Zhen Li
Zhen Li
① For computational efficiency. ② You will obtain better reconstruction performance if you enlarge T or G.
You can first check the dataset options in your `.json` file.
We will test the compatibility of multi-gpu and give your feedback in a few days. If you want to inject a BN layer in one conv/deconv block, you can change...
You can use [this function](https://github.com/Paper99/SRFBN_CVPR19/blob/da28bcf985543f7c24f456d43125c3d6fb7badc0/solvers/SRSolver.py#L385) to calculate the parameters of the SRFBN. The number of parameters in SRFBN is 3,631K.
Thank you for reminding me of this mistake. I haven't noticed this yet. I will correct it as soon as possible.
Hello everyone, first of all, thank you for your discussions, which helped us fix many known issues in environment construction (in the latest version). Now, you could update the environment...
pip install -U transformers
Hello everyone, first of all, thank you for your discussions, which helped us fix many known issues in environment construction (in the latest version). Now, you could update the environment...
Hello everyone, first of all, thank you for your discussions, which helped us fix many known issues in environment construction (in the latest version). Now, you could update the environment...
Please give us more details. Thank you.