add tensor subclass printing support in fx/graph.py#164403
Closed
bobrenjc93 wants to merge 6 commits intogh/bobrenjc93/623/basefrom
Closed
add tensor subclass printing support in fx/graph.py#164403bobrenjc93 wants to merge 6 commits intogh/bobrenjc93/623/basefrom
bobrenjc93 wants to merge 6 commits intogh/bobrenjc93/623/basefrom
Conversation
it was previously quite misleading since it looks like the inputs to the
dynamo graph are plain tensors when in reality they are dtensors
before
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "i64[2, 512][512, 1]cuda:0", L_self_parameters_weight_: "f32[202048, 256][256, 1]cuda:0"):
```
after
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "DTensor(dtype=i64, shape=[2, 512], device=cuda:0, mesh=DeviceMesh((dim1=8, dim2=2, dim3=2), device: 'cuda', stride: (4, 2, 1)), placements=(Replicate(), Replicate(), Replicate()))", L_self_parameters_weight_: "DTensor(dtype=f32, shape=[202048, 256], device=cuda:0, mesh=DeviceMesh((dim1=8, dim2=2, dim3=2), device: 'cuda', stride: (4, 2, 1)), placements=(Replicate(), Replicate(), Replicate()))"):
```
[ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/164403
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit f6fc5cb with merge base dfda239 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This was referenced Oct 1, 2025
bdhirsh
reviewed
Oct 1, 2025
bobrenjc93
changed the title
it was previously quite misleading since it looks like the inputs to the
dynamo graph are plain tensors when in reality they are tensor subclasses
before
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "i64[2, 512][512, 1]cuda:0", L_self_parameters_weight_: "f32[202048, 256][256, 1]cuda:0"):
```
after
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "DTensor(i64[2, 512][512, 1]cuda:0)", L_self_parameters_weight_: "DTensor(f32[202048, 256][256, 1]cuda:0)"):
```
cc ezyang EikanWang jgong5 wenzhe-nrv
[ghstack-poisoned]
bobrenjc93
added a commit
that referenced
this pull request
Oct 1, 2025
it was previously quite misleading since it looks like the inputs to the
dynamo graph are plain tensors when in reality they are tensor
subclasses
before
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "i64[2, 512][512, 1]cuda:0", L_self_parameters_weight_: "f32[202048, 256][256, 1]cuda:0"):
```
after
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "DTensor(i64[2, 512][512, 1]cuda:0)", L_self_parameters_weight_: "DTensor(f32[202048, 256][256, 1]cuda:0)"):
```
ghstack-source-id: 4054d27
Pull Request resolved: #164403
Contributor
Author
|
updated the implementation for now to be more generic and check if it's a tensor subclass and if so just wrap the existing tensor descriptor with the subclass name. maybe in the future i'll extend this to have custom printers, but this solves the high level problem for now. |
it was previously quite misleading since it looks like the inputs to the
dynamo graph are plain tensors when in reality they are tensor subclasses
before
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "i64[2, 512][512, 1]cuda:0", L_self_parameters_weight_: "f32[202048, 256][256, 1]cuda:0"):
```
after
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "DTensor(i64[2, 512][512, 1]cuda:0)", L_self_parameters_weight_: "DTensor(f32[202048, 256][256, 1]cuda:0)"):
```
cc ezyang EikanWang jgong5 wenzhe-nrv
[ghstack-poisoned]
bobrenjc93
added a commit
that referenced
this pull request
Oct 1, 2025
it was previously quite misleading since it looks like the inputs to the
dynamo graph are plain tensors when in reality they are tensor
subclasses
before
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "i64[2, 512][512, 1]cuda:0", L_self_parameters_weight_: "f32[202048, 256][256, 1]cuda:0"):
```
after
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "DTensor(i64[2, 512][512, 1]cuda:0)", L_self_parameters_weight_: "DTensor(f32[202048, 256][256, 1]cuda:0)"):
```
ghstack-source-id: 0810646
Pull Request resolved: #164403
it was previously quite misleading since it looks like the inputs to the
dynamo graph are plain tensors when in reality they are tensor subclasses
before
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "i64[2, 512][512, 1]cuda:0", L_self_parameters_weight_: "f32[202048, 256][256, 1]cuda:0"):
```
after
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "DTensor(i64[2, 512][512, 1]cuda:0)", L_self_parameters_weight_: "DTensor(f32[202048, 256][256, 1]cuda:0)"):
```
cc ezyang EikanWang jgong5 wenzhe-nrv
[ghstack-poisoned]
bobrenjc93
added a commit
