3-turbo) on an unknown dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.7572
|
| 22 |
+
- Wer: 0.2624
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 1e-05
|
| 42 |
+
- train_batch_size: 4
|
| 43 |
+
- eval_batch_size: 8
|
| 44 |
+
- seed: 42
|
| 45 |
+
- gradient_accumulation_steps: 4
|
| 46 |
+
- total_train_batch_size: 16
|
| 47 |
+
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 48 |
+
- lr_scheduler_type: constant_with_warmup
|
| 49 |
+
- lr_scheduler_warmup_steps: 500
|
| 50 |
+
- num_epochs: 8
|
| 51 |
+
- mixed_precision_training: Native AMP
|
| 52 |
+
|
| 53 |
+
### Training results
|
| 54 |
+
|
| 55 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 56 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 57 |
+
| 2.5146 | 1.0 | 92 | 2.3205 | 0.2746 |
|
| 58 |
+
| 2.4798 | 2.0 | 184 | 2.1513 | 0.2734 |
|
| 59 |
+
| 2.1105 | 3.0 | 276 | 1.7245 | 0.2647 |
|
| 60 |
+
| 1.6642 | 4.0 | 368 | 1.2785 | 0.2463 |
|
| 61 |
+
| 1.2627 | 5.0 | 460 | 1.0579 | 0.2395 |
|
| 62 |
+
| 1.076 | 6.0 | 552 | 0.9416 | 0.2771 |
|
| 63 |
+
| 1.0108 | 7.0 | 644 | 0.8642 | 0.2746 |
|
| 64 |
+
| 0.9166 | 8.0 | 736 | 0.7572 | 0.2624 |
|
| 65 |
+
|
| 66 |
+
|
| 67 |
+
### Framework versions
|
| 68 |
+
|
| 69 |
+
- PEFT 0.15.2
|
| 70 |
+
- Transformers 4.52.4
|
| 71 |
+
- Pytorch 2.6.0+cu124
|
| 72 |
+
- Datasets 3.6.0
|
| 73 |
+
- Tokenizers 0.21.1
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 26237160
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d541c951118c903f24cc100702212baf2d1589eac19acf35e92e050d6320ae2
|
| 3 |
size 26237160
|