README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
model-index:
  - name: psst_batch_size_8_base_model
    results: []

psst_batch_size_8_base_model

This model is a fine-tuned version of facebook/wav2vec2-base-960h on the None dataset. It achieves the following results on the evaluation set:

• Loss: 3.6856

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: 0.0003
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss
13.113	3.33	100	3.6542
3.9221	6.67	200	3.6851
3.8468	10.0	300	3.7162
3.7095	13.33	400	3.6988
3.6987	16.67	500	3.6870
3.7247	20.0	600	3.6951
3.6922	23.33	700	3.6682
3.6862	26.67	800	3.6812
3.6838	30.0	900	3.6856

Framework versions

• Transformers 4.28.0
• Pytorch 2.0.1
• Datasets 2.12.0
• Tokenizers 0.13.2