Make SynthSkillsFlow honor the num_iters parameter #82
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
The base `Flow` class takes a `num_iters` parameter. This is used by
the `ilab` CLI to adjust how much data is generated by a given run.
The old value for this parameter in the CLI was 100, so while testing
this in the short term, it should be specified explicitly like this:
ilab data generate --num-instructions 30
When we release a version of this library that includes this effective
rewrite, the default value will be 30 and the option and its
description will better reflect the new behavior.
ilab data generate --sdg-scale-factor 30
More detail about how this is exposed via the CLI can be found in this
PR:
instructlab/instructlab#1570
Honoring this parameter for the full pipeline will be used immediately
in CI integration, where we're testing that the code can run
successfully, but want to do so as quickly as is reasonable.
Currently, E2E always runs with this setting set to 1 for speed
purposes.
Signed-off-by: Russell Bryant <[email protected]>
|
lgtm 👍 |
|
This is a trivial fix and it's causing CI problems, so I'm going to merge it with @markmc 's review |
markmc
added a commit
to markmc/sdg
that referenced
this pull request
Jul 11, 2024
Two pipelines include an LLMBlock which use `{num_samples}` in their
instructions to the teacher model. There needs to be some way to
configure the LLMBlock so that `num_samples` will be included, but
as per instructlab#82 (commit a01b04e) the value of `num_samples` should be
based on the `num_instructions_to_generate` parameter.
Signed-off-by: Mark McLoughlin <[email protected]>
jwm4
pushed a commit
to jwm4/sdg
that referenced
this pull request
Dec 13, 2024
Proposal for new repository for UI work
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.
The base
Flowclass takes anum_itersparameter. This is used bythe
ilabCLI to adjust how much data is generated by a given run.The old value for this parameter in the CLI was 100, so while testing
this in the short term, it should be specified explicitly like this:
When we release a version of this library that includes this effective
rewrite, the default value will be 30 and the option and its
description will better reflect the new behavior.
More detail about how this is exposed via the CLI can be found in this
PR:
instructlab/instructlab#1570
Honoring this parameter for the full pipeline will be used immediately
in CI integration, where we're testing that the code can run
successfully, but want to do so as quickly as is reasonable.
Currently, E2E always runs with this setting set to 1 for speed
purposes.
Signed-off-by: Russell Bryant [email protected]