Update inference.py

inference.py

@@ -205,6 +205,11 @@ def format_predictions(image_shapes, outputs, width, height, gp=None, batch_size
         gp: a callback function which takes as input 1 parameter, (int) percent complete
         prep_for_marking: re-index fish for manual marking output
     """
+
+    print(type(image_shapes))
+    print(len(image_shapes))
+    print(type(outputs))
+    print(len(outputs))
     
     if (gp): gp(0, "Formatting...")
     # keep predictions to feed them ordered into the Tracker
@@ -216,7 +221,9 @@ def format_predictions(image_shapes, outputs, width, height, gp=None, batch_size
             if gp: gp(batch_i / len(image_shapes), pbar.__str__())
 
             batch_shapes = image_shapes[batch_i]
+            print(type(batch_shapes))
             print(len(batch_shapes))
+            print(type(batch))
             print(len(batch))
             print(batch_shapes[0])
             print(batch[0])
@@ -402,16 +409,16 @@ def json_dump_round_float(some_object, out_path, num_digits=4):
 def filter_detection_size(inference, image_meter_width, width, max_length):
 
     outputs = []
-    for prediction in inference:
+    for batch in inference:
 
         pix2width = image_meter_width/width
         print("pix2w", pix2width)
-        width = prediction[..., 2]*pix2width
+        width = batch[..., 2]*pix2width
         wc = width < max_length
-        bs = prediction.shape[0]  # batches
+        bs = batch.shape[0]  # batches
 
-        output = [torch.zeros((0, 6), device=prediction.device)] * bs
-        for xi, x in enumerate(prediction):
+        output = [torch.zeros((0, 6), device=batch.device)] * bs
+        for xi, x in enumerate(batch):
             x = x[wc[xi]]  # confidence
             output[xi] = x