fix multilabel

deploy/configs/inference_cls_multilabel.yaml

@@ -25,11 +25,10 @@ PreProcess:
         order: ''
         channel_num: 3
     - ToCHWImage:
-    
+
 PostProcess:
-  main_indicator: MultiLabelTopk
-  MultiLabelTopk:
-    topk: 5
-    class_id_map_file: None
+  main_indicator: MultiLabelThreshOutput
+  MultiLabelThreshOutput:
+    threshold: 0.5 
   SavePreLabel:
-    save_dir: ./pre_label/
\ No newline at end of file
+    save_dir: ./pre_label/

deploy/python/postprocess.py

@@ -138,12 +138,29 @@ class Topk(object):
         return y
 
 
-class MultiLabelTopk(Topk):
-    def __init__(self, topk=1, class_id_map_file=None):
-        super().__init__()
+class MultiLabelThreshOutput(object):
+    def __init__(self, threshold=0.5):
+        self.threshold = threshold
 
     def __call__(self, x, file_names=None):
-        return super().__call__(x, file_names, multilabel=True)
+        y = []
+        for idx, probs in enumerate(x):
+            index = np.where(probs >= self.threshold)[0].astype("int32")
+            clas_id_list = []
+            score_list = []
+            for i in index:
+                clas_id_list.append(i.item())
+                score_list.append(probs[i].item())
+            result = {
+                "class_ids": clas_id_list,
+                "scores": np.around(
+                    score_list, decimals=5).tolist(),
+                "label_names": []
+            }
+            if file_names is not None:
+                result["file_name"] = file_names[idx]
+            y.append(result)
+        return y
 
 
 class SavePreLabel(object):

docs/en/quick_start/quick_start_multilabel_classification_en.md

@@ -107,7 +107,7 @@ Inference and prediction through predictive engines:
 
 ```
 python3 python/predict_cls.py \
-     -c configs/inference_multilabel_cls.yaml
+     -c configs/inference_cls_multilabel.yaml
 ```
 
 Obtain an output silimar to the following:

docs/zh_CN/quick_start/quick_start_multilabel_classification.md

@@ -100,7 +100,7 @@ cd ./deploy
 
 ```
 python3 python/predict_cls.py \
-     -c configs/inference_multilabel_cls.yaml
+     -c configs/inference_cls_multilabel.yaml
 ```
 
 得到类似下面的输出：

ppcls/configs/quick_start/professional/MobileNetV1_multilabel.yaml

@@ -99,7 +99,7 @@ DataLoader:
       use_shared_memory: True
 
 Infer:
-  infer_imgs: ./deploy/images/0517_2715693311.jpg
+  infer_imgs: deploy/images/0517_2715693311.jpg
   batch_size: 10
   transforms:
     - DecodeImage:
@@ -116,9 +116,8 @@ Infer:
         order: ''
     - ToCHWImage:
   PostProcess:
-    name: MultiLabelTopk
-    topk: 5
-    class_id_map_file: None
+    name: MultiLabelThreshOutput
+    threshold: 0.5
 
 Metric:
   Train:

ppcls/data/postprocess/__init__.py

@@ -16,8 +16,8 @@ import importlib
 
 from . import topk, threshoutput
 
-from .topk import Topk, MultiLabelTopk
-from .threshoutput import ThreshOutput
+from .topk import Topk
+from .threshoutput import ThreshOutput, MultiLabelThreshOutput
 from .attr_rec import VehicleAttribute, PersonAttribute
 
 

ppcls/data/postprocess/threshoutput.py

@@ -12,6 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+import numpy as np
 import paddle.nn.functional as F
 
 
@@ -34,3 +35,28 @@ class ThreshOutput(object):
                 result["file_name"] = file_names[idx]
             y.append(result)
         return y
+
+
+class MultiLabelThreshOutput(object):
+    def __init__(self, threshold=0.5):
+        self.threshold = threshold
+
+    def __call__(self, x, file_names=None):
+        y = []
+        x = F.sigmoid(x).numpy()
+        for idx, probs in enumerate(x):
+            index = np.where(probs >= self.threshold)[0].astype("int32")
+            clas_id_list = []
+            score_list = []
+            for i in index:
+                clas_id_list.append(i.item())
+                score_list.append(probs[i].item())
+            result = {
+                "class_ids": clas_id_list,
+                "scores": np.around(
+                    score_list, decimals=5).tolist(),
+            }
+            if file_names is not None:
+                result["file_name"] = file_names[idx]
+            y.append(result)
+        return y

ppcls/data/postprocess/topk.py

@@ -79,10 +79,3 @@ class Topk(object):
             y.append(result)
         return y
 
-
-class MultiLabelTopk(Topk):
-    def __init__(self, topk=1, class_id_map_file=None):
-        super().__init__()
-
-    def __call__(self, x, file_names=None):
-        return super().__call__(x, file_names, multilabel=True)

ppcls/engine/engine.py

@@ -501,7 +501,7 @@ class Engine(object):
         assert self.mode == "export"
         use_multilabel = self.config["Global"].get(
             "use_multilabel",
-            False) and "ATTRMetric" in self.config["Metric"]["Eval"][0]
+            False) or "ATTRMetric" in self.config["Metric"]["Eval"][0]
         model = ExportModel(self.config["Arch"], self.model, use_multilabel)
         if self.config["Global"]["pretrained_model"] is not None:
             load_dygraph_pretrain(model.base_model,