You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
There has been a recent number of issues found in various frontends for Relay. I propose to ensure robustness and compatibility with popular model architectures we make an effort to import commonly used models and test their correctness.
We should consider more complex scenarios such as PyTorch -> ONNX -> Relay. I have added some initial tests in #3084 but there are still some issues in supporting all of torchvision.
The text was updated successfully, but these errors were encountered:
@jroesch@tqchen is anyone taking this? if not, i'm interested in taking this. I'm interested in making more contributions to tvm in the longer run, this testing infrastructure could be a good start for me to understand some of the code :)
There has been a recent number of issues found in various frontends for Relay. I propose to ensure robustness and compatibility with popular model architectures we make an effort to import commonly used models and test their correctness.
We should consider more complex scenarios such as PyTorch -> ONNX -> Relay. I have added some initial tests in #3084 but there are still some issues in supporting all of
torchvision
.The text was updated successfully, but these errors were encountered: