|
--- |
|
library_name: transformers |
|
license: cc-by-nc-4.0 |
|
base_model: facebook/mms-1b-all |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- fleurs |
|
metrics: |
|
- wer |
|
model-index: |
|
- name: mms-1b-all-lg-CV-Fleurs_filtered-100hrs-v2 |
|
results: |
|
- task: |
|
name: Automatic Speech Recognition |
|
type: automatic-speech-recognition |
|
dataset: |
|
name: fleurs |
|
type: fleurs |
|
config: lg_ug |
|
split: test |
|
args: lg_ug |
|
metrics: |
|
- name: Wer |
|
type: wer |
|
value: 0.4090379008746356 |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# mms-1b-all-lg-CV-Fleurs_filtered-100hrs-v2 |
|
|
|
This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the fleurs dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.2933 |
|
- Wer: 0.4090 |
|
- Cer: 0.0749 |
|
|
|
## 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: 0.001 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 2 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 2 |
|
- total_train_batch_size: 8 |
|
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
|
- lr_scheduler_type: linear |
|
- num_epochs: 70 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer | |
|
|:-------------:|:-----:|:------:|:---------------:|:------:|:------:| |
|
| 0.7083 | 1.0 | 7125 | 0.3181 | 0.4363 | 0.0781 | |
|
| 0.2153 | 2.0 | 14250 | 0.3057 | 0.4327 | 0.0775 | |
|
| 0.2092 | 3.0 | 21375 | 0.2982 | 0.4040 | 0.0738 | |
|
| 0.207 | 4.0 | 28500 | 0.3020 | 0.4057 | 0.0740 | |
|
| 0.2047 | 5.0 | 35625 | 0.3008 | 0.4136 | 0.0790 | |
|
| 0.2025 | 6.0 | 42750 | 0.3010 | 0.4156 | 0.0763 | |
|
| 0.1989 | 7.0 | 49875 | 0.3064 | 0.4101 | 0.0754 | |
|
| 0.1989 | 8.0 | 57000 | 0.2903 | 0.4086 | 0.0751 | |
|
| 0.1973 | 9.0 | 64125 | 0.2927 | 0.4 | 0.0737 | |
|
| 0.1961 | 10.0 | 71250 | 0.2882 | 0.3986 | 0.0736 | |
|
| 0.1952 | 11.0 | 78375 | 0.2895 | 0.4068 | 0.0741 | |
|
| 0.1943 | 12.0 | 85500 | 0.2950 | 0.4096 | 0.0754 | |
|
| 0.1933 | 13.0 | 92625 | 0.2945 | 0.4086 | 0.0749 | |
|
| 0.1926 | 14.0 | 99750 | 0.2933 | 0.4004 | 0.0734 | |
|
| 0.1912 | 15.0 | 106875 | 0.2925 | 0.4180 | 0.0755 | |
|
| 0.1909 | 16.0 | 114000 | 0.2949 | 0.4149 | 0.0751 | |
|
| 0.1902 | 17.0 | 121125 | 0.2888 | 0.4045 | 0.0740 | |
|
| 0.189 | 18.0 | 128250 | 0.2856 | 0.4086 | 0.0744 | |
|
| 0.1885 | 19.0 | 135375 | 0.2933 | 0.4125 | 0.0745 | |
|
| 0.187 | 20.0 | 142500 | 0.2930 | 0.4115 | 0.0746 | |
|
| 0.1877 | 21.0 | 149625 | 0.2886 | 0.4023 | 0.0737 | |
|
| 0.1867 | 22.0 | 156750 | 0.2933 | 0.4009 | 0.0730 | |
|
| 0.1863 | 23.0 | 163875 | 0.2893 | 0.4040 | 0.0738 | |
|
| 0.1846 | 24.0 | 171000 | 0.2920 | 0.4146 | 0.0753 | |
|
| 0.185 | 25.0 | 178125 | 0.2907 | 0.4017 | 0.0730 | |
|
| 0.1836 | 26.0 | 185250 | 0.2939 | 0.3992 | 0.0730 | |
|
| 0.1827 | 27.0 | 192375 | 0.2934 | 0.4144 | 0.0760 | |
|
| 0.1827 | 28.0 | 199500 | 0.2962 | 0.4038 | 0.0736 | |
|
| 0.1818 | 29.0 | 206625 | 0.2917 | 0.4063 | 0.0750 | |
|
| 0.1818 | 30.0 | 213750 | 0.2933 | 0.4090 | 0.0749 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.46.2 |
|
- Pytorch 2.1.0+cu118 |
|
- Datasets 3.1.0 |
|
- Tokenizers 0.20.3 |
|
|
