add pp-yoloe quantization model (#5894)

configs/slim/README.md

@@ -126,6 +126,8 @@ python tools/export_model.py -c configs/{MODEL.yml} --slim_config configs/slim/{
 
 | 模型               | 压缩策略     | 输入尺寸 |  模型体积(MB) | 预测时延(V100) | 预测时延(SD855) |  Box AP    |                  下载         |   Inference模型下载    |            模型配置文件                  |                     压缩算法配置文件                   |
 | ------------------ | ------------ | -------- | :---------: | :---------: |:---------: | :---------: | :----------------------------------------------: | :----------------------------------------------: |:------------------------------------------: | :------------------------------------: |
+| PP-YOLOE-l | baseline     | 640      | - |   11.2ms(trt_fp32) | 7.7ms(trt_fp16)  | -- | 50.9     | [下载链接](https://paddledet.bj.bcebos.com/models/ppyoloe_crn_l_300e_coco.pdparams) | -  | [配置文件](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/ppyoloe/ppyoloe_crn_l_300e_coco.yml) | -  |
+| PP-YOLOE-l | 普通在线量化     | 640      | - |   6.7ms(trt_int8)  | -- | 48.8     | [下载链接](https://paddledet.bj.bcebos.com/models/slim/ppyoloe_l_coco_qat.pdparams) | -  | [配置文件](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/ppyoloe/ppyoloe_crn_l_300e_coco.yml) | [配置文件](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/slim/quant/ppyoloe_l_qat.yml) |
 | PP-YOLOv2_R50vd | baseline     | 640      | 208.6  |   19.1ms  | -- | 49.1     | [下载链接](https://paddledet.bj.bcebos.com/models/ppyolov2_r50vd_dcn_365e_coco.pdparams) | [下载链接](https://paddledet.bj.bcebos.com/models/slim/ppyolov2_r50vd_dcn_365e_coco.tar)  | [配置文件](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/ppyolo/ppyolo_r50vd_dcn_1x_coco.yml) |                            -                               |
 | PP-YOLOv2_R50vd | PACT在线量化     | 640      | --  |   17.3ms  | -- | 48.1     | [下载链接](https://paddledet.bj.bcebos.com/models/slim/ppyolov2_r50vd_dcn_qat.pdparams) | [下载链接](https://paddledet.bj.bcebos.com/models/slim/ppyolov2_r50vd_dcn_qat.tar)  | [配置文件](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/ppyolo/ppyolo_r50vd_dcn_1x_coco.yml) |         [配置文件](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/slim/quant/ppyolov2_r50vd_dcn_qat.yml)   |
 | PP-YOLO_R50vd | baseline     | 608      | 183.3  |  17.4ms  | -- | 44.8     | [下载链接](https://paddledet.bj.bcebos.com/models/ppyolo_r50vd_dcn_1x_coco.pdparams) | [下载链接](https://paddledet.bj.bcebos.com/models/slim/ppyolo_r50vd_dcn_1x_coco.tar)  | [配置文件](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/ppyolo/ppyolo_r50vd_dcn_1x_coco.yml) |                  -         |
@@ -146,6 +148,7 @@ python tools/export_model.py -c configs/{MODEL.yml} --slim_config configs/slim/{
 说明：
 - 上述V100预测时延非量化模型均是使用TensorRT-FP32测试，量化模型均使用TensorRT-INT8测试，并且都包含NMS耗时。
 - SD855预测时延为使用PaddleLite部署，使用arm8架构并使用4线程(4 Threads)推理时延。
+- 上述PP-YOLOE模型均在V100，开启TensorRT环境中测速，不包含NMS。（导出模型时指定：-o trt=True exclude_nms=True）
 
 ### 离线量化
 需要准备val集，用来对离线量化模型进行校准，运行方式：

configs/slim/README_en.md

@@ -124,7 +124,9 @@ Description:
 
