Allan Zelener

No, the softmax tree method described by YOLO9000 is not currently implemented here. Most of the other YOLO_v2 features are.

No, separating x and y is not the issue here. I don't remember exactly why rectangular images were broken but I think how `conv_index` is computed is the most likely...

`conv_dims` are the dimensions of the final convolutional layer. If the input image size is (416, 416) then conv_dims are (13, 13). `conv_index` contains every possible index into the convolution...

See `train_overfit.py` for an example of how to train on a single image. Extending this to a larger dataset isn't difficult and basically just requires zero-padding the `boxes` tensor to...

I found that the model needs to be trained for about 1000 steps to consistently fit to my single training image. Haven't tried it but setting the score threshold to...

The loss probably won't fall much below 5 because the loss Keras reports includes the L2 regularization on the model's weights. But I have seen it get closer to about...

A script for (re)training a full model was just contributed. I haven't tested it myself but see `retrain_yolo.py` for details.

`overfit_weights.h5` are just the weights of the model, not the model itself. I did it that way because there seems to be an issue with serializing the model with the...

@chairath The `yolo_loss` function and the `preprocess_true_boxes` function expect the ground truth boxes to be in the form "x_center, y_center, width, height, class" where these values have also been normalized...