From 435e9a33dcc1b2a3c73a9d066f5751380bfc9433 Mon Sep 17 00:00:00 2001
From: Manix <50542248+manickavela29@users.noreply.github.com>
Date: Fri, 5 Apr 2024 10:08:20 +0530
Subject: [PATCH] Update quantization.md for  QDQ  and dynamic quantization old
 doc (#20130)

- doc suggests that QDQ model are created with dynamic quant, which is
not the case anymore.
- updating and restructuring the doc

### Description
<!-- Describe your changes. -->
- QDQ model format representation doesn't come for dynamic quantization,
but the doc was suggesting.
- May be a couple of years back this support was there but not anymore


### Motivation and Context
<!-- - Why is this change required? What problem does it solve?
- If it fixes an open issue, please link to the issue here. -->
- Clears misunderstanding that QDQ Onnx models can be created with
dynamic quantization.

https://github.com/microsoft/onnxruntime/issues/20125
---
 .../model-optimizations/quantization.md           | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

diff --git a/docs/performance/model-optimizations/quantization.md b/docs/performance/model-optimizations/quantization.md
index 0769e7f5bc..c769b0889f 100644
--- a/docs/performance/model-optimizations/quantization.md
+++ b/docs/performance/model-optimizations/quantization.md
@@ -38,11 +38,16 @@ Quantization in ONNX Runtime refers to 8 bit linear quantization of an ONNX mode
 
 ## ONNX quantization representation format
 There are two ways to represent quantized ONNX models:
-- Operator-oriented (QOperator). All the quantized operators have their own ONNX definitions, like QLinearConv, MatMulInteger and etc.
-- Tensor-oriented (QDQ; Quantize and DeQuantize). This format inserts DeQuantizeLinear(QuantizeLinear(tensor)) between the original operators to simulate the quantization and dequantization process. In Static Quantization, the QuantizeLinear and DeQuantizeLinear operators also carry the quantization parameters. In Dynamic Quantization, a ComputeQuantizationParameters function proto is inserted to calculate quantization parameters on the fly. Models generated in the following ways are in the QDQ format:
-  - Models quantized by quantize_static or quantize_dynamic API, explained below, with `quant_format=QuantFormat.QDQ`.
-  - Quantization-Aware training (QAT) models converted from Tensorflow or exported from PyTorch.
-  - Quantized models converted from TFLite and other frameworks.
+- Operator-oriented (QOperator) :  
+  All the quantized operators have their own ONNX definitions, like QLinearConv, MatMulInteger and etc.
+- Tensor-oriented (QDQ; Quantize and DeQuantize) :  
+  This format inserts DeQuantizeLinear(QuantizeLinear(tensor)) between the original operators to simulate the quantization and dequantization process.  
+  In Static Quantization, the QuantizeLinear and DeQuantizeLinear operators also carry the quantization parameters.  
+  In Dynamic Quantization, a ComputeQuantizationParameters function proto is inserted to calculate quantization parameters on the fly.
+- Models generated in the following ways are in the QDQ format:
+    1. Models quantized by quantize_static, explained below, with `quant_format=QuantFormat.QDQ`.
+    2. Quantization-Aware training (QAT) models converted from Tensorflow or exported from PyTorch.
+    3. Quantized models converted from TFLite and other frameworks.  
 
 For the latter two cases, you don't need to quantize the model with the quantization tool. ONNX Runtime can run them directly as a quantized model.