read_fast.py script breaks without cuda

[vicbois390]

Running CPU only just to test things out as I have an AMD GPU. I am able to run do_tts.py and read.py (eventually), but when attempting to run read_fast.py, the following error appears:
Traceback (most recent call last): File "c:\Users\victor\tortoise-tts\tortoise\read_fast.py", line 33, in <module> tts = TextToSpeech(models_dir=args.model_dir, use_deepspeed=args.use_deepspeed, kv_cache=args.kv_cache, half=args.half) File "c:\Users\victor\tortoise-tts\tortoise\api_fast.py", line 222, in __init__ hifi_model = torch.load(get_model_path('hifidecoder.pth')) File "D:\miniconda3\envs\tortoise\lib\site-packages\torch\serialization.py", line 1026, in load return _load(opened_zipfile, File "D:\miniconda3\envs\tortoise\lib\site-packages\torch\serialization.py", line 1438, in _load result = unpickler.load() File "D:\miniconda3\envs\tortoise\lib\site-packages\torch\serialization.py", line 1408, in persistent_load typed_storage = load_tensor(dtype, nbytes, key, _maybe_decode_ascii(location)) File "D:\miniconda3\envs\tortoise\lib\site-packages\torch\serialization.py", line 1382, in load_tensor wrap_storage=restore_location(storage, location), File "D:\miniconda3\envs\tortoise\lib\site-packages\torch\serialization.py", line 391, in default_restore_location result = fn(storage, location) File "D:\miniconda3\envs\tortoise\lib\site-packages\torch\serialization.py", line 266, in _cuda_deserialize device = validate_cuda_device(location) File "D:\miniconda3\envs\tortoise\lib\site-packages\torch\serialization.py", line 250, in validate_cuda_device raise RuntimeError('Attempting to deserialize object on a CUDA ' RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with map_location=torch.device('cpu') to map your storages to the CPU.

Using no programming knowledge and only my basic ability to read, I changed line 33 in api_fast.py from
hifi_model = torch.load(get_model_path('hifidecoder.pth'))
to
hifi_model = torch.load(get_model_path('hifidecoder.pth'), map_location=torch.device('cpu'))

This resolved the error.

It seems like something is failing to check for CUDA at or before this point in the script, but I really don't know enough to say for sure. Since it is working for me now I thought it best to bring it up.