|
About the compiler category
|
|
0
|
221
|
January 22, 2021
|
|
Where do the 2000+ PyTorch operators come from?: More than you wanted to know
|
|
12
|
2286
|
May 12, 2022
|
|
Tracing with Primitives: Update 1, nvFuser and its Primitives
|
|
0
|
184
|
April 25, 2022
|
|
Tracing with Primitives: Update 0
|
|
14
|
551
|
April 18, 2022
|
|
TorchDynamo Update 5: Improved Capture & Bigger Graphs
|
|
4
|
381
|
April 14, 2022
|
|
TorchDynamo Update 7: Inference with FX2TRT
|
|
0
|
159
|
April 8, 2022
|
|
TorchDynamo Update 6: Training support with AOTAutograd
|
|
0
|
312
|
March 29, 2022
|
|
Where we are headed and why it looks a lot like Julia (but not exactly like Julia)
|
|
5
|
25201
|
March 28, 2022
|
|
Universal binaries for libtorch Mac?
|
|
10
|
926
|
March 14, 2022
|
|
TorchDynamo Update 4: LazyTensor & nvFuser Experiments
|
|
2
|
628
|
February 25, 2022
|
|
The nuances of PyTorch Graph Capture
|
|
0
|
686
|
February 16, 2022
|
|
Python Operator Authoring w/ NNC
|
|
4
|
874
|
February 4, 2022
|
|
Min-cut optimal(*) recomputation (i.e. activation checkpointing) with AOTAutograd
|
|
6
|
1552
|
January 20, 2022
|
|
TorchDynamo: An Experiment in Dynamic Python Bytecode Transformation
|
|
4
|
2081
|
January 19, 2022
|
|
TorchDynamo Update 3: GPU Inference Edition
|
|
5
|
878
|
January 10, 2022
|
|
Prim::autocast_promote operation?
|
|
4
|
351
|
December 31, 2021
|
|
TorchDynamo Update: 1.48x geomean speedup on TorchBench CPU Inference
|
|
0
|
916
|
November 12, 2021
|
|
NNC walkthrough: how PyTorch ops get fused
|
|
10
|
1193
|
November 3, 2021
|
|
Next Steps for PyTorch Compilers
|
|
9
|
2896
|
October 21, 2021
|
|
Loop_tool's lazy frontend - Experimenting with symbolic laziness
|
|
0
|
177
|
October 11, 2021
|
|
torch-MLIR presentation from Google / nod.ai
|
|
1
|
354
|
October 8, 2021
|
|
Depthwise conv2d: An NNC Case Study
|
|
0
|
483
|
April 7, 2021
|
|
Symbolic Shape Inference
|
|
1
|
364
|
March 31, 2021
|
|
Summary of Issues Found in Adapting Models to use TorchScript
|
|
1
|
321
|
March 16, 2021
|
|
ExecutionPlan caching in ProfilingGraphExecutorImpl?
|
|
0
|
197
|
March 3, 2021
|
|
Adapting Models to use TorchScript and Getting them to Produce Fusions
|
|
15
|
1440
|
February 26, 2021
|
|
JIT scripting & Autocast
|
|
11
|
906
|
February 23, 2021
|
|
Single-op fusion benchmarking
|
|
0
|
271
|
February 4, 2021
|
|
NNC Per-Operator Benchmarks (on CPU)
|
|
5
|
307
|
January 27, 2021
|
|
Reducing Framework Overhead with Static Runtime
|
|
1
|
565
|
January 27, 2021
|