Closed lumoslnt closed 1 week ago
4 files 4 suites 4m 6s :stopwatch: 238 tests 203 :white_check_mark: 35 :zzz: 0 :x: 952 runs 812 :white_check_mark: 140 :zzz: 0 :x:
Results for commit 115cdbbb.
:recycle: This comment has been updated with latest results.
0 tests 0 :white_check_mark: 0s :stopwatch: 0 suites 0 :zzz: 0 files 0 :x:
Results for commit 115cdbbb.
:recycle: This comment has been updated with latest results.
6 tests 6 :white_check_mark: 1m 20s :stopwatch: 1 suites 0 :zzz: 1 files 0 :x:
Results for commit 115cdbbb.
:recycle: This comment has been updated with latest results.
792 tests 792 :white_check_mark: 3m 44s :stopwatch: 1 suites 0 :zzz: 1 files 0 :x:
Results for commit 115cdbbb.
:recycle: This comment has been updated with latest results.
242 tests 237 :white_check_mark: 5m 11s :stopwatch: 1 suites 5 :zzz: 1 files 0 :x:
Results for commit 115cdbbb.
:recycle: This comment has been updated with latest results.
4 files 4 suites 59m 44s :stopwatch: 668 tests 655 :white_check_mark: 13 :zzz: 0 :x: 2 672 runs 2 620 :white_check_mark: 52 :zzz: 0 :x:
Results for commit 115cdbbb.
:recycle: This comment has been updated with latest results.
Description
Currently, if user call OpenAI API with streaming mode in dag flow, the completion tokens in run info will be zero. In this PR, we fix the error by calculating tokens after the output is consumed.
All Promptflow Contribution checklist:
General Guidelines and Best Practices
Testing Guidelines