Debugging story: The case of the garbage text generation
|
|
0
|
2042
|
March 30, 2023
|
How is triton_mm called
|
|
1
|
710
|
March 29, 2023
|
TorchInductor CPP Backend Vectorization Status Analysis
|
|
0
|
994
|
March 27, 2023
|
How to get ir.ExternKernel instance to initialize inductor Scheduler?
|
|
1
|
502
|
March 24, 2023
|
PrimTorch: decompose ATen ops
|
|
1
|
1396
|
March 23, 2023
|
How do we do mid layer integration after Aten fx graph
|
|
4
|
826
|
March 22, 2023
|
PrimTorch: could we get pure core-aten-ops or prims-ops after aot_autograd
|
|
6
|
4646
|
March 21, 2023
|
Registering new compiler backend in Pytorch2.0
|
|
5
|
2265
|
March 20, 2023
|
Partial graph allocation for accelerators
|
|
3
|
624
|
March 13, 2023
|
TorchInductor Update 5: CPU backend backend performance update and deep dive on key optimizations
|
|
0
|
3336
|
March 9, 2023
|
TorchDynamo: An Experiment in Dynamic Python Bytecode Transformation
|
|
7
|
17463
|
March 9, 2023
|
What is the recommend serialization format when considering the upcoming pt2?
|
|
1
|
793
|
March 6, 2023
|
How to customize tracing granularity for GraphAppendingTracer?
|
|
0
|
547
|
February 13, 2023
|
TorchDynamo Update 3: GPU Inference Edition
|
|
12
|
6764
|
February 2, 2023
|
Tracing with Primitives: Update 2
|
|
4
|
7001
|
January 13, 2023
|
PyTorch/XLA 2022 Q4 Dev update
|
|
0
|
3351
|
January 5, 2023
|
PyTorch 2.0 Manifesto and Architecture docs
|
|
4
|
1677
|
December 11, 2022
|
The nuances of PyTorch Graph Capture
|
|
9
|
16201
|
December 9, 2022
|
TorchDynamo Update 7: Inference with FX2TRT
|
|
2
|
3542
|
December 8, 2022
|
TorchInductor Update 4: CPU backend started to show promising performance boost
|
|
1
|
2958
|
November 25, 2022
|
Dynamo/FX: patching a function to add more outputs not working
|
|
2
|
731
|
November 21, 2022
|
Where we are headed and why it looks a lot like Julia (but not exactly like Julia)
|
|
8
|
37686
|
November 14, 2022
|
Skipping Dispatcher with LazyTensor
|
|
10
|
1622
|
October 19, 2022
|
TorchInductor Update 3: E2E model training with TorchDynamo + Inductor gets 1.67x/2.1x speedup
|
|
3
|
2928
|
October 11, 2022
|
Reducing Framework Overhead with Static Runtime
|
|
4
|
2359
|
October 4, 2022
|
Tracing with Primitives: Update 0
|
|
17
|
8397
|
September 26, 2022
|
NNC walkthrough: how PyTorch ops get fused
|
|
10
|
7378
|
November 3, 2021
|
TorchDynamo Update 8: TorchDynamo passed correctness check on 7k+ github models
|
|
7
|
6345
|
July 1, 2022
|
Python Operator Authoring w/ NNC
|
|
5
|
2511
|
June 7, 2022
|
Tracing with Primitives: Update 1, nvFuser and its Primitives
|
|
0
|
6199
|
April 25, 2022
|