Convert tiny-yolov3 to Tflite.

[apoorvopen]

Can you help me in converting the tiny-yolov3 model to .Tflite to run it on android? Please let me know if you have any leads for me.

[weilanShi]

Can you help me in converting the tiny-yolov3 model to .Tflite to run it on android? Please let me know if you have any leads for me. Hello, have you succeeded in converting. weight to. tflite?

[weilanShi]

Hello, I have succeeded in converting. weigths to. tflite. The steps are as follows: first, convert. weights. cfg to. pb and .ckpt, referring specifically to: https://github.com/jinyu121/DW2TF/issues/30 https://github.com/jinyu121/DW2TF/releases/tag/freez/tensorflow tool e_graph, then generate. pb. Finally, the tensorflow tool toco is used to convert to. tflite.

[nscotto]

Could you provide the commands you entered to get the tflite version from the darknet weights? I can't figure out what to enter for freezing the graph, I get errors on every trials, and your links don't work. Especially, I would like to convert weights of a self trained YoloV3 from darknet to TFlite.

Thank you in advance.

[peace195]

@apoorvopen I did it here: https://github.com/peace195/tensorflow-lite-yolo-v3

Please try it. I would appreciate if you give me a star for this project 👍

[nscotto]

Thanks, I will. Before I try, does it support different number of channels? (1 for grayscale images)

[peace195]

@nscotto Yes, it does. Please change the size of your image in this line: https://github.com/peace195/tensorflow-lite-yolo-v3/blob/master/convert_weights_pb.py#L42

[nscotto]

Thanks! Star deserved.

[peace195]

@nscotto I need some helps: https://github.com/peace195/tensorflow-lite-yolo-v3/issues/1

[spoorgholi74]

@nscotto I need some helps: peace195/tensorflow-lite-yolo-v3#1

your repo only gets the common yolov3 config right? so if pruned or change it should be change accordingly?

[peace195]

@nscotto I need some helps: peace195/tensorflow-lite-yolo-v3#1

your repo only gets the common yolov3 config right? so if pruned or change it should be change accordingly?

Yes. You are correct.

[peace195]

@spoorgholi74 You can change the YOLO's structure at here: https://github.com/peace195/tensorflow-lite-yolo-v3/blob/master/yolo_v3.py

[spoorgholi74]

@peace195 yeah but it would be better to write a function to read the .cfg file instead of changing the lines

[peace195]

@spoorgholi74 I have a version of it. I will push the code when I have time. 👍