Update graphsage #96
Merged
Update graphsage #96
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
Yelrose
reviewed
Aug 30, 2022
| int* actual_size_array = (int*)(node_info_list + shard_len) + i * shard_len; | ||
| uint64_t* sample_array = | ||
| (uint64_t*)(actual_size_array + (shard_len + shard_len % 2) * edge_type_len) + | ||
| i * shard_len * sample_size; |
There was a problem hiding this comment.
这里为什么后面还有 + i * shard_len * sample_size的？
DesmonDay
commented
Sep 2, 2022
| @@ -1296,7 +1296,7 @@ std::pair<uint64_t, uint64_t> GraphTable::parse_node_file( | |||
| std::string parse_node_type = vals[0].to_string(); | |||
| auto it = feature_to_id.find(parse_node_type); | |||
| if (it == feature_to_id.end()) { | |||
| VLOG(0) << parse_node_type << "type error, please check"; | |||
| VLOG(1) << parse_node_type << "type error, please check"; | |||
There was a problem hiding this comment.
改这个是因为如果把node_type或者edge_type减少一些的话，有些type找不到，这里就会疯狂打log。
huwei02
reviewed
Sep 2, 2022
| @@ -1296,7 +1296,7 @@ std::pair<uint64_t, uint64_t> GraphTable::parse_node_file( | |||
| std::string parse_node_type = vals[0].to_string(); | |||
| auto it = feature_to_id.find(parse_node_type); | |||
| if (it == feature_to_id.end()) { | |||
| VLOG(0) << parse_node_type << "type error, please check"; | |||
| VLOG(1) << parse_node_type << "type error, please check"; | |||
There was a problem hiding this comment.
这个主要是因为，如果我在测试删掉某个类型来跑（比如node_types和edge_types删掉一些），这块地方就会疯狂打很多很多log，打了很久log之后才开始训练。所以觉得降低log等级会比较好。
huwei02
reviewed
Sep 2, 2022
paddle/fluid/framework/data_feed.cu
Outdated
| auto gpu_graph_ptr = GraphGpuWrapper::GetInstance(); | ||
| auto edge_to_id = gpu_graph_ptr->edge_to_id; | ||
| int64_t all_sample_size = 0; | ||
|
|
||
| /* |
huwei02
reviewed
Sep 2, 2022
| @@ -293,6 +293,25 @@ struct NeighborSampleResult { | |||
| ~NeighborSampleResult() {} | |||
| }; | |||
|
|
|||
| struct NeighborSampleResultV2 { | |||
There was a problem hiding this comment.
为啥不直接掉NeighborSampleResult，而要重新弄一个NeighborSampleResultV2呢？
There was a problem hiding this comment.
这个也是我正在尝试的优化代码。弄成V2是因为分配的空间大小也不一样了，这里输入了一个新的参数叫edge_type_len。
seemingwang
approved these changes
Sep 5, 2022
lxsbupt
pushed a commit
to lxsbupt/Paddle
that referenced
this pull request
Nov 29, 2022
* add sample update, update reindex * temp commit * fix offset calc * optimize reindex * delete unnecessary code * add for loop speed up * add speed optimize * change common_graph VLOG type * add kernel3 update for review * temp commit * delete unused code
lxsbupt
pushed a commit
to lxsbupt/Paddle
that referenced
this pull request
Dec 17, 2022
* add sample update, update reindex * temp commit * fix offset calc * optimize reindex * delete unnecessary code * add for loop speed up * add speed optimize * change common_graph VLOG type * add kernel3 update for review * temp commit * delete unused code
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
Performance optimization
PR changes
Others
Describe
Optimize graphsage speed