Cryst4L

A problem when compiled with CUDA11.1 and RTX 3090

3

I've follow all the steps to build the llvm、terra and Opt one by one, and i"m sure all of the tests of terra are passed, so it must be a...

Augmented Data Available?

2

Since I want to reproduce the entire training process,I'm wondering if the augmented data of Ho3d-v1 can be released.

avoid bad image message

1

In addition, considering the error when transferring the information and other noise, we mutate some of the image message and there will be a crash like following: ``` maze_solver: Traceback...

More robust initialization against a delay of info

1

Hi! I'm a researcher on bug detection and software robustness focusing on robotic applications. It's our new idea to design a language-irrelevant algorithm, so we are trying to check and...

Simple question about using the script preprocess_in_the_wild.py to get the poses.

1

Hi! Thanks for the excellent work. and detailed code. I just have a question about the dataset preprocessing: For the script preprocess_in_the_wild.py, should I set the INPUT_IMAGE_FOLDER as the original...

something wrong with the concatdataloader for the newer torch

1

Hi! I'm going to run the code on the ubuntu with a 3090Ti,so it's necessary that I should use the pytorch>=1.7. but here is something wrong with the concatdataloader: Traceback...

Where is the testing_emonet_split.csv?

1

I'm trying to reproduce the training and testing of emotion recognition, but it seems the testing_emonet_split.csv is not provided.

About dataset_mead.json and dataset_exp.json

2

Hi! Thanks for the excellent work! I just have a simple question about the training code: With the param --gen_exp_cond and --gen_pose_cond, I think there should be the data dataset_mead.json...