Loading pipeline components...: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 6/6 [00:00<00:00, 9.06it/s]
| load 'model' from 'checkpoints/0228_opencpop_ds100_rel/model_ckpt_steps_160000.ckpt'.
Traceback (most recent call last):
File "/data1/muzic/musicagent/gradio_agent.py", line 141, in
agent = MusicAgent(args.config, mode="gradio")
File "/data1/muzic/musicagent/agent.py", line 41, in init
self.pipes = init_plugins(self.config)
File "/data1/muzic/musicagent/plugins.py", line 102, in init_plugins
pipes["DiffSinger"] = DiffSinger(config)
File "/data1/muzic/musicagent/plugins.py", line 294, in init
self._init_toolkit(config)
File "/data1/muzic/musicagent/plugins.py", line 308, in _init_toolkit
self.model = DiffSingerE2EInfer(hparams, device="cuda:0")
File "/data1/muzic/musicagent/models/DiffSinger/inference/svs/base_svs_infer.py", line 33, in init
self.model = self.build_model()
File "/data1/muzic/musicagent/models/DiffSinger/inference/svs/ds_e2e.py", line 28, in build_model
utils.load_ckpt(self.pe, hparams['pe_ckpt'], 'model', strict=True)
File "/data1/muzic/musicagent/models/DiffSinger/utils/init.py", line 207, in load_ckpt
assert False, e_msg
AssertionError: | ckpt not found in checkpoints/0102_xiaoma_pe.
### Tasks
从wget https://github.com/MoonInTheRiver/DiffSinger/releases/download/pretrain-model/0102_xiaoma_pe.zip地方下载文件后，再次加载还是报错，如下：| load 'model' from 'checkpoints/0102_xiaoma_pe/model_ckpt_steps_60000.ckpt'.
Traceback (most recent call last):
File "/data1/muzic/musicagent/agent.py", line 351, in
agent = MusicAgent(args.config, mode="cli")
File "/data1/muzic/musicagent/agent.py", line 41, in init
self.pipes = init_plugins(self.config)
File "/data1/muzic/musicagent/plugins.py", line 102, in init_plugins
pipes["DiffSinger"] = DiffSinger(config)
File "/data1/muzic/musicagent/plugins.py", line 294, in init
self._init_toolkit(config)
File "/data1/muzic/musicagent/plugins.py", line 308, in init_toolkit
self.model = DiffSingerE2EInfer(hparams, device="cuda:0")
File "/data1/muzic/musicagent/models/DiffSinger/inference/svs/base_svs_infer.py", line 36, in init
self.vocoder = self.build_vocoder()
File "/data1/muzic/musicagent/models/DiffSinger/inference/svs/base_svs_infer.py", line 49, in build_vocoder
ckpt = sorted(glob.glob(f'{base_dir}/model_ckpt_steps*.ckpt'), key=
IndexError: list index out of range，请教各位大神
I also meet this question. you need to download the all the checkpoint file form Hugging face repository. Below is the download link. https://huggingface.co/spaces/Silentlin/DiffSinger/tree/main. good luck
microsoft
