Closed jingfanc closed 3 years ago
i used python main.py --input_height 96 --input_width 96 --output_height 48 --output_width 48 --dataset anime2 --crop --train --ckpt_freq 5 --sample_freq 4 visualize=True --epoch 200 --input_fname_pattern "*.jpg"
and it can save samples. so weird
i use
python main.py --dataset anime2 --input_height 96 --output_height 48 --checkpoint_dir="D:\machine_learning\DCGAN-tensorflow-master\out\20210509.001024 - data - anime2 - x96.z100.uniform_signed.y48.b64\checkpoint" --crop False --visualise True
to test. But there was no results in the sample folder.the record ended like this:
[*] Reading checkpoints... D:\machine_learning\DCGAN-tensorflow-master\out\20210507.233835 - data - anime2 - x96.z100.uniform_signed.y48.b64\checkpoint WARNING:tensorflow:From D:\Anaconda3\lib\site-packages\tensorflow\python\training\saver.py:1266: checkpoint_exists (from tensorflow.python.training.checkpoint_management) is deprecated and will be removed in a future version. Instructions for updating: Use standard file APIs to check for files with this prefix. [*] Success to read model.b64-1600
do i need to provide some pictures to be processed?