search

BobLiu20 / YOLOv3_PyTorch

Full implementation of YOLOv3 in PyTorch
624 stars 191 forks source link

Prediction #6

Open crankson123 opened 6 years ago

crankson123 commented 6 years ago

Thanks for your excellent and easily understanding script.

My simple question is how to predict, because there are no prediction module, modifying train.py or eval.py, I have no idea, hope to be answered.

themathgeek13 commented 6 years ago

Hi @crankson123, you can check the eval.py script, there is a portion where it returns x1, y1, x2, y2, obj_conf and obj_pred and you can use this to draw the bounding boxes and show the class labels and confidence scores on the image. This should not be too difficult to implement, I think, but yes, it is not included currently in this repository.

BobLiu20 commented 6 years ago

@crankson123 @themathgeek13 Hi, I had add a test_images.py to test folder. Please update code to use it. Any issue please let me know.

BCWang93 commented 5 years ago

@crankson123 @themathgeek13 Hi, I had add a test_images.py to test folder. Please update code to use it. Any issue please let me know.

Hi ,I change the dataset to voc,but,when I trained the model,when I test the model,the output is nothing,the output is the origin image,there is haven't dections result in the image.Can you tell me this is why?Thank you very much!(I think it is the way of load model has some error?)