search

TobyPDE / FRRN

Full Resolution Residual Networks for Semantic Image Segmentation
MIT License
278 stars 93 forks source link

I find a bug in train.py #4

Closed lliuz closed 7 years ago

lliuz commented 7 years ago

@TobyPDE The batch_size in test_classification_loss is same to the batch_size in train_loss, which is default by 3 different to the value 1 in validation_provider. 

    # Validation classification loss (supervised)
    test_classification_loss = dltools.utility.bootstrapped_categorical_cross_entropy4d_loss(
        test_predictions,
        target_var,
        batch_size = config["batch_size"],
        multiplier = 64)
    validation_provider = dltools.data.CityscapesHDDDataProvider(
        config["cityscapes_folder"],
        file_folder="val",
        batch_size=1,
        augmentor=None,
        sampling_factor=config["sample_factor"],
        random=False
    )
TobyPDE commented 7 years ago

Thanks for pointing that out. I will fix it on the weekend.

TobyPDE commented 7 years ago

Thank you very much! It was a stupid bug that resulted from hastily extracting the code from a bigger project. I fixed it now.