add paddle nv-embed-v1 #8785
Merged
add paddle nv-embed-v1 #8785
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 ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## develop #8785 +/- ##
===========================================
+ Coverage 55.44% 55.58% +0.14%
===========================================
Files 626 630 +4
Lines 98065 98382 +317
===========================================
+ Hits 54368 54683 +315
- Misses 43697 43699 +2 ☔ View full report in Codecov by Sentry. |
sijunhe
reviewed
Jul 24, 2024
Comment on lines
38
to
39
| hf_model = LoRAModel.from_pretrained(base_model, peft_model_name, lora_config=lora_config, dtype="bfloat16") | ||
| return hf_model |
Comment on lines
81
to
89
| self.cross_attend_blocks_0_fn_to_kv = paddle.nn.Linear(in_features=4096, out_features=65536, bias_attr=False) | ||
| self.cross_attend_blocks_0_fn_to_out = paddle.nn.Linear(in_features=32768, out_features=4096, bias_attr=False) | ||
| self.cross_attend_blocks_0_fn_to_q = paddle.nn.Linear(in_features=4096, out_features=32768, bias_attr=False) | ||
| self.cross_attend_blocks_0_norm = paddle.nn.LayerNorm(4096) | ||
| self.cross_attend_blocks_0_norm_context = paddle.nn.LayerNorm(4096) | ||
|
|
||
| self.cross_attend_blocks_1_fn_net_0 = paddle.nn.Linear(in_features=4096, out_features=32768) | ||
| self.cross_attend_blocks_1_fn_net_2 = paddle.nn.Linear(in_features=16384, out_features=4096) | ||
| self.cross_attend_blocks_1_norm = paddle.nn.LayerNorm(4096) |
There was a problem hiding this comment.
这个模型里面的参数大小为什么是写死的？ 既然是基于Mistral，应该基于MistralConfig里面的大小
There was a problem hiding this comment.
已修改为：使用config.json中的参数值
| ``` | ||
| export CUDA_VISIBLE_DEVICES=0 | ||
| python eval_mteb.py \ | ||
| --base_model_name_or_path NV-Embed-v1 \ |
There was a problem hiding this comment.
这个NV-Embed-v1 是怎么得到的呢？从torch 转过来的吗？
There was a problem hiding this comment.
是的，陆老师发您的文件就是从torch转过来的paddle版本的NV-Embed-v1模型权重
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
Models
Description
add paddle nv-embed-v1 embedding model, and integrate it into the MTEB evaluation framework