How to set --config config.ini during Prediction?

Question

I have train model using keras-retinanet for object Detection and Changing Anchor size as per below in config.ini file:

[anchor_parameters]
sizes = 16 32 64 128 256
strides = 8 16 32 64 128 
ratios = 0.5 1 2 3 
scales = 1 1.2 1.6

I have save this config in file config.ini and I put this as input to training as below:

!python keras_retinanet/bin/train.py \ 
--freeze-backbone \ 
--random-transform \ 
--weights {PRETRAINED_MODEL} \ 
--batch-size 1 \ 
--steps 500 \ 
--epochs 5 \ 
--config config.ini \ 
csv annotations.csv classes.csv

And Training Goes Good but how to use this file during Prediction in given function?

convert_model(model,nms=True, class_specific_filter=True, anchor_params=None)??

I am using Below Code to Load Model

model_path = os.path.join('snapshots', sorted(os.listdir('snapshots'), reverse=True)[0])
model = models.load_model(model_path, backbone_name='resnet50')
model = models.convert_model(model,anchor_params=anchor_parameters)
labels_to_names = pd.read_csv(CLASSES_FILE, header=None).T.loc[0].to_dict()

Convert Model Works like as Below:

def convert_model(model, nms=True, class_specific_filter=True, anchor_params=None):
    """ Converts a training model to an inference model.

    Args
        model                 : A retinanet training model.
        nms                   : Boolean, whether to add NMS filtering to the converted model.
        class_specific_filter : Whether to use class specific filtering or filter for the best scoring class only.
        anchor_params         : Anchor parameters object. If omitted, default values are used.

    Returns
        A keras.models.Model object.

    Raises
        ImportError: if h5py is not available.
        ValueError: In case of an invalid savefile.
    """
    from .retinanet import retinanet_bbox
    return retinanet_bbox(model=model, nms=nms, class_specific_filter=class_specific_filter, anchor_params=anchor_params)

How can I set Config.ini or Anchor Parameters during Prediction or load model as above code???

Answer 1

Convert your trained model for inference -

keras_retinanet/bin/convert.py --config config.ini trained_model.h5 converted_model.h5