search

e-lab / ENet-training

351 stars 88 forks source link

Reproducing Results #38

Closed dylanheld closed 7 years ago

dylanheld commented 7 years ago

Hello, I've been able to successfully retrain an encoder and decoder model using the Cityscape dataset and the output from the visualization scripts looks great; however I was wondering how one would go about outputting the IoU mAP results on the 1525 test images (Cityscape) stated in the paper with this model. I'd like to confirm that my model produces similar results to the following: image

There doesn't appear to be a dedicated script to do this in the repo.

codeAC29 commented 7 years ago

We followed the instructions provided in the cityscapes website and submitted our results to get performance values on validation data.

dylanheld commented 7 years ago

Thank you; I followed a similar process and received comparable results.