hamzaqureshi5 opened a new issue, #18008:
URL: https://github.com/apache/tvm/issues/18008
### Expected behavior
import tvm
mod = tvm.relax.frontend.onnx.from_onnx(onnx_model, shape_dict=shape_dict) #
type: ignore
Should work
### Actual behavior
---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
Cell In[2], line 11
5 # Properly define shape_dict using symbolic TVM variables
6 shape_dict = {
7 "input_ids": [batch, seq_len],
8 "attention_mask": [batch, seq_len],
9 }
---> 11 mod = tvm.relax.frontend.onnx.from_onnx(onnx_model,
shape_dict=shape_dict) # type: ignore
12 mod
AttributeError: module 'tvm.relax.frontend' has no attribute 'onnx'
Operating System: Ubuntu, TVM version: 0.21 dev Python: 3.11.11
### Steps to reproduce
# Define symbolic variables with int64 dtype
batch = tvm.tir.Var("batch", "int64")
seq_len = tvm.tir.Var("seq_len", "int64")
# Properly define shape_dict using symbolic TVM variables
shape_dict = {
"input_ids": [batch, seq_len],
"attention_mask": [batch, seq_len],
}
mod = tvm.relax.frontend.onnx.from_onnx(onnx_model, shape_dict=shape_dict) #
type: ignore
mod
--
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]
