[inductor] add a config to specify the shape attribute for the generated svg graphs #114811
Conversation
…ted svg graphs We draw our fx graphs with the "record" shape attribute by default. Sometimes, when the graph is very complex, we may hit dot errors like below: "flat edge between adjacent nodes one of which has a record shape - replace records with HTML-like labels" and thus fail to generate a graph. So, let's give the user an option to specify the shape attribute for the dot graph. For example, passing INDUCTOR_DOT_GRAPH_SHAPE_SVG = "none" would let us generate HTML-like lables to workaround the above failure. [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/114811
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 778b8f0 with merge base 0ced55e ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
…ted svg graphs We draw our fx graphs with the "record" shape attribute by default. Sometimes, when the graph is very complex, we may hit dot errors like below: "flat edge between adjacent nodes one of which has a record shape - replace records with HTML-like labels" and thus fail to generate a graph. So, let's give the user an option to specify the shape attribute for the dot graph. For example, passing INDUCTOR_DOT_GRAPH_SHAPE_SVG = "none" would let us generate HTML-like lables to workaround the above failure. ghstack-source-id: e287acd Pull Request resolved: #114811
There was a problem hiding this comment.
thanks for the fix. I met the same error before and shape=record seem to complain a lot when drawing siblings. there are some CIs in progress but I assume it's good to land if CI passed.
there are some subclass in fbcode inherits FxGraphDrawer. since we are adding a new param, you might need to be keep an eye on the diff train if it breaks fbcode CI. I did a similar fix before D49749670)
Thanks. Will keep an eye on those internal cases. |
|
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
…ted svg graphs (pytorch#114811) We draw our fx graphs with the "record" shape attribute by default. Sometimes, when the graph is very complex, we may hit dot errors like below: "flat edge between adjacent nodes one of which has a record shape - replace records with HTML-like labels" and thus fail to generate a graph. So, let's give the user an option to specify the shape attribute for the dot graph. For example, passing INDUCTOR_DOT_GRAPH_SHAPE_SVG = "none" would let us generate HTML-like lables to workaround the above failure. Pull Request resolved: pytorch#114811 Approved by: https://github.com/weifengpy
Stack from ghstack (oldest at bottom):
We draw our fx graphs with the "record" shape attribute by default.
Sometimes, when the graph is very complex, we may hit dot errors like below:
"flat edge between adjacent nodes one of which has a record shape -
replace records with HTML-like labels"
and thus fail to generate a graph. So, let's give the user an option
to specify the shape attribute for the dot graph. For example, passing
INDUCTOR_DOT_GRAPH_SHAPE_SVG = "none" would let us generate HTML-like lables
to workaround the above failure.
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @peterbell10 @ipiszy @yf225 @kadeng @muchulee8 @aakhundov @ColinPeppler