Sjdan/f03_main

1---2license: apache-2.03tags:4- generated_from_trainer5model-index:6- name: f03_main7  results: []8---9 10<!-- This model card has been generated automatically according to the information the Trainer had access to. You11should probably proofread and complete it, then remove this comment. -->12 13# f03_main14 15This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the None dataset.16It achieves the following results on the evaluation set:17- Loss: 0.076818- Wer: 1.059719 20## Model description21 22More information needed23 24## Intended uses & limitations25 26More information needed27 28## Training and evaluation data29 30More information needed31 32## Training procedure33 34### Training hyperparameters35 36The following hyperparameters were used during training:37- learning_rate: 0.000138- train_batch_size: 839- eval_batch_size: 840- seed: 4241- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-0842- lr_scheduler_type: linear43- lr_scheduler_warmup_steps: 100044- num_epochs: 745- mixed_precision_training: Native AMP46 47### Training results48 49| Training Loss | Epoch | Step | Validation Loss | Wer    |50|:-------------:|:-----:|:----:|:---------------:|:------:|51| 9.4936        | 0.93  | 500  | 4.0912          | 1.0    |52| 2.6311        | 1.87  | 1000 | 1.7825          | 2.4286 |53| 1.0606        | 2.8   | 1500 | 0.3506          | 1.6247 |54| 0.3205        | 3.74  | 2000 | 0.1833          | 1.1471 |55| 0.1583        | 4.67  | 2500 | 0.1143          | 1.3198 |56| 0.0759        | 5.61  | 3000 | 0.0881          | 1.3561 |57| 0.0621        | 6.54  | 3500 | 0.0768          | 1.0597 |58 59 60### Framework versions61 62- Transformers 4.17.063- Pytorch 1.13.1+cu11664- Datasets 1.18.365- Tokenizers 0.13.266