[Bug]
Closed this issue · 3 comments
MalaJeans commented
test code is as follows:
#!/usr/bin/python3
import torch
# Define pytorch model
model = torch.hub.load('pytorch/vision:v0.6.0', 'resnet18', pretrained=True).cuda().eval()
x = torch.rand(1, 3, 224, 224).cuda()
# Compile the model through Hidet
# Optional: set optimization options (see our documentation for more details)
# import hidet
# hidet.torch.dynamo_config.search_space(2) # tune each tunable operator
# hidet.torch.dynamo_config.use_fp16() # use float16 for acceleration
model_opt = torch.compile(model, backend='hidet')
# Run the optimized model
y = model_opt(x)
Traceback (most recent call last):
File "./test.py", line 13, in <module>
model_opt = torch.compile(model, backend='hidet')
File "/home/user02/.local/lib/python3.8/site-packages/torch/__init__.py", line 1441, in compile
return torch._dynamo.optimize(backend=backend, nopython=fullgraph, dynamic=dynamic, disable=disable)(model)
File "/home/user02/.local/lib/python3.8/site-packages/torch/_dynamo/eval_frame.py", line 424, in optimize
backend = get_compiler_fn(backend)
File "/home/user02/.local/lib/python3.8/site-packages/torch/_dynamo/eval_frame.py", line 363, in get_compiler_fn
compiler_fn = lookup_backend(compiler_fn)
File "/home/user02/.local/lib/python3.8/site-packages/torch/_dynamo/backends/registry.py", line 58, in lookup_backend
_lazy_import_entry_point(compiler_fn)
File "/home/user02/.local/lib/python3.8/site-packages/torch/_dynamo/backends/registry.py", line 100, in _lazy_import_entry_point
eps = [ep for ep in backend_eps[group_name] if ep.name == backend_name]
KeyError: 'torch_dynamo_backends'
May I ask why this error occurs and how to solve it?Thanks.
yaoyaoding commented
Hi @MalaJeans , have you installed hidet?
pip install hidet
MalaJeans commented
It's ready to work. Thank you very much!
MalaJeans commented
It's ready to work. Thank you very much!