feat: add final evaluation #84
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
leseb
force-pushed
the
final-eval
branch
3 times, most recently
from
5b9f7f2 to
832f13a
Compare
MichaelClifford
force-pushed
the
final-eval
branch
from
832f13a to
8122f5b
Compare
MichaelClifford
force-pushed
the
final-eval
branch
from
adaf720 to
891f726
Compare
sallyom
reviewed
Oct 11, 2024
| updated_lines.append(new_line) | ||
|
|
||
| if changed: | ||
| with open(file_path, "w", encoding="utf-8") as file: |
There was a problem hiding this comment.
@MichaelClifford the "utf-8" to fix the check_printable error, 🙏
sallyom
force-pushed
the
final-eval
branch
2 times, most recently
from
55dfafd to
64c9224
Compare
leseb
force-pushed
the
final-eval
branch
5 times, most recently
from
bc44932 to
6aa2b1b
Compare
tumido
reviewed
Oct 15, 2024
eval/final/components.py
Outdated
Comment on lines
54
to
71
| if gpu_count > 0: | ||
| command = [ | ||
| sys.executable, | ||
| "-m", | ||
| "vllm.entrypoints.openai.api_server", | ||
| "--model", | ||
| model_path, | ||
| "--tensor-parallel-size", | ||
| str(gpu_count), | ||
| ] | ||
| else: | ||
| command = [ | ||
| sys.executable, | ||
| "-m", | ||
| "vllm.entrypoints.openai.api_server", | ||
| "--model", | ||
| model_path, | ||
| ] |
There was a problem hiding this comment.
Nit, making this less repetitive
Suggested change
| if gpu_count > 0: | |
| command = [ | |
| sys.executable, | |
| "-m", | |
| "vllm.entrypoints.openai.api_server", | |
| "--model", | |
| model_path, | |
| "--tensor-parallel-size", | |
| str(gpu_count), | |
| ] | |
| else: | |
| command = [ | |
| sys.executable, | |
| "-m", | |
| "vllm.entrypoints.openai.api_server", | |
| "--model", | |
| model_path, | |
| ] | |
| command = [ | |
| sys.executable, | |
| "-m", | |
| "vllm.entrypoints.openai.api_server", | |
| "--model", | |
| model_path, | |
| ] | |
| if gpu_count > 0: | |
| command += [ | |
| "--tensor-parallel-size", | |
| str(gpu_count), | |
| ] |
Signed-off-by: Sébastien Han <[email protected]>
Signed-off-by: Michael Clifford <[email protected]>
Signed-off-by: sallyom <[email protected]>
It used to be `/data/generated`. Signed-off-by: Sébastien Han <[email protected]>
The shell from the python executor will interpret the content of the backtick. Signed-off-by: Sébastien Han <[email protected]>
Signed-off-by: Sébastien Han <[email protected]>
- do not print final eval scores in logs - use the correct model location for final push - fix job/cr watch - add progress bar for download/upload - add a few globale variables for eval steps - handle the case where pods are removed but the job exists Signed-off-by: Sébastien Han <[email protected]>
THe functions from the helper are embedded into the components themselves. So they are left unused now. Signed-off-by: Sébastien Han <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
8ff3780 feat: add final evaluation
113078c wip:final eval
fd52c08 update sdg generated line in knowledge-yaml
fc7067b fix typo on task generate
339ae90 fix: do not use backticks
5e74166 fix: add final details to push model to S3
a28d739 bulk: mt-bench eval, final eval and trained model push to S3
0885fa5 chore: remove helpers directory
commit 8ff3780
Author: Sébastien Han [email protected]
Date: Thu Oct 10 17:39:25 2024 +0200
commit 113078c
Author: Michael Clifford [email protected]
Date: Thu Oct 10 20:04:04 2024 -0400
commit fd52c08
Author: sallyom [email protected]
Date: Fri Oct 11 13:21:38 2024 -0400
commit fc7067b
Author: Sébastien Han [email protected]
Date: Mon Oct 14 09:40:50 2024 +0200
commit 339ae90
Author: Sébastien Han [email protected]
Date: Mon Oct 14 09:41:33 2024 +0200
commit 5e74166
Author: Sébastien Han [email protected]
Date: Mon Oct 14 09:43:24 2024 +0200
commit a28d739
Author: Sébastien Han [email protected]
Date: Mon Oct 14 23:00:15 2024 +0200
commit 0885fa5
Author: Sébastien Han [email protected]
Date: Tue Oct 15 17:12:32 2024 +0200