that referenced
this pull request
Oct 1, 2025
it was previously quite misleading since it looks like the inputs to the
dynamo graph are plain tensors when in reality they are tensor
subclasses
before
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "i64[2, 512][512, 1]cuda:0", L_self_parameters_weight_: "f32[202048, 256][256, 1]cuda:0"):
```
after
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "DTensor(i64[2, 512][512, 1]cuda:0)", L_self_parameters_weight_: "DTensor(f32[202048, 256][256, 1]cuda:0)"):
```
ghstack-source-id: 02236d5
Pull Request resolved: #164403
ezyang
reviewed
Oct 1, 2025
it was previously quite misleading since it looks like the inputs to the
dynamo graph are plain tensors when in reality they are tensor subclasses
before
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "i64[2, 512][512, 1]cuda:0", L_self_parameters_weight_: "f32[202048, 256][256, 1]cuda:0"):
```
after
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "DTensor(i64[2, 512][512, 1]cuda:0)", L_self_parameters_weight_: "DTensor(f32[202048, 256][256, 1]cuda:0)"):
```
cc ezyang EikanWang jgong5 wenzhe-nrv
[ghstack-poisoned]
bobrenjc93
added a commit
that referenced
this pull request
Oct 1, 2025
it was previously quite misleading since it looks like the inputs to the
dynamo graph are plain tensors when in reality they are tensor
subclasses
before
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "i64[2, 512][512, 1]cuda:0", L_self_parameters_weight_: "f32[202048, 256][256, 1]cuda:0"):
```
after
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "DTensor(i64[2, 512][512, 1]cuda:0)", L_self_parameters_weight_: "DTensor(f32[202048, 256][256, 1]cuda:0)"):
```
ghstack-source-id: 0d1cfaf
Pull Request resolved: #164403
ezyang
reviewed
Oct 2, 2025
ezyang
reviewed
Oct 2, 2025
Contributor
|
I'd like this to land but it needs more work |
bdhirsh
reviewed
Oct 2, 2025
it was previously quite misleading since it looks like the inputs to the
dynamo graph are plain tensors when in reality they are tensor subclasses
before
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "i64[2, 512][512, 1]cuda:0", L_self_parameters_weight_: "f32[202048, 256][256, 1]cuda:0"):
```
after
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "DTensor(i64[2, 512][512, 1]cuda:0)", L_self_parameters_weight_: "DTensor(f32[202048, 256][256, 1]cuda:0)"):
```
cc ezyang EikanWang jgong5 wenzhe-nrv voznesenskym penguinwu Guobing-Chen XiaobingSuper zhuhaozhe blzheng jiayisunx chenyang78 kadeng chauhang amjames Lucaskabela
[ghstack-poisoned]
bobrenjc93
added a commit
that referenced
this pull request
Oct 2, 2025
it was previously quite misleading since it looks like the inputs to the
dynamo graph are plain tensors when in reality they are tensor
subclasses
before
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "i64[2, 512][512, 1]cuda:0", L_self_parameters_weight_: "f32[202048, 256][256, 1]cuda:0"):
```
after
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "DTensor(i64[2, 512][512, 1]cuda:0)", L_self_parameters_weight_: "DTensor(f32[202048, 256][256, 1]cuda:0)"):
```
ghstack-source-id: a6fcc2d
Pull Request resolved: #164403
Contributor
Author
|
refactored PR to be more DRY |
ezyang
approved these changes
Oct 2, 2025
Contributor
|
@pytorchbot merge |
Collaborator
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 |
Chao1Han
pushed a commit
to Chao1Han/pytorch
that referenced
this pull request
Oct 21, 2025
it was previously quite misleading since it looks like the inputs to the
dynamo graph are plain tensors when in reality they are tensor subclasses
before
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "i64[2, 512][512, 1]cuda:0", L_self_parameters_weight_: "f32[202048, 256][256, 1]cuda:0"):
```
after
```
class GraphModule(torch.nn.Module):
def forward(self, L_input_batch_inputs_: "DTensor(i64[2, 512][512, 1]cuda:0)", L_self_parameters_weight_: "DTensor(f32[202048, 256][256, 1]cuda:0)"):
```
Pull Request resolved: pytorch#164403
Approved by: https://github.com/ezyang
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 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.
Stack from ghstack (oldest at bottom):
it was previously quite misleading since it looks like the inputs to the
dynamo graph are plain tensors when in reality they are tensor subclasses
before
after
cc @ezyang @EikanWang @jgong5 @wenzhe-nrv @voznesenskym @penguinwu @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @jiayisunx @chenyang78 @kadeng @chauhang @amjames @Lucaskabela