[BUG]: RuntimeError: Given group Error

[FrankieDong]

🐛 Describe the bug

我换了一批图片进行finetune，目前在推理阶段出现了以下错误： （之前finetune的模型推理没有问题） Traceback (most recent call last): File "scripts/txt2img.py", line 345, in main() File "scripts/txt2img.py", line 296, in main samples_ddim, _ = sampler.sample(S=opt.ddim_steps, File "/opt/conda/envs/pytorch/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context return func(*args, **kwargs) File "/workspace/ColossalAI-0.1.10/ColossalAI/examples/images/diffusion/./ldm/models/diffusion/plms.py", line 97, in sample samples, intermediates = self.plms_sampling(conditioning, size, File "/opt/conda/envs/pytorch/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context return func(*args, **kwargs) File "/workspace/ColossalAI-0.1.10/ColossalAI/examples/images/diffusion/./ldm/models/diffusion/plms.py", line 152, in plms_sampling outs = self.p_sample_plms(img, cond, ts, index=index, use_original_steps=ddim_use_original_steps, File "/opt/conda/envs/pytorch/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context return func(*args, **kwargs) File "/workspace/ColossalAI-0.1.10/ColossalAI/examples/images/diffusion/./ldm/models/diffusion/plms.py", line 218, in p_sample_plms e_t = get_model_output(x, t) File "/workspace/ColossalAI-0.1.10/ColossalAI/examples/images/diffusion/./ldm/models/diffusion/plms.py", line 185, in get_model_output e_t_uncond, e_t = self.model.apply_model(x_in, t_in, c_in).chunk(2) File "/workspace/ColossalAI-0.1.10/ColossalAI/examples/images/diffusion/./ldm/models/diffusion/ddpm.py", line 1068, in apply_model x_recon = self.model(x_noisy, t, **cond) File "/opt/conda/envs/pytorch/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl return forward_call(*input, **kwargs) File "/workspace/ColossalAI-0.1.10/ColossalAI/examples/images/diffusion/./ldm/models/diffusion/ddpm.py", line 1493, in forward out = self.diffusion_model(x, t, context=cc) File "/opt/conda/envs/pytorch/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl return forward_call(*input, **kwargs) File "/workspace/ColossalAI-0.1.10/ColossalAI/examples/images/diffusion/./ldm/modules/diffusionmodules/openaimodel.py", line 923, in forward h = module(h, emb, context) File "/opt/conda/envs/pytorch/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl return forward_call(*input, **kwargs) File "/workspace/ColossalAI-0.1.10/ColossalAI/examples/images/diffusion/./ldm/modules/diffusionmodules/openaimodel.py", line 91, in forward x = layer(x) File "/opt/conda/envs/pytorch/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl return forward_call(*input, **kwargs) File "/opt/conda/envs/pytorch/lib/python3.8/site-packages/torch/nn/modules/conv.py", line 457, in forward return self._conv_forward(input, self.weight, self.bias) File "/opt/conda/envs/pytorch/lib/python3.8/site-packages/torch/nn/modules/conv.py", line 453, in _conv_forward return F.conv2d(input, weight, bias, self.stride, RuntimeError: Given groups=1, weight of size [320, 4, 3, 3], expected input[4, 2, 64, 64] to have 4 channels, but got 2 channels instead

Environment

No response

[Issues-translate-bot]

Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑‍🤝‍🧑👫🧑🏿‍🤝‍🧑🏻👩🏾‍🤝‍👨🏿👬🏿

---

🐛 Describe the bug

I changed a batch of pictures for finetune, and the following errors occurred in the inference stage: (There was no problem with finetune's model reasoning before) Traceback (most recent call last): File "scripts/txt2img.py", line 345, in main() File "scripts/txt2img.py", line 296, in main samples_ddim, _ = sampler.sample(S=opt.ddim_steps, File "/opt/conda/envs/pytorch/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context return func(*args, **kwargs) File "/workspace/ColossalAI-0.1.10/ColossalAI/examples/images/diffusion/./ldm/models/diffusion/plms.py", line 97, in sample samples, intermediates = self.plms_sampling(conditioning, size, File "/opt/conda/envs/pytorch/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context return func(*args, **kwargs) File "/workspace/ColossalAI-0.1.10/ColossalAI/examples/images/diffusion/./ldm/models/diffusion/plms.py", line 152, in plms_sampling outs = self.p_sample_plms(img, cond, ts, index=index, use_original_steps=ddim_use_original_steps, File "/opt/conda/envs/pytorch/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context return func(*args, **kwargs) File "/workspace/ColossalAI-0.1.10/ColossalAI/examples/images/diffusion/./ldm/models/diffusion/plms.py", line 218, in p_sample_plms e_t = get_model_output(x, t) File "/workspace/ColossalAI-0.1.10/ColossalAI/examples/images/diffusion/./ldm/models/diffusion/plms.py", line 185, in get_model_output e_t_uncond, e_t = self.model.apply_model(x_in, t_in, c_in).chunk(2) File "/workspace/ColossalAI-0.1.10/ColossalAI/examples/images/diffusion/./ldm/models/diffusion/ddpm.py", line 1068, in apply_model x_recon = self.model(x_noisy, t, **cond) File "/opt/conda/envs/pytorch/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl return forward_call(*input, **kwargs) File "/workspace/ColossalAI-0.1.10/ColossalAI/examples/images/diffusion/./ldm/models/diffusion/ddpm.py", line 1493, in forward out = self. diffusion_model(x, t, context=cc) File "/opt/conda/envs/pytorch/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl return forward_call(*input, **kwargs) File "/workspace/ColossalAI-0.1.10/ColossalAI/examples/images/diffusion/./ldm/modules/diffusionmodules/openaimodel.py", line 923, in forward h = module(h, emb, context) File "/opt/conda/envs/pytorch/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl return forward_call(*input, **kwargs) File "/workspace/ColossalAI-0.1.10/ColossalAI/examples/images/diffusion/./ldm/modules/diffusionmodules/openaimodel.py", line 91, in forward x = layer(x) File "/opt/conda/envs/pytorch/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl return forward_call(*input, **kwargs) File "/opt/conda/envs/pytorch/lib/python3.8/site-packages/torch/nn/modules/conv.py", line 457, in forward return self._conv_forward(input, self.weight, self.bias) File "/opt/conda/envs/pytorch/lib/python3.8/site-packages/torch/nn/modules/conv.py", line 453, in _conv_forward return F.conv2d(input, weight, bias, self.stride, RuntimeError: Given groups=1, weight of size [320, 4, 3, 3], expected input[4, 2, 64, 64] to have 4 channels, but got 2 channels instead

Environment

No response