Add Support for MMdetection RetinaNet&&FSAF&&SSD&&Faster R-CNN (#614)

* fix flatten op bug for retinanet&&fsaf

* fix RoiAlign op for fasterrcnn

* fix NMS op bugs for ap test

* update model_zoo.md

docs/introduction/x2paddle_model_zoo.md

@@ -73,7 +73,11 @@
 |GPT2| [pytorch(huggingface)](https://github.com/huggingface/transformers/blob/master/notebooks/04-onnx-export.ipynb)|11|转换时需指定input shape，见[文档Q3](../inference_model_convertor/FAQ.md)|
 |CifarNet | [tensorflow](https://github.com/tensorflow/models/blob/master/research/slim/nets/cifarnet.py)|9|
 |Fcos | [pytorch(mmdetection)](https://github.com/open-mmlab/mmdetection/blob/master/configs/fcos/fcos_r50_caffe_fpn_gn-head_1x_coco.py)|11|
-|Yolov3 | [pytorch(mmdetection)](https://github.com/open-mmlab/mmdetection/blob/master/configs/yolo/yolov3_d53_mstrain-608_273e_coco.py)|11||
+|Yolov3 | [pytorch(mmdetection)](https://github.com/open-mmlab/mmdetection/blob/master/configs/yolo/yolov3_d53_mstrain-608_273e_coco.py)|11|
+|RetinaNet | [pytorch(mmdetection)](https://github.com/open-mmlab/mmdetection/blob/master/configs/retinanet/retinanet_r50_fpn_1x_coco.py)|11|
+|FSAF | [pytorch(mmdetection)](https://github.com/open-mmlab/mmdetection/blob/master/configs/fsaf/fsaf_r50_fpn_1x_coco.py)|11|
+|SSD | [pytorch(mmdetection)](https://github.com/open-mmlab/mmdetection/blob/master/configs/ssd/ssd300_coco.py)|11|
+|Faster R-CNN | [pytorch(mmdetection)](https://github.com/open-mmlab/mmdetection/blob/master/configs/faster_rcnn/faster_rcnn_r50_fpn_1x_coco.py)|11||
 
 
 ## PyTorch预测模型

x2paddle/op_mapper/onnx2paddle/onnx_custom_layer/nms.py

@@ -83,18 +83,18 @@ def multiclass_nms(bboxes,
 
 
 class NMS(object):
-    def __init__(self, score_threshold, nms_top_k, nms_threshold):
+    def __init__(self, score_threshold, keep_top_k, nms_threshold):
         self.score_threshold = score_threshold
-        self.nms_top_k = nms_top_k
+        self.keep_top_k = keep_top_k
         self.nms_threshold = nms_threshold
 
     def __call__(self, bboxes, scores):
         attrs = {
             'background_label': -1,
             'score_threshold': self.score_threshold,
-            'nms_top_k': self.nms_top_k,
+            'nms_top_k': -1,
             'nms_threshold': self.nms_threshold,
-            'keep_top_k': -1,
+            'keep_top_k': self.keep_top_k,
             'nms_eta': 1.0,
             'normalized': False,
             'return_index': True

x2paddle/op_mapper/onnx2paddle/opset9/opset.py

@@ -465,12 +465,20 @@ class OpSet9():
             inputs={"input": val_rois.name},
             outputs=[val_rois_shape])
         val_rois_num = val_rois.name + '_num'
-        self.paddle_graph.add_layer(
-            'paddle.split',
-            inputs={"x": val_rois_shape},
-            outputs=[val_rois_num, '_', '_', '_'],
-            num_or_sections=[1, 1, 1, 1],
-            axis=0)
+        if len(val_rois.out_shapes[0]) == 4:
+            self.paddle_graph.add_layer(
+                'paddle.split',
+                inputs={"x": val_rois_shape},
+                outputs=[val_rois_num, ' _', ' _', ' _'],
+                num_or_sections=[1, 1, 1, 1],
+                axis=0)
+        elif len(val_rois.out_shapes[0]) == 2:
+            self.paddle_graph.add_layer(
+                'paddle.split',
+                inputs={"x": val_rois_shape},
+                outputs=[val_rois_num, ' _'],
+                num_or_sections=[1, 1],
+                axis=0)
         layer_attrs = {
             'pooled_height': pooled_height,
             'pooled_width': pooled_width,
@@ -1329,7 +1337,7 @@ class OpSet9():
     @print_mapping_info
     def Flatten(self, node):
         val_x = self.graph.get_input_node(node, idx=0, copy=True)
-        output_shape = node.out_shapes[0]
+        output_shape = val_x.out_shapes[0]
         axis = node.get_attr('axis', 1)
         shape_list = [1, 1]
         if axis == 0:
@@ -2192,15 +2200,16 @@ class OpSet9():
         layer_outputs = [nn_op_name, output_name]
         boxes = self.graph.get_input_node(node, idx=0, copy=True)
         scores = self.graph.get_input_node(node, idx=1, copy=True)
+        num_classes = scores.out_shapes[0][1]
         inputs_len = len(node.layer.input)
         layer_attrs = dict()
         if inputs_len > 2:
             max_output_boxes_per_class = self.graph.get_input_node(
                 node, idx=2, copy=True)
-            layer_attrs["nms_top_k"] = _const_weight_or_none(
-                max_output_boxes_per_class).tolist()[0]
+            layer_attrs["keep_top_k"] = _const_weight_or_none(
+                max_output_boxes_per_class).tolist()[0] * num_classes
         else:
-            layer_attrs["nms_top_k"] = 0
+            layer_attrs["keep_top_k"] = 0
         if inputs_len > 3:
             iou_threshold = self.graph.get_input_node(node, idx=3, copy=True)
             layer_attrs["nms_threshold"] = _const_weight_or_none(