训练时报错 PytorchStreamReader failed reading zip archive: failed finding central directory

**Summary[问题简述（一句话）]**
A clear and concise description of what the issue is.
训练时报错 PytorchStreamReader failed reading zip archive: failed finding central directory
**Env & To Reproduce[复现与环境]**
描述你用的环境、代码版本、模型
gtx1060
py3.9.7

**Screenshots[截图（如有）]**
If applicable, add screenshots to help

`(mygui) PS D:\video\MockingBird> python synthesizer_train.py mandarin D:\video/SV2TTS/synthesizer
Arguments:
run_id: mandarin
syn_dir: D:\video/SV2TTS/synthesizer
models_dir: synthesizer/saved_models/
save_every: 1000
backup_every: 25000
log_every: 200
force_restart: False
hparams:

Checkpoint path: synthesizer\saved_models\mandarin\mandarin.pt
Loading training data from: D:\video\SV2TTS\synthesizer\train.txt
Using model: Tacotron
Using device: cuda

Initialising Tacotron Model...

Traceback (most recent call last):
File "D:\video\MockingBird\synthesizer_train.py", line 37, in
train(**vars(args))
File "D:\video\MockingBird\synthesizer\train.py", line 75, in train
loaded_shape = torch.load(str(weights_fpath), map_location=device)["model_state"]["encoder.embedding.weight"].shape
File "E:\Anaconda3\envs\mygui\lib\site-packages\torch\serialization.py", line 705, in load
with _open_zipfile_reader(opened_file) as opened_zipfile:
File "E:\Anaconda3\envs\mygui\lib\site-packages\torch\serialization.py", line 243, in __init__
super(_open_zipfile_reader, self).__init__(torch._C.PyTorchFileReader(name_or_buffer))
RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory`

Start with synthesizer_train.py and synthesizer/train.py, then reproduce the provided command and inspect the checkpoint at synthesizer\saved_models\mandarin\mandarin.pt. Verify the checkpoint and model-loading path around torch.load; the issue is done when the failure's cause is confirmed and an actionable resolution is documented.

Written by the indexing model from the issue text.