jikechao opened a new issue, #15006: URL: https://github.com/apache/tvm/issues/15006
For the following script, it produced different inference results between PyTorch and TVM. ### Actual behavior  ### Steps to reproduce ``` import torch from tvm import relay import tvm import numpy as np from torch.nn import Module input_data = torch.randn([1, 3, 7], dtype=torch.float64) class lp_pool1d(Module): def forward(self, *args): return torch.nn.functional.lp_pool1d(args[0], 1.5, 2) m = lp_pool1d().float().eval() torch_outputs = m(input_data) trace = torch.jit.trace(m, input_data) input_shapes = [('input0', torch.Size([1, 3, 7]))] mod, params = relay.frontend.from_pytorch(trace, input_shapes) print(mod) with tvm.transform.PassContext(opt_level=3): exe = relay.create_executor('graph', mod=mod, params=params, device=tvm.device('llvm', 0), target='llvm').evaluate() input_tvm = {'input0': np.array(input_data, dtype='float64')} tvm_outputs = exe(**input_tvm).asnumpy() np.testing.assert_allclose(torch_outputs, tvm_outputs, rtol=1e-3, atol=1e-3) ``` ### Triage * needs-triage * frontend:pytorch ### Analysis This inconsistency was triggered only we use `lpool1d` and with input_shape=[1, 3, 7]. other input_shape will not produce any inconsistency. Wish your further analyzing and fixing. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
