add mixed precision part on tensorflow.md

Signed-off-by: zehao-intel <[email protected]>

docs/source/3x/TensorFlow.md

@@ -2,12 +2,16 @@ TensorFlow
 ===============
 
 
-1. [Introduction](#introduction)
-2. [API for TensorFlow](#api-for-tensorflow)
-3. [Support Matrix](#support-matrix)  
-   3.1 [Quantization Scheme](#quantization-scheme)   
-   3.2 [Quantization Approaches](#quantization-approaches)   
-   3.3 [Backend and Device](#backend-and-device)   
+- [TensorFlow](#tensorflow)
+  - [Introduction](#introduction)
+  - [API for TensorFlow](#api-for-tensorflow)
+    - [Support Matrix](#support-matrix)
+      - [Quantization Scheme](#quantization-scheme)
+      - [Quantization Approaches](#quantization-approaches)
+        - [Post Training Static Quantization](#post-training-static-quantization)
+        - [Smooth Quantization](#smooth-quantization)
+        - [Mixed Precison](#mixed-precison)
+      - [Backend and Device](#backend-and-device)
 
 ## Introduction
 
@@ -152,9 +156,16 @@ The supported Quantization methods for TensorFlow and Keras are listed below:
             <td align="center"><a href="https://github.com/tensorflow/tensorflow">TensorFlow</a>/<a href="https://github.com/Intel-tensorflow/tensorflow">Intel TensorFlow</a></td>
         </tr>
         <tr>
-            <td rowspan="2" align="center">Smooth Quantization(SQ)</td>
-            <td rowspan="2" align="center">weights</td>
-            <td rowspan="2" align="center">calibration</td>
+            <td rowspan="1" align="center">Smooth Quantization(SQ)</td>
+            <td rowspan="1" align="center">weights</td>
+            <td rowspan="1" align="center">calibration</td>
+            <td align="center">Tensorflow</td>
+            <td align="center"><a href="https://github.com/tensorflow/tensorflow">TensorFlow</a>/<a href="https://github.com/Intel-tensorflow/tensorflow">Intel TensorFlow</a></td>
+        </tr>
+        <tr>
+            <td rowspan="1" align="center">Mixed Precision(MP)</td>
+            <td rowspan="1" align="center">weights and activations</td>
+            <td rowspan="1" align="center">NA</td>
             <td align="center">Tensorflow</td>
             <td align="center"><a href="https://github.com/tensorflow/tensorflow">TensorFlow</a>/<a href="https://github.com/Intel-tensorflow/tensorflow">Intel TensorFlow</a></td>
         </tr>
@@ -175,6 +186,10 @@ Smooth Quantization (SQ) is an advanced quantization technique designed to optim
 
 Refer to the [SQ Guide](./TF_SQ.md) for detailed information.
 
+##### Mixed Precison
+The Mixed Precision (MP) is enabled with Post Training Static Quantization. Once `BF16` is supported on machine, the matched operators will be automatically converted.
+
+
 #### Backend and Device
 Intel(R) Neural Compressor supports TF GPU with [ITEX-XPU](https://github.com/intel/intel-extension-for-tensorflow). We will automatically run model on GPU by checking if it has been installed.
 

neural_compressor/tensorflow/keras/quantization/config.py

@@ -138,7 +138,7 @@ def get_model_info(model) -> List[Tuple[str, Callable]]:
 
     @classmethod
     def get_config_set_for_tuning(cls) -> Union[None, "StaticQuantConfig", List["StaticQuantConfig"]]:
-        """get a default config set for tunning."""
+        """Get a default config set for tunning."""
         return StaticQuantConfig(weight_sym=[True, False])