Get scores, boxes, labels and logits from detr model
#13
by khasaad - opened
Hello everyone,
To get 'scores', 'labels' and 'boxes' , we can do like that:
model = DetrForSegmentation.from_pretrained("facebook/detr-resnet-50-panoptic")
outputs = model(**inputs)
results = image_processor.post_process_object_detection(outputs)[0] # {'scores': ..., 'labels': ..., 'boxes': ...}
How we can get {'scores': ..., 'labels': ..., 'boxes': ..., 'logits': ...} at the same time with the same length ??
detections = []
for score, label, box in zip(results["scores"], results["labels"], results["boxes"]):
detections.append({
"label": model.config.id2label[label.item()],
"confidence": round(score.item(), 3),
"box": [round(i, 2) for i in box.tolist()]