Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[BUG] Unable to serve a session-based topk_model on Triton due to TypeError: Unable to serialize item_id-list to JSON. Unrecognized type <class 'merlin.schema.schema.ColumnSchema'>. #1212

Closed
rnyak opened this issue Sep 21, 2023 · 0 comments · Fixed by #1214
Labels
bug Something isn't working P0 status/needs-triage

Comments

@rnyak
Copy link
Contributor

rnyak commented Sep 21, 2023

Bug description

I created this bug ticket on systems and cross linking it here.

The issue might be solved from Merlin models side, not from Systems side.

The main reason for this bug is that the topk model cannot be saved properly. we get an error like below:

...
File /usr/lib/python3.8/json/encoder.py:257, in JSONEncoder.iterencode(self, o, _one_shot)
    252 else:
    253     _iterencode = _make_iterencode(
    254         markers, self.default, _encoder, self.indent, floatstr,
    255         self.key_separator, self.item_separator, self.sort_keys,
    256         self.skipkeys, _one_shot)
--> 257 return _iterencode(o, 0)

TypeError: Unable to serialize item_id-list to JSON. Unrecognized type <class 'merlin.schema.schema.ColumnSchema'>.

Steps/Code to reproduce bug

Please run the code in this gist to repro the issue.

Expected behavior

We should be able to serve a session-based topk_model on Triton and be able to return topK scores and corresponding item-ids

Environment details

  • Merlin version:
  • Platform:
  • Python version:
  • PyTorch version (GPU?):
  • Tensorflow version (GPU?):
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working P0 status/needs-triage
Projects
None yet
Development

Successfully merging a pull request may close this issue.

1 participant