Man M. Ho

I met the same problem. Have you solved it @freefly-huanshao @Parker-Lyu ?

Sorry I'm new in tensorflow. There are 3 files .data, .index and .meta, how could you specify the ckpt @Parker-Lyu ?

> Sorry I'm new in tensorflow. There are 3 files .data, .index and .meta, how could you specify the ckpt @Parker-Lyu ? Sorry for the stupid question. Problem solved!

@CJHFUTURE Here is how I resolved this problem. Perhaps it's just a style of tensorflow framework to load checkpoint. Please directly give the directory. ``` init_fn = slim.assign_from_checkpoint_fn("/path/to/cvpr-18_avatar-net/data/model.ckpt-120000", slim.get_model_variables(), ignore_missing_vars=True)...

Hi Ofer, Thank you, and thank you so much for your question. It is not the ideal result but the expected result. The idea of PPL is to minimize distances...

Hi Ofer, Yes, I noticed that. The ideal result is that the preset loss is decreased gradually. However, estimating an exact preset (predicting accurate settings) for the actual color transformation...

Hi, thank you for being interested. I may publish it with the extended version.

Hi, Thank you for asking. As described in the script for retouching: ``` local exportImage = require "exportImage" -- .... -- -- .... local _out = outfolder .. tostring(i) LrFileUtils.createDirectory(_out)...

Hi, Thank you for asking. I'm glad to hear that. You can find a way of generating dataset at https://github.com/minhmanho/deep_preset#regarding-training-data Regarding the training code, unfortunately, I have paused my research...

Hi, Yes, sure. I'm working on the tutorial. I'm not sure about your error. Could you please elaborate it with screenshots?