aalto-speech / ner-asr Goto Github PK

Hey,

Apparently while loading the model in evaluate_document.py there is a mistake
RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False

You can update the script this way to make it work for those loading the model with CPU :)

# load the model
    if lowercase_model == False:
        model.load_state_dict(torch.load('weights/model_upper.pt', map_location=torch.device('cpu')))
    else:
        model.load_state_dict(torch.load('weights/model_lower.pt', map_location=torch.device('cpu')))