4

I have trained the model using resnet18 for mask R-CNN detection. For every epoch, it created a ".ckpt" file only. Now I want to use that .ckpt file as the detector for detecting images. I have python code that takes a ".h5" file for detection. Please help me with How I can use the ".ckpt" file for detection. or How can I convert it to the ".h5"? Thanks

I have tried generating ".h5" file instead of ".ckpt" with training process but it did not work for me. Now I need a way to use ".ckpt" files for detecting objects in Images.

Dhanraj Jain
  • 61
  • 4
  • 1
    You can use model.load_weights() to load the weights to model from checkpoint folder to use model again for saving to h5 format . Attached reference [document](https://www.tensorflow.org/tutorials/keras/save_and_load#checkpoint_callback_options) for reference. –  Dec 06 '22 at 05:14

0 Answers0