[PaddleSlim] add skip_quant test in doc (#3719)

PaddleSlim/classification/models/resnet.py

@@ -44,6 +44,8 @@ class ResNet():
 
         # TODO(wanghaoshuang@baidu.com):
         # fix name("conv1") conflict between student and teacher in distillation.
+        
+        #with fluid.name_scope('skip_quant'):
         conv = self.conv_bn_layer(
             input=input,
             num_filters=64,
@@ -51,8 +53,7 @@ class ResNet():
             stride=2,
             act='relu',
             name=prefix_name + conv1_name)
-        with fluid.name_scope("skip_quant"):
-            conv = fluid.layers.pool2d(
+        conv = fluid.layers.pool2d(
                 input=conv,
                 pool_size=3,
                 pool_stride=2,

PaddleSlim/classification/quantization/README.md

@@ -159,6 +159,38 @@ python infer.py \
 ### PaddleLite预测
 FP32模型可使用Paddle-Lite进行加载预测，可参见教程[Paddle-Lite如何加载运行量化模型](https://github.com/PaddlePaddle/Paddle-Lite/wiki/model_quantization)。
 
+## 如何进行部分量化
+
+通过在定义op时指定 ``name_scope``为 ``skip_quant``可对这个op跳过量化。比如在<a href="../models/resnet.py">PaddleSlim/classification/models/resnet.py</a>中，将某个conv的定义作如下改变：
+
+原定义：
+```
+conv = self.conv_bn_layer(
+                input=input,
+                num_filters=64,
+                filter_size=7,
+                stride=2,
+                act='relu',
+                name=prefix_name + conv1_name)
+
+```
+
+跳过量化时的定义：
+
+```
+
+with fluid.name_scope('skip_quant'):
+    conv = self.conv_bn_layer(
+                input=input,
+                num_filters=64,
+                filter_size=7,
+                stride=2,
+                act='relu',
+                name=prefix_name + conv1_name)
+
+```
+在脚本 <a href="./compress.py">PaddleSlim/classification/quantization/compress.py</a> 中，统计了``conv`` op的数量和以``fake_quantize``开头的量化op的数量，在对一些``conv`` op跳过之后，可发现以``fake_quantize``开头的量化op的数量变少。
+
 
 ## 示例结果
 

PaddleSlim/classification/quantization/compress.py

@@ -99,6 +99,16 @@ def compress(args):
         distiller_optimizer=None)
     com_pass.config(args.config_file)
     com_pass.run()
+    conv_op_num = 0
+    fake_quant_op_num = 0
+    for op in com_pass.context.eval_graph.ops():
+        if op._op.type == 'conv2d':
+            conv_op_num += 1
+        elif op._op.type.startswith('fake_quantize'):
+            fake_quant_op_num += 1
+    print('conv op num {}'.format(conv_op_num))
+    print('fake quant op num {}'.format(fake_quant_op_num))
+
 
 
 def main():