search

FangGet / tf-monodepth2

Tensorflow implementation(unofficial) of "Digging into Self-Supervised Monocular Depth Prediction"
MIT License
81 stars 16 forks source link

is pretrained model broken ? #1

Closed ixtiyoruz closed 5 years ago

ixtiyoruz commented 5 years ago

when i run the command : python train_monodepth2.py train config/monodepth2_kitti.yml ./model/pretrained_model/model.latest

i am getting the error :

_Traceback (most recent call last): File "train_monodepth2.py", line 51, in _cli_test(config, output_dir, args) File "train_monodepth2.py", line 25, in _cli_test monodepth2_learner.test(output_dir) File "C:\Users\essys\Desktop\project_1\depth_estimation\tf-monodepth2-master\tf-monodepth2-master\model\monodepth2_learner.py", line 371, in test self.saver.restore(sess, latest_ckpt) File "C:\Users\essys\Anaconda3\envs\tf_gpu\lib\site-packages\tensorflow\python\training\saver.py", line 1274, in restore raise ValueError("Can't load save_path when it is None.") ValueError: Can't load savepath when it is None.

FangGet commented 5 years ago

Sorry for late reply, have you solved this problem? I will have a try on it.

ixtiyoruz commented 5 years ago

I haven't tried since, did you update the model ?

FangGet commented 5 years ago

@ixtiyoruz I just have a try. and soory, it's my fault, you need to add a checkpoint txt file to tell tensorflow which model you decide to use, example like that: filename: checkpoint (no extension) file detail: model_checkpoint_path: "/YOUR/PATH/TO/pretrained_model/model.latest"

put this file inside pretrained_model directory and everything will be ok.

ixtiyoruz commented 5 years ago

Sorry for late reply, Ok, Thank you.