size mismatch for emb_g.weight: copying a param with shape torch.Size([5, 256]) from checkpoint, the shape in current model is torch.Size([7, 256]). about vits-simple-api HOT 4 CLOSED

可以看一下后台日志吗

from vits-simple-api.

可以看一下后台日志吗

moegoe_1 | Traceback (most recent call last):
moegoe_1 | File "/app/app.py", line 19, in
moegoe_1 | voice_obj, voice_speakers = merge_model(app.config["MODEL_LIST"])
moegoe_1 | File "/app/utils.py", line 183, in merge_model
moegoe_1 | obj = Voice(model=i[0], config=i[1], out_path=out_path)
moegoe_1 | File "/app/voice.py", line 45, in init
moegoe_1 | utils.load_checkpoint(model, self.net_g_ms)
moegoe_1 | File "/app/utils.py", line 70, in load_checkpoint
moegoe_1 | model.load_state_dict(new_state_dict)
moegoe_1 | File "/usr/local/lib/python3.9/site-packages/torch/nn/modules/module.py", line 2041, in load_state_dict
moegoe_1 | raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
moegoe_1 | RuntimeError: Error(s) in loading state_dict for SynthesizerTrn:
moegoe_1 | size mismatch for enc_p.emb.weight: copying a param with shape torch.Size([52, 192]) from checkpoint, the shape in current model is torch.Size([40, 192]).
moegoe_1 | size mismatch for emb_g.weight: copying a param with shape torch.Size([5, 256]) from checkpoint, the shape in current model is torch.Size([7, 256]).

from vits-simple-api.

尝试复现出了这个错误，原因是你的模型和config.json是不匹配的，就会出现这个错误，一定要使用该模型对应的config.json

from vits-simple-api.

已解决，config.json确实不对应，谢谢作者的解答

from vits-simple-api.