What is it?

A text classification model for 200 languages using Meta-AI nllb200 and Google BERT model.

how to run this example?

1. install torch and huggingface transformers library through pip3 install torch huggingface
2. download BERT base mode and nllb200 600m model through the following two command: huggingface-cli download facebook/nllb-200-distilled-600M huggingface-cli download google-bert/bert-base-uncased
3. train model through python3 hello_bert.py
4. comment and/or uncomment the function callings in hello_bert.py to test the model

def main():
    device = torch.device("cuda:0")
    model = MyClassifierNet(output_count=7, device=device).to(device)
    model.load_state_dict(torch.load("led_controller.pth"))
    #train_model(model, device)
    test_model(model, device)
    
    torch.save(model.state_dict(), "led_controller.pth")

Screenshots: