siju-samuel commented on a change in pull request #5383:
URL: https://github.com/apache/incubator-tvm/pull/5383#discussion_r412094229
##########
File path: python/tvm/relay/frontend/pytorch.py
##########
@@ -337,6 +337,55 @@ def _impl(inputs, input_types):
return _op.transform.repeat(data, repeats=repeats, axis=axis)
return _impl
+
+def _parse_input_data(inp):
+ import torch
+ if isinstance(inp, _expr.Var):
Review comment:
if one of the input is `torch.ones,` it will first convert to
`_expr.Call `for `torch.ones`, for these kind of inputs it will happen.
actually we only need to differentiate `torch.Tensor` as tvm ops cannot input
`torch.tensor`.
Any tvm expr as input, tvm will handle.
Whereever having check for `_expr.var` & `torch.tensor` is there, need to
handle `_expr.call` also.
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
