add valid

configs/example.yaml

@@ -18,12 +18,10 @@ DATASET: #DATASET field
         format: "VideoDataset" #Mandatory, indicate the type of dataset, associate to the 'paddlevidel/loader/dateset'
         data_prefix: "" #Mandatory, train data root path
         file_path: "data/ucf101/ucf101_train_split_1_videos.txt" #Mandatory, train data index file path
-        # suffix: '{:05}.jpg'
     valid:
         format: "VideoDataset" #Mandatory, indicate the type of dataset, associate to the 'paddlevidel/loader/dateset'
-        data_prefix: "data/ucf101/videos_val" #Mandatory, valid data root path
-        file_path: "./data/dataset/ucf101/valid_video.txt" #Mandatory, valid data index file path
-        # suffix: '{:05}.jpg'
+        data_prefix: "" #Mandatory, valid data root path
+        file_path: "./data/ucf101/ucf101_val_split_1_videos.txt" #Mandatory, valid data index file path
 
 PIPELINE: #PIPELINE field
     train: #Mandotary, indicate the pipeline to deal with the training data, associate to the 'paddlevideo/loader/pipelines/'
@@ -36,7 +34,7 @@ PIPELINE: #PIPELINE field
             valid_mode: False
         transform: #Mandotary, image transfrom operator
             - Scale:
-                short_size: 224
+                short_size: 256
             - RandomCrop:
                 target_size: 224
             - RandomFlip:
@@ -46,14 +44,23 @@ PIPELINE: #PIPELINE field
                 std: [0.229, 0.224, 0.225]
 
     valid: #Mandatory, indicate the pipeline to deal with the validing data. associate to the 'paddlevideo/loader/pipelines/'
+        decode:
+            name: "VideoDecoder"
         sample:
             name: "Sampler"
             valid_mode: True
-            clip_len: 1
-            num_clips: 8 
+            num_seg: 8
+            seg_len: 1 
         transform:
-            - Resize:
-            - Norm:
+            - Scale:
+                short_size: 256
+            - CenterCrop:
+                target_size: 224
+            - Image2Array:
+            - Normalization:
+                mean: [0.485, 0.456, 0.406]
+                std: [0.229, 0.224, 0.225]
+
 
 OPTIMIZER: #OPTIMIZER field
     name: 'Momentum' #Mandatory, the type of optimizer, associate to the 'paddlevideo/solver/'

main.py

@@ -58,6 +58,11 @@ def main():
         paddle.distributed.init_parallel_env()
 
     model = build_model(cfg.MODEL)
+
+    #NOTE: To debug dataloader or wanna inspect your data, please try to print(dataset) here.
+    #    for i in dataset[0]:
+    #        logger.error(i)
+
     dataset = [build_dataset((cfg.DATASET.train, cfg.PIPELINE.train))]
     if args.validate:
         dataset.append(build_dataset((cfg.DATASET.valid, cfg.PIPELINE.valid)))

paddlevideo/modeling/framework/recognizers/base.py

@@ -81,7 +81,7 @@ def train_step(self, data_batch, **kwargs):
         return loss_metrics
 
     def val_step(self, data_batch, **kwargs):
-        """Interface to valid
+        """Validating setp.
         """
         imgs = data_batch[0]
         labels = data_batch[1]

paddlevideo/tasks/train.py

@@ -27,7 +27,7 @@ def train_model(model,
                 dataset,
                 cfg,
                 parallel=True,
-	        validate=False):
+	        validate=True):
     """Train model entry
 
     Args:
@@ -63,7 +63,9 @@ def train_model(model,
         model = paddle.DataParallel(model)
 
     train_loader = data_loaders[0]
-
+    if  validate:
+        valid_loader = data_loaders[1]
+
     metric_list = [
         ("loss", AverageMeter('loss', '7.5f')),
         ("lr", AverageMeter(
@@ -116,17 +118,47 @@ def train_model(model,
             if i == 20:
                 break
 
-        """
-        Note: validate is not ready now!
 
-        """
-    end_str = ' '.join([str(m.mean) for m in metric_list.values()] +
+        end_str = ' '.join([str(m.mean) for m in metric_list.values()] +
                        [metric_list['batch_time'].total])
-    end_epoch_str = "END epoch:{:<3d}".format(epoch)
+        end_epoch_str = "END epoch:{:<3d}".format(epoch)
 
-    logger.info("{:s} {:s} {:s}s".format(
+        logger.info("{:s} {:s} {:s}s".format(
             coloring(end_epoch_str, "RED"),
             coloring("TRAIN", "PURPLE"),
             coloring(end_str, "OKGREEN")))
 
-    logger.info('[TRAIN] training finished')
+        if validate:
+            model.eval()
+            tic = time.time()
+            for i, data in enumerate(valid_loader):
+                if parallel:
+                    outputs = model._layers.val_step(data)
+                else:
+                    outputs = model.val_step(data)
+                for name, value in outputs.items():
+                    metric_list[name].update(value.numpy()[0], batch_size)
+                metric_list['batch_time'].update(time.time() - tic)
+                tic = time.time()
+
+
+                if i % cfg.get("log_interval", 10) == 0:
+                    fetchs_str = ' '.join([str(m.value) for m in metric_list.values()])
+                    epoch_str = "epoch:[{:>3d}/{:<3d}]".format(epoch, cfg.epochs)
+                    step_str = "{:s} step:{:<4d}".format("valid", i)
+                    logger.info("{:s} {:s} {:s}s".format(
+                        coloring(epoch_str, "HEADER")
+                        if i == 0 else epoch_str,
+                        coloring(step_str, "PURPLE"),
+                        coloring(fetchs_str, 'OKGREEN')))
+                if i == 20:
+                    break
+
+        logger.info("{:s} {:s} {:s}s".format(
+            coloring(end_epoch_str, "RED"),
+            coloring("VALID", "PURPLE"),
+            coloring(end_str, "OKGREEN")))
+
+    #save and resume is not ready!!!
+
+    logger.info('training finished') #info of yaml

run.sh

@@ -0,0 +1,2 @@
+export CUDA_VISIBLE_DEVICES=0
+python3 -B -m paddle.distributed.launch --selected_gpus=0 main.py -c configs/example.yaml -o log_level="INFO"