 | Model                     | Compression Strategy       | Input Size  | Model Volume(MB) | Prediction Delay(V100) | Prediction Delay(SD855) |        Box AP         |                                          Download                                           |                                 Download of Inference Model                                 |                                                          Model Configuration File                                                          |                                                 Compression Algorithm Configuration File                                                 |
 | ------------------------- | -------------------------- | ----------- | :--------------: | :--------------------: | :---------------------: | :-------------------: | :-----------------------------------------------------------------------------------------: | :-----------------------------------------------------------------------------------------: | :----------------------------------------------------------------------------------------------------------------------------------------: | :--------------------------------------------------------------------------------------------------------------------------------------: |
-| PP-YOLOv2_R50vd           | baseline                   | 640         |      208.6       |         19.1ms         |           --            |         49.1          |    [link](https://paddledet.bj.bcebos.com/models/ppyolov2_r50vd_dcn_365e_coco.pdparams)     |    [link](https://paddledet.bj.bcebos.com/models/slim/ppyolov2_r50vd_dcn_365e_coco.tar)     |      [Configuration File ](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/ppyolo/ppyolo_r50vd_dcn_1x_coco.yml)       |                                                                    -                                                                     |
+| PP-YOLOE-l | baseline     | 640      | - |   11.2ms(trt_fp32) | 7.7ms(trt_fp16)   | -- | 50.9     | [link](https://paddledet.bj.bcebos.com/models/ppyoloe_crn_l_300e_coco.pdparams) | -  | [Configuration File](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/ppyoloe/ppyoloe_crn_l_300e_coco.yml) |  - |
+| PP-YOLOE-l | Common Online quantitative     | 640      | - |   6.7ms(trt_int8)  | -- | 48.8     | [link](https://paddledet.bj.bcebos.com/models/slim/ppyoloe_l_coco_qat.pdparams) | -  | [Configuration File](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/ppyoloe/ppyoloe_crn_l_300e_coco.yml) | [Configuration File](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/slim/quant/ppyoloe_l_qat.yml) |
+| PP-YOLOv2_R50vd           | baseline                   | 640         |      208.6       |         19.1ms         |           --            |         49.1          |    [link](https://paddledet.bj.bcebos.com/models/ppyolov2_r50vd_dcn_365e_coco.pdparams)     |    [link](https://paddledet.bj.bcebos.com/models/slim/ppyolov2_r50vd_dcn_365e_coco.tar)     |      [Configuration File](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/ppyolo/ppyolo_r50vd_dcn_1x_coco.yml)       |                                                                    -                                                                     |
 | PP-YOLOv2_R50vd           | PACT Online quantitative   | 640         |        --        |         17.3ms         |           --            |         48.1          |     [link](https://paddledet.bj.bcebos.com/models/slim/ppyolov2_r50vd_dcn_qat.pdparams)     |       [link](https://paddledet.bj.bcebos.com/models/slim/ppyolov2_r50vd_dcn_qat.tar)        |      [Configuration File ](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/ppyolo/ppyolo_r50vd_dcn_1x_coco.yml)       |    [Configuration File ](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/slim/quant/ppyolov2_r50vd_dcn_qat.yml)     |
 | PP-YOLO_R50vd             | baseline                   | 608         |      183.3       |         17.4ms         |           --            |         44.8          |      [link](https://paddledet.bj.bcebos.com/models/ppyolo_r50vd_dcn_1x_coco.pdparams)       |      [link](https://paddledet.bj.bcebos.com/models/slim/ppyolo_r50vd_dcn_1x_coco.tar)       |      [Configuration File ](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/ppyolo/ppyolo_r50vd_dcn_1x_coco.yml)       |                                                                    -                                                                     |
 | PP-YOLO_R50vd             | PACT Online quantitative   | 608         |       67.3       |         13.8ms         |           --            |         44.3          |     [link](https://paddledet.bj.bcebos.com/models/slim/ppyolo_r50vd_qat_pact.pdparams)      |        [link](https://paddledet.bj.bcebos.com/models/slim/ppyolo_r50vd_qat_pact.tar)        |      [Configuration File ](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/ppyolo/ppyolo_r50vd_dcn_1x_coco.yml)       |     [Configuration File ](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/slim/quant/ppyolo_r50vd_qat_pact.yml)     |

configs/slim/quant/ppyoloe_l_qat.yml

+pretrain_weights: https://paddledet.bj.bcebos.com/models/ppyoloe_crn_l_300e_coco.pdparams
+slim: QAT
+
+QAT:
+  quant_config: {
+    'weight_quantize_type': 'channel_wise_abs_max', 'activation_quantize_type': 'moving_average_abs_max',
+    'weight_bits': 8, 'activation_bits': 8, 'dtype': 'int8', 'window_size': 10000, 'moving_rate': 0.9,
+    'quantizable_layer_type': ['Conv2D', 'Linear']}
+  print_model: True
+
+epoch: 30
+snapshot_epoch: 5
+LearningRate:
+  base_lr: 0.001
+  schedulers:
+  - !PiecewiseDecay
+    gamma: 0.1
+    milestones:
+    - 10
+    - 20
+  - !LinearWarmup
+    start_factor: 0.
+    steps: 100
+
+TrainReader:
+   batch_size: 8

ppdet/engine/trainer.py

@@ -701,9 +701,10 @@ class Trainer(object):
         if hasattr(self.model, 'deploy'):
             self.model.deploy = True
 
-        for layer in self.model.sublayers():
-            if hasattr(layer, 'convert_to_deploy'):
-                layer.convert_to_deploy()
+        if 'slim' not in self.cfg:
+            for layer in self.model.sublayers():
+                if hasattr(layer, 'convert_to_deploy'):
+                    layer.convert_to_deploy()
 
         export_post_process = self.cfg['export'].get(
             'post_process', False) if hasattr(self.cfg, 'export') else True

ppdet/modeling/heads/ppyoloe_head.py

@@ -111,6 +111,7 @@ class PPYOLOEHead(nn.Layer):
                     in_c, 4 * (self.reg_max + 1), 3, padding=1))
         # projection conv
         self.proj_conv = nn.Conv2D(self.reg_max + 1, 1, 1, bias_attr=False)
+        self.proj_conv.skip_quant = True
         self._init_weights()
 
     @classmethod

ppdet/slim/quant.py

@@ -38,6 +38,11 @@ class QAT(object):
             logger.info("Model before quant:")
             logger.info(model)
 
+        # For PP-YOLOE, convert model to deploy firstly.
+        for layer in model.sublayers():
+            if hasattr(layer, 'convert_to_deploy'):
+                layer.convert_to_deploy()
+
         self.quanter.quantize(model)
 
         if self.print_model: