Edit model card
YAML Metadata Error: "language[0]" must only contain lowercase characters
YAML Metadata Error: "language[0]" with value "en-US" is not valid. It must be an ISO 639-1, 639-2 or 639-3 code (two/three letters), or a special value like "code", "multilingual". If you want to use BCP-47 identifiers, you can specify them in language_bcp47.

xtreme_s_xlsr_300m_mt5-small_minds14.en-US

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the GOOGLE/XTREME_S - MINDS14.EN-US dataset. It achieves the following results on the evaluation set:

  • Loss: 4.7321
  • F1: 0.0154
  • Accuracy: 0.0638

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: 64
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 100
  • num_epochs: 50.0
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss F1 Accuracy
2.6067 3.95 20 2.6501 0.0112 0.0851
2.5614 7.95 40 2.8018 0.0133 0.0603
2.2836 11.95 60 3.0786 0.0084 0.0603
1.9597 15.95 80 3.2288 0.0126 0.0638
1.5566 19.95 100 3.6934 0.0178 0.0567
1.3168 23.95 120 3.9135 0.0150 0.0638
1.0598 27.95 140 4.2618 0.0084 0.0603
0.5721 31.95 160 3.7973 0.0354 0.0780
0.4402 35.95 180 4.6233 0.0179 0.0638
0.6113 39.95 200 4.6149 0.0208 0.0674
0.3938 43.95 220 4.7886 0.0159 0.0638
0.2473 47.95 240 4.7321 0.0154 0.0638

Framework versions

  • Transformers 4.18.0
  • Pytorch 1.11.0+cu113
  • Datasets 2.1.0
  • Tokenizers 0.12.1
Downloads last month
2
Unable to determine this model’s pipeline type. Check the docs .