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

Better chains error propagation (+various fixes). #1271

Merged
merged 2 commits into from
Dec 10, 2024
Merged

Conversation

marius-baseten
Copy link
Contributor

@marius-baseten marius-baseten commented Dec 10, 2024

🚀 What

  • Skip pre-/post-process (and tracing span) if not defined on model.
  • Remove left-over print statememtn in chains streaming.
  • Add some helper functions to integration test chain (for more interesting stack traces).
  • Improve chainlet error chaining by filtering out all non-user code and making the formatting nicer.
  • Add test with regex to assert that this filtering is matching in case functions are renamed.
  • Simplify code gen by avoiding chainlet name in exception_to_http_error.

💻 How

🔬 Testing

@marius-baseten marius-baseten changed the title Better chains error propagation. Better chains error propagation (+various fixes). Dec 10, 2024
Copy link
Contributor

@tyranitar tyranitar left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good!

@marius-baseten marius-baseten enabled auto-merge (squash) December 10, 2024 18:25
@marius-baseten marius-baseten merged commit c570bb0 into main Dec 10, 2024
4 checks passed
@marius-baseten marius-baseten deleted the marius/errors branch December 10, 2024 18:27
joostinyi added a commit that referenced this pull request Dec 10, 2024
* update (#1267)

Co-authored-by: Tianshu Cheng <tianshuc@Tianshus-MBP.localdomain>

* send truss version on patch (#1268)

* Speculative Decoding Interface refactor (#1270)

* spec dec config

* add optional dict of trt llm configs

* fix bad merge

* add extensions support

* fix fixture

* cli push fixes

* constants

* fix ordering

* fix merge

* refactor interface

* add tp validation error

* self review

* use constant

* fix tests

* fix tests

* add request_default_max_tokens

* fix default on trtllm runtime

* update copy

* bump to 54rc0

* add total token limit to toplevel config

* bump briton to 0.3.10

* fix import

* 54rc2

* fix rc3

* rc4

* bump briton server image

* bump rc6 for briton 0.3.12.dev3

* bump rc7

* revert trtllm serialization changes

* bump briton

* interface refactor

* add validation + tests

* 56rc0

* reduce property

* Update trt_llm_config.py (#1274)

* Update trt_llm_config.py -> revision (#1269)

* Better chains error propagation (+various fixes). (#1271)

* Bump briton in truss library (#1273)

* bump briton to briton==0.3.12.dev8

* bump truss to 0.9.56rc1

* Bump version to 0.9.56

---------

Co-authored-by: Tianshu <26018552+tianshuc0731@users.noreply.github.com>
Co-authored-by: Tianshu Cheng <tianshuc@Tianshus-MBP.localdomain>
Co-authored-by: rcano-baseten <raymond.cano@baseten.co>
Co-authored-by: joostinyi <63941848+joostinyi@users.noreply.github.com>
Co-authored-by: Michael Feil <63565275+michaelfeil@users.noreply.github.com>
Co-authored-by: Marius Killinger <155577904+marius-baseten@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants