[PUBLISHER] Merge #43

* PUSH NOTE : PyTorch Conference 2024.md

* PUSH NOTE : UCLA.md

* PUSH NOTE : PyTorch.md

* PUSH NOTE : Meta.md

* PUSH NOTE : Jesse Cai.md

* PUSH NOTE : FAIR.md

* PUSH NOTE : Representation (Group Theory).md

* PUSH NOTE : PyTorch Quantization for TensorRT.md

docs/000 Zettelkasten/PyTorch Quantization for TensorRT.md

@@ -6,7 +6,17 @@ share: true
 ---
 
 There seems to be quite a few possible ways to do this:
-- [[PyTorch Eager Mode Quantization TensorRT Acceleration|PyTorch Eager Mode Quantization TensorRT Acceleration]] 
-	- 1. torchao quantization 2. ONNX conversion 3. Graph Surgery (changing some ops in the onnx graph) 4.. tensorrt conversion
-	- Seems very cumbersome
-- 
+- [[PyTorch Eager Mode Quantization TensorRT Acceleration|PyTorch Eager Mode Quantization TensorRT Acceleration]] , seems a bit cumbersome:
+	1. torchao quantization 
+	2. ONNX conversion
+	3. Graph Surgery (changing some ops in the onnx graph)
+	4. tensorrt conversion
+- Not sure if it works, but would be ideal
+	1. torch.export
+	2. torchao quantization
+	3. tensorrt conversion
+- Less ideal would be:
+	1. torchao quantization
+	2. torch.export
+	3. tensorrt conversion
+	- I've already sort of tried this using the [vgg ptq example from tensorrt](https://github.com/pytorch/TensorRT/blob/main/examples/dynamo/vgg16_ptq.py), but torch.export complained that it couldn't translate the quantized operations

docs/000 Zettelkasten/Representation (Group Theory).md

@@ -6,11 +6,13 @@ tags:
 ---
 
 Property required:
+
 $$
 p(g)p(h) = p(g \cdot h)
 $$
 
 A representation of a group action can be a linear operator like:
+
 $$
 p(\theta) = [sin(\theta) ...]
 $$

docs/100 Reference notes/102 Authors/Jesse Cai.md

@@ -0,0 +1,7 @@
+---
+affiliation:
+  - "[[Meta|Meta]]"
+  - "[[UCLA|UCLA]]"
+  - "[[PyTorch|PyTorch]]"
+share: true
+---

docs/100 Reference notes/103 Affiliations/FAIR.md

@@ -1,3 +1,4 @@
 ---
 share: true
 ---
+Related: [[FAIR|FAIR]]

docs/100 Reference notes/103 Affiliations/Meta.md

@@ -0,0 +1,4 @@
+---
+share: true
+---
+

docs/100 Reference notes/103 Affiliations/PyTorch.md

@@ -0,0 +1,3 @@
+---
+share: true
+---

docs/100 Reference notes/103 Affiliations/UCLA.md

@@ -0,0 +1,3 @@
+---
+share: true
+---

docs/100 Reference notes/104 Other/PyTorch Conference 2024.md

@@ -0,0 +1,11 @@
+---
+authors: 
+year: 2024
+tags:
+  - conference
+url: 
+share: true
+---
+Some interesting talks for #efficient_dl :
+- [[PyTorch Conference 2024 - What’s new in torch.export?|PyTorch Conference 2024 - What’s new in torch.export?]]
+- [[PyTorch Conference 2024 - Fast Sparse Vision Transformers with minimal accuracy loss|PyTorch Conference 2024 - Fast Sparse Vision Transformers with minimal accuracy loss]]