[llm]support dpo pp #9039
Merged
[llm]support dpo pp #9039
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
|
Thanks for your contribution! |
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## develop #9039 +/- ##
===========================================
- Coverage 53.26% 53.07% -0.20%
===========================================
Files 652 656 +4
Lines 105607 106095 +488
===========================================
+ Hits 56254 56309 +55
- Misses 49353 49786 +433 ☔ View full report in Codecov by Sentry. |
ZHUI
reviewed
Aug 30, 2024
ZHUI
reviewed
Sep 12, 2024
Comment on lines
+279
to
+301
| for key in batch.keys(): | ||
| if key not in "response_indexs": | ||
| concatenated_inputs[key] = [ | ||
| batch[key][i * per_device_train_batch_size : (i + 1) * per_device_train_batch_size] | ||
| for i in range(gradient_accumulation_steps) | ||
| ] | ||
| else: | ||
| concatenated_inputs["response_indexs"] = [[] for _ in range(gradient_accumulation_steps)] | ||
| for i in range(gradient_accumulation_steps): | ||
| for response_index in batch[key]: | ||
| if response_index[0] in list( | ||
| range(i * per_device_train_batch_size, (i + 1) * per_device_train_batch_size) | ||
| ): | ||
| response_index[0] -= i * per_device_train_batch_size | ||
| concatenated_inputs["response_indexs"][i].append(response_index) | ||
| concatenated_inputs["response_indexs"][i] = paddle.stack(concatenated_inputs["response_indexs"][i]) | ||
| if model._layers.config.use_sparse_head_and_loss_fn: | ||
| last_batch_response_length = concatenated_inputs["response_indexs"][i][0, 1] | ||
| concatenated_inputs["response_indexs"][i][:, 1:] -= last_batch_response_length | ||
|
|
||
| concatenated_inputs["reference_chosen_logps"] = None | ||
| concatenated_inputs["reference_rejected_logps"] = None | ||
|
|
ZHUI
reviewed
Sep 18, 2024
|
|
ZHUI
reviewed
Sep 20, 2024
| get_last_checkpoint, | ||
| set_seed, | ||
| from dpo_argument import ( | ||
| DPOConfig, |
There was a problem hiding this comment.
DPOConfig,DPOTrainingArguments 这些看是否加到主repo？
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.
PR types
New features
PR changes
APIs
Description