update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: whisper-base-full-data-language-v2-20ep
|
| 7 |
+
results: []
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 11 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 12 |
+
|
| 13 |
+
# whisper-base-full-data-language-v2-20ep
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the None dataset.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 0.1957
|
| 18 |
+
|
| 19 |
+
## Model description
|
| 20 |
+
|
| 21 |
+
More information needed
|
| 22 |
+
|
| 23 |
+
## Intended uses & limitations
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Training and evaluation data
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training procedure
|
| 32 |
+
|
| 33 |
+
### Training hyperparameters
|
| 34 |
+
|
| 35 |
+
The following hyperparameters were used during training:
|
| 36 |
+
- learning_rate: 0.00015
|
| 37 |
+
- train_batch_size: 16
|
| 38 |
+
- eval_batch_size: 8
|
| 39 |
+
- seed: 42
|
| 40 |
+
- distributed_type: tpu
|
| 41 |
+
- num_devices: 8
|
| 42 |
+
- gradient_accumulation_steps: 2
|
| 43 |
+
- total_train_batch_size: 256
|
| 44 |
+
- total_eval_batch_size: 64
|
| 45 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
+
- lr_scheduler_type: linear
|
| 47 |
+
- lr_scheduler_warmup_steps: 5000
|
| 48 |
+
- training_steps: 63840
|
| 49 |
+
|
| 50 |
+
### Training results
|
| 51 |
+
|
| 52 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 53 |
+
|:-------------:|:-----:|:-----:|:---------------:|
|
| 54 |
+
| 0.3116 | 1.57 | 5000 | 0.5301 |
|
| 55 |
+
| 0.2104 | 3.13 | 10000 | 0.4066 |
|
| 56 |
+
| 0.1729 | 4.7 | 15000 | 0.3555 |
|
| 57 |
+
| 0.1472 | 6.27 | 20000 | 0.3208 |
|
| 58 |
+
| 0.128 | 7.83 | 25000 | 0.2923 |
|
| 59 |
+
| 0.1065 | 9.4 | 30000 | 0.2719 |
|
| 60 |
+
| 0.0995 | 10.97 | 35000 | 0.2516 |
|
| 61 |
+
| 0.0812 | 12.53 | 40000 | 0.2368 |
|
| 62 |
+
| 0.066 | 14.1 | 45000 | 0.2230 |
|
| 63 |
+
| 0.0574 | 15.67 | 50000 | 0.2119 |
|
| 64 |
+
| 0.0463 | 17.23 | 55000 | 0.2028 |
|
| 65 |
+
| 0.04 | 18.8 | 60000 | 0.1957 |
|
| 66 |
+
|
| 67 |
+
|
| 68 |
+
### Framework versions
|
| 69 |
+
|
| 70 |
+
- Transformers 4.28.1
|
| 71 |
+
- Pytorch 2.1.0a0+gitcc01568
|
| 72 |
+
- Datasets 2.13.1
|
| 73 |
+
- Tokenizers 0.13.3
|