You can deploy this model and start earning money today!

tonykorol/
t5_phoneme_decoder

Name: t5_phoneme_decoder
Availability: InStock

generated_from_trainer

google-t5/t5-small

finetune

apache-2.0

text-generation-inference

endpoints_compatible

Views

No views yet

t5_phoneme_decoder

This model is a fine-tuned version of t5-small on the Neural Prosthesis Phoneme-Text dataset. It achieves the following results on the evaluation set:

Loss: 0.0892

Model description

This model accepts phoneme text sequences and decodes them into full sentences. In this case, phoneme sequences are generated by decoding neural spike data with a GRU based decoder. This model allows us to see the phonemes more naturally as English language.

Intended uses & limitations

This model was trained for a course project for ECE243 at UCLA. This is not intended for any other uses, besides for this project/future personal experimentation

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.0002
train_batch_size: 12
eval_batch_size: 8
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: polynomial
lr_scheduler_warmup_ratio: 0.1
num_epochs: 3
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.1303	1.0	726	0.1084
0.1084	2.0	1452	0.0939
0.0998	3.0	2178	0.0892

Framework versions

Transformers 4.57.2
Pytorch 2.9.0+cu126
Datasets 4.0.0
Tokenizers 0.22.1

tonykorol/t5_phoneme_decoder