problem when predicting an image with a new dataset

[jafarMajidpour]

when training and predicting with your dataset there is not problem and predicts the images correctly but when using a new dataset to predict images it gives a weird result, see the image below:

in all epochs shows the same predicted image as in the above mentioned image.

[junyanz]

Not sure if your new input label map has the same color scheme with Cityscape color scheme. If I were you, I would train a new model for the new dataset.

[jafarMajidpour]

Not sure if your new input label map has the same color scheme with Cityscape color scheme. I will train a new model for the new dataset.

thank you very much for response. kindly, could you please send me the model that you had created for the new dataset? and how we can labeling a new dataset programmatically? could you please send me that program?

[junyanz]

Sorry. I meant if I were you, I would train a new dataset. We don't have models for your datasets. It seems that you already have the ground truth annotation for each image in your dataset. You can just train a model on your datasets with our code.