Yolo Custom Training - couldn't open file: data/obj.names

computer-vision object-detection darknet
cabritoIncognito picture cabritoIncognito · Feb 14, 2018 · Viewed 8.8k times · Source

I recently trained Yolo using the steps provided by https://github.com/AlexeyAB/darknet on Windows10 and had great results on the PC when using my custom training set. However, I am having issues deploying to a linux machine when running the command ./darknet detector test data/obj.data cfg/yolo-obj.cfg yolo-obj_25000.weights which returns:

Couldn't open file: data/obj.names

I've checked that the paths are correct, the files exist and I'm able to cat and return results on all files in their respective directories. I used relative paths and still had the same issue. Any insight on what I might be missing?

Answer

Shailaja picture Shailaja · Jun 15, 2018

Just copy the obj.names file in the darknet parent directory (outside the cfg folder) and it worked for me. Note that this is the temporary solution and you need to fix that by changing paths appropriately.

Related questions

How to do transfer learning in darknet for YoloV3

I want to do transfer learning in YOLOv3 in Darknet so I want to use the pre-trained model of YOLOv3 that was trained on COCO dataset and then further train it on my own dataset to detect additional objects. So …

Read More
Image Processing: What are occlusions?

I'm developing an image processing project and I come across the word occlusion in many scientific papers, what do occlusions mean in the context of image processing? The dictionary is only giving a general definition. Can anyone describe them using …

Read More
What are good algorithms for vehicle license plate detection?

Background For my final project at university, I'm developing a vehicle license plate detection application. I consider myself an intermediate programmer, however my mathematics knowledge lacks anything above secondary school, which makes producing the right formulas harder than it probably …

Read More