Add high-frequency subgraph sequence extraction script #372
Conversation
This file contains hidden or 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! |
lixinqi
reviewed
Nov 20, 2025
| token_tensor[start:end] | ||
| for consecutive_tensor in consecutive_tensors | ||
| for start, end in [get_range(int(consecutive_tensor.size(0)))] | ||
| for start, end in [get_range(len(consecutive_tensor))] |
Collaborator
There was a problem hiding this comment.
之前的语义是已知consecutive_tensor 是个 1D的 tensor，其 size(0)其实就是这个张量的长度。
后面的代码看起来意思是一样的，这么改的原因是啥呢？
Contributor
Author
There was a problem hiding this comment.
TypeError: 'int' object is not callable
这里语义上是tensor，但是实际上是一个np的数组，这样调用是有问题的
Collaborator
这个图里的用法不是真实情况。应该是一下子塞入 resnet18 和 resnet34，综合两个计算图的情况来抽取高频子图 |
fangfangssj
commented
Nov 20, 2025
Comment on lines
+157
to
+168
| inputs = [] | ||
| valid_model_names = [] | ||
|
|
||
| for model_path in args.models: | ||
| seq = extract_ops_via_compile(model_path, args.device) | ||
| inputs.append(seq) | ||
| valid_model_names.append(os.path.basename(model_path)) | ||
|
|
||
| rp_parser = RpExprParser( | ||
| window_size=args.window, fold_policy="default", fold_times=0 | ||
| ) | ||
| rp_expr, token_id2primitive_id = rp_parser(inputs) |
Contributor
Author
There was a problem hiding this comment.
这里就是把计算图一起输入的，并不是单个分开输入
lixinqi
reviewed
Nov 20, 2025
| description="Extract graph patterns and split points from multiple models." | ||
| ) | ||
| parser.add_argument( | ||
| "--models", |
Collaborator
There was a problem hiding this comment.
这种调用方式并不方便。改成 model-dirs-file。从文本里读取多个 model_path
| help="List of paths to model directories (e.g. --models path/to/m1 path/to/m2)", | ||
| ) | ||
| parser.add_argument("--device", type=str, default="cuda") | ||
| parser.add_argument("--window", type=int, default=10) |
Collaborator
There was a problem hiding this comment.
add argument output_split_points.
put results into output_split_points file with scheme like
<model-path> <space> 3,5,7
Collaborator
|
todo things:
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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 Category
Feature Enhancement
Description
添加高频子图序列提取脚本
运行命令 python split_points.py --models ./samples/torchvision/resnet18 ./samples/torchvision/resnet34
下图为windows为10时，测试的结果