From d641d8ab6db7a926cc770b626e048f57684f81e2 Mon Sep 17 00:00:00 2001 From: ArDiouscuros <72071512+ArDiouscuros@users.noreply.github.com> Date: Tue, 27 Sep 2022 08:59:16 +0200 Subject: [PATCH] Update documents, add new options information --- docs/features/CLI.md | 2 ++ docs/help/TROUBLESHOOT.md | 10 ++++++++++ docs/other/CONTRIBUTORS.md | 1 + 3 files changed, 13 insertions(+) diff --git a/docs/features/CLI.md b/docs/features/CLI.md index 89b872d46b..4a1580fc0d 100644 --- a/docs/features/CLI.md +++ b/docs/features/CLI.md @@ -98,6 +98,8 @@ overridden on a per-prompt basis (see [List of prompt arguments] | `--gfpgan_dir` | | `src/gfpgan` | Path to where GFPGAN is installed. | | `--gfpgan_model_path` | | `experiments/pretrained_models/GFPGANv1.4.pth` | Path to GFPGAN model file, relative to `--gfpgan_dir`. | | `--device ` | `-d` | `torch.cuda.current_device()` | Device to run SD on, e.g. "cuda:0" | +| `--free_gpu_mem` | | `False` | Free GPU memory after sampling, to allow image decoding and saving in low VRAM conditions | +| `--precision` | | `auto` | Set model precision, default is selected by device. Options: auto, float32, float16, autocast | #### deprecated diff --git a/docs/help/TROUBLESHOOT.md b/docs/help/TROUBLESHOOT.md index abf5bc2663..d91867fef3 100644 --- a/docs/help/TROUBLESHOOT.md +++ b/docs/help/TROUBLESHOOT.md @@ -113,3 +113,13 @@ branch that contains the pull request: You will need to go through the install procedure again, but it should be fast because all the dependencies are already loaded. + +--- + +### **QUESTION** + +Image generation crashed with CUDA out of memory error after successful sampling. + +### **SOLUTION** + +Try to run script with option `--free_gpu_mem` This will free memory before image decoding step. diff --git a/docs/other/CONTRIBUTORS.md b/docs/other/CONTRIBUTORS.md index 8f40419791..be4f3f407c 100644 --- a/docs/other/CONTRIBUTORS.md +++ b/docs/other/CONTRIBUTORS.md @@ -57,6 +57,7 @@ We thank them for all of their time and hard work. - [Kyle Schouviller](https://github.com/kyle0654) - [rabidcopy](https://github.com/rabidcopy) - [Dominic Letz](https://github.com/dominicletz) +- [Dmitry T.](https://github.com/ArDiouscuros) ## **Original CompVis Authors:**