Interpretation of 06_classify_image.py outputs (again)

[tonyboston-au]

@jensleitloff I'm having trouble interpreting the classes created by a model in the label.tif file. 02_train_rgb_finetuning.py outputs the classes as:
Found 18900 images belonging to 10 classes.
{'AnnualCrop': 0, 'Forest': 1, 'HerbaceousVegetation': 2, 'Highway': 3, 'Industrial': 4, 'Pasture': 5, 'PermanentCrop': 6, 'Residential': 7, 'River': 8, 'SeaLake': 9}
Found 8100 images belonging to 10 classes.
...
When I review the grayscale TIF image created by 06_classify_image.py, the classes don't make sense. For example, the forest class (identified by overlaying the Sentinel-2 satellite image) is displayed as class #9 #090909, which according to the above legend should be the class 'SeaLake'. I must be doing something wrong!

Any advice is really appreciated.

[jensleitloff]

Unfortunately, it's hard to give any useful advice based on your comment. In our tests the classification was fine (even so, not good as segmentation approaches). What do you mean by

#090909

Normally you should have one label per pixel. Have you tested our pretrained models? Maybe something went wrong during training.

[jensleitloff]

Please also check, that the data you intend to classify, has the same preprocessing as the EuroSAT dataset. We already had this problem in a previous issue

[tonyboston-au]

Thanks @jensleitloff for those suggestions. Yes I downloaded your models a few months ago so can try those and will check that the Sentinel data is prepared the same way as described in that issue.

[felixriese]

Are you on the latest version of the repository, @tonyboston-au ? Please do a

git pull

since in e956484 we fixed a bug.

[tonyboston-au]

okay will do - thanks @felixriese