msr-fiddle pipedream issues

msr-fiddle / pipedream

MIT License

376 stars 117 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Can you tell me how to run this code with the RTX 4090D? How should I configure it and run this patch?

#81 hnust-xxq opened 1 week ago
0
same train_loader but got different loader size

#80 Hyaloid closed 10 months ago
2
optimizer got an empty parameter list when rank=1

#79 Hyaloid closed 10 months ago
1
When I was testing the pipedream code with version-updated torch, I encountered the following error (1.1.0 -> 1.11.0):

#78 lengien opened 1 year ago
9
what is the role of pre_hook_pytorch_latest.patch？

#77 matrix97317 opened 1 year ago
1
Running in docker will give you an error that you can't find a physical address

#76 guanyonglai opened 1 year ago
1
AttributeError: module 'torch.distributed' has no attribute 'P2POp'

#75 guanyonglai closed 1 year ago
1
Is there any 2bw code that will run on the native GPU

#74 guanyonglai closed 1 year ago
1
AttributeError: module 'models.resnet50.resnet50' has no attribute 'model'

#73 guanyonglai closed 1 year ago
1
CVE-2007-4559 Patch

#72 TrellixVulnTeam opened 2 years ago
0
Question about PipeDream's optimizer

#71 lllukehuang opened 2 years ago
0
Question about time complexity of PipeDream-2BW's planner algorithm

#70 ConnollyLeon opened 2 years ago
0
The arguments of self.start_helper_thread() should be more flexible instead of fixed as int64.

#69 gouchangjiang opened 2 years ago
0
Supporting T5

#68 gperrotta closed 1 year ago
0
modify runtime/image_classification/models/resnet50/gpus=2/__init__.py

#67 SHEELE41 opened 3 years ago
0
How is the Double-Buffered Weight Mechanism implemented?

#66 BinhangYuan opened 3 years ago
0
Is there AllReduce in data parallelism?

#65 Allen-Czyysx closed 3 years ago
6
GPT2 355m model convergence with 2BW training

#64 nitikasaran68 opened 3 years ago
0
The BLEU score of translation model seems abnormal. The model doesn't seem to train effectively.

#63 njuyexiangyu opened 3 years ago
0
To run PipeDream_2BW branch without --recompute_step

#62 Shigangli closed 3 years ago
0
Resource temporarily unavailable

#61 liulixinkerry closed 3 years ago
0
GPU Peer2Peer communication via --num_ranks_in_server argument

#60 siddharth9820 opened 3 years ago
1
Handling uneven number of batches per replicated instance of a layer

#59 siddharth9820 opened 3 years ago
0
Running a transformer module

#58 oranichu closed 4 years ago
0
Planner for PipeDream-2BW

#57 nict-wisdom opened 4 years ago
5
Communication error when training with Pipedream

#56 gudiandian closed 4 years ago
6
Hanging with [4,3,1] GPU assignment

#55 BestSonny closed 4 years ago
0
Can the profiler handle dynamic graphs?

#54 rahul003 opened 4 years ago
0
What is the meaning of `antichain` in optimizer_graph_hierarchical.py ?

#53 sergei-mironov opened 4 years ago
0
What's the latest version of PyTorch supported?

#52 SimonZsx opened 4 years ago
10
How to expose the "register_pre_hook()" interface?

#51 letian-zhang opened 4 years ago
1
Actual results did not match the optimizer expectation

#50 nirandaperera opened 4 years ago
1
Translation demo: Division by zero

#49 sergei-mironov opened 4 years ago
4
Translation demo: Installation instrutions issue. Missing CUDA kernels.

#48 sergei-mironov closed 4 years ago
3
Batch size and optimizer

#47 nirandaperera closed 4 years ago
6
Is this the version used in SOSP paper?

#46 nirandaperera closed 4 years ago
1
Infinite loop problem in convert_graph_to_model.py

#45 jayhpark530 opened 4 years ago
3
Error occurred in profiling

#44 gudiandian closed 4 years ago
3
Some error about communication

#43 jglicat closed 10 months ago
4
docker pull error

#42 cnzhanj closed 4 years ago
1
I can't get the same result as your model after segmentation

#41 ADAM-CT opened 4 years ago
2
whats the role of optimizer/inference_optimizer_graph.py

#40 ADAM-CT closed 4 years ago
1
All ranks are not trained. They are blocked all the time

#39 ADAM-CT opened 4 years ago
3
"stage_to_depth_map" not found

#38 ADAM-CT closed 4 years ago
2
How to determine replication factors

#37 ADAM-CT opened 4 years ago
4
Multi node training

#36 ADAM-CT opened 4 years ago
6
bandwidth parameter

#35 ADAM-CT opened 4 years ago
2
Multi-machine distribution problem

#34 ADAM-CT opened 4 years ago
6
I would like to know what the role of the following driver.py is？

#33 ADAM-CT opened 4 years ago
2
RuntimeError: [enforce fail at ../third_party/gloo/gloo/transport/tcp/device.cc:127] rp != nullptr. Unable to find address for: dgx-1.ai

#32 ADAM-CT closed 4 years ago
3