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

Remove imports of code from instructlab #6

Closed
russellb opened this issue Jun 2, 2024 · 1 comment
Closed

Remove imports of code from instructlab #6

russellb opened this issue Jun 2, 2024 · 1 comment

Comments

@russellb
Copy link
Member

russellb commented Jun 2, 2024

Since this code was pulled from the instructlab repo, it still imports some code from there. We do not want a bidirectional dependency between the packages. We need to remove all instructlab imports from here.

russellb added a commit that referenced this issue Jun 3, 2024
All of these need to be removed eventually. Issue #6 tracks this.

Signed-off-by: Russell Bryant <[email protected]>
@russellb
Copy link
Member Author

closing since #11 is the same and has more comments

makelinux added a commit to makelinux/instructlab-sdg that referenced this issue Aug 19, 2024
It is not required anymore because
instructlab#6 is resolved.

Signed-off-by: Costa Shulyupin <[email protected]>
makelinux added a commit to makelinux/instructlab-sdg that referenced this issue Aug 19, 2024
It is not required anymore because
instructlab#6 is resolved.

Signed-off-by: Costa Shulyupin <[email protected]>
bbrowning pushed a commit that referenced this issue Dec 10, 2024
…ses (#6)

* feat: Add a simple flow to get granite model responses

Signed-off-by: shiv-sr <[email protected]>

* refactor: Split dataset into smaller batches for improved processing

This commit modifies the  class in  to split the dataset into smaller batches for improved processing. The code now uses start and end ids of the batch to split dataset and creates a batch on the fly rather than creating the batches apriori. This is necessary for large datasets.

Signed-off-by: shiv-sr <[email protected]>

* refactor: Update LLMBlock to use default model if model_id is not provided

This commit modifies the LLMBlock class in llmblock.py to use the default model if the model_id is not provided. It checks if the model_id is None and if so, retrieves the default model id from the client. This ensures that the LLMBlock can still function even if the model_id is not explicitly specified.

Signed-off-by: shiv-sr <[email protected]>

* refactor: Remove commented out code for dataset batching

The code changes in sdg.py remove the commented out code that was used for dataset batching. This code is no longer needed and can be safely removed.

Signed-off-by: shiv-sr <[email protected]>

---------

Signed-off-by: shiv-sr <[email protected]>
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

No branches or pull requests

1 participant