T5 Small ASLG PC12 Open-Source Model - Efficient Sign Language Translation Tasks with Free Deployment

T5 Small Aslg Pc12

Developed by HamdanXI

This model is a fine-tuned version of t5-small on the aslg_pc12 dataset, primarily used for sign language translation tasks.

Machine Translation

Transformers

Open Source License:Apache-2.0 #Sign language translation #Low-resource optimization #Sequence generation

Downloads 22

Release Time : 10/17/2023

Model Overview

A translation model based on the T5-small architecture, specifically optimized for the ASLG-PC12 sign language dataset, capable of translating sign language text into natural language.

Model Features

Efficient translation

Achieves a BLEU score of 73.84 on the ASLG-PC12 dataset, demonstrating excellent performance

Lightweight model

Based on T5-small architecture, suitable for deployment in resource-limited environments

Multi-metric evaluation

Supports various evaluation metrics including BERTScore, BLEU, COMET, and ROUGE

Model Capabilities

Sign language text translation

Natural language generation

Use Cases

Accessibility technology

Sign language translation system

Automatically translates sign language text into natural language text

Achieves a BLEU score of 73.84 on the ASLG-PC12 dataset

Educational technology

Sign language learning aid

Helps learners understand the natural language expressions corresponding to sign language texts

🚀 t5_small_aslg_pc12

This model is a fine - tuned version of [t5 - small](https://huggingface.co/t5 - small) on the aslg_pc12 dataset, which can be used for translation tasks.

📚 Documentation

Model Information

Property	Details
Model Type	Fine - tuned version of t5 - small
Training Data	aslg_pc12
Metrics	bertscore, bleu, comet, rouge
Pipeline Tag	translation

Model Index

Name: t5_small_aslg_pc12
Results:
- Task:
  - Type: translation
  - Name: Translation
- Dataset:
  - Name: aslg_pc12
  - Type: aslg_pc12
  - Config: default
  - Split: train
- Metrics:
  - Type: bleu
  - Value: 73.8405
  - Name: BLEU
  - Verified: true
  - VerifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiYzM4ODViYTVlYjVjZjUwNzI2YzM4YTYxMjBlZjIxNWI2YjNmM2RkOWU1NGU3NTZlYWYxNDU3YjRlNzFmNWQ4MCIsInZlcnNpb24iOjF9.KNo - oNa4YBfVvNzs7 - x5b2 - J1MThZX9lgztxklJVR7uwrRMvNnJb32mThwK_4Ge_WqPcy - zFHEeF6mCKZ - QWCA
  - Type: loss
  - Value: 0.2336091846227646
  - Name: loss
  - Verified: true
  - VerifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiNGZlMDZlMzYxNWUzNjk0ZmM4MDg0YTc1YjUyYjcyMTJmMTQxNmVlOTAxZGU3MTY1M2FjZDBhMmIwYzQwMmIwMyIsInZlcnNpb24iOjF9.PEWz - fUp1QjRztcRLHhmInmEGbTefHq - 6a9M4HUh7Krdd1Ih8aoWoMdZE8 - CCKy_zS6vhZFLUbWocaJw8TH0BA
  - Type: gen_len
  - Value: 15.4908
  - Name: gen_len
  - Verified: true
  - VerifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiMzk3MmNkNDUxOGRjNmQxZDhjNGM4N2Y0NjFhMWQyOTViMjU3NzRiMTJiMzAwZjFkZjkxMTg0YzY4MTZkNjBjZiIsInZlcnNpb24iOjF9.QIx8UAWOLibfiqNhWP3e4m69rMOzrGhk4iRH2rdwN8NEFUGDJnHrnruhD6qU7doc7W770GCFOo0ZxUV01V7xDQ

Train - Eval Index

Config: default
Task: translation
Task ID: translation
Splits:
- Eval Split: train
Col Mapping:
- gloss: source
- text: target

🔧 Technical Details

Training Procedure

Training Hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e - 05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9, 0.999) and epsilon = 1e - 08
lr_scheduler_type: linear
num_epochs: 3.0

Framework Versions

Transformers 4.34.0
Pytorch 2.0.1+cu118
Datasets 2.14.5
Tokenizers 0.14.1

📄 License

This model is released under the Apache - 2.0 license.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご