how to train on my own dataset?

[JimmyLauren]

Hi, thanks for ur excellent work. What modifications should i made to train on my own dataset?

[yijingru]

You can modify dataset_dota.py in datasets by replacing the following functions

    def load_img_ids(self):
        return None

    def load_image(self, index):
        return None

    def load_annoFolder(self, img_id):
        return None

    def load_annotation(self, index):
        return None

    def dec_evaluation(self, result_path):
        return None

[JimmyLauren]

thanks for reply. Another question, these code uses lib DOTA_devkit, but i could not find it, can you provide this file?

[yijingru]

It's this link: https://github.com/CAPTAIN-WHU/DOTA_devkit

[JimmyLauren]

thanks a lot