2022-12-17 20:32:48 -07:00
# EveryDream Trainer 2.0
2022-12-17 22:43:25 -07:00
Welcome to v2.0 of EveryDream trainer! Now with more diffusers and even more features!
2022-12-17 20:32:48 -07:00
Please join us on Discord! https://discord.gg/uheqxU6sXN
2023-01-08 16:52:39 -07:00
If you find this tool useful, please consider subscribing to the project on [Patreon ](https://www.patreon.com/everydream ) or a one-time donation at [Ko-fi ](https://ko-fi.com/everydream ).
2022-12-17 20:32:48 -07:00
2023-02-26 17:11:42 -07:00
If you're coming from Dreambooth, please [read this ](doc/NOTDREAMBOOTH.md ) for an explanation of why EveryDream is not Dreambooth.
2023-02-28 08:57:49 -07:00
## Requirements
Windows 10/11, Linux (Ubuntu 20.04+ recommended), or use the linux Docker container
Python 3.10.x
Nvidia GPU with 11GB VRAM or more (note: 1080 Ti and 2080 Ti may require compiling xformers yourself)
16GB system RAM recommended minimum
Single GPU is currently supported
32GB of system RAM recommended for 50k+ training images, but may get away with sufficient swap file and 16GB
Ampere or newer 24GB+ (3090/A5000/4090, etc) recommended for 10k+ images unless you want to wait a long time
...Or use any computer with a web browser and run on Vast/Runpod/Colab. See [Cloud ](#cloud ) section below.
2022-12-18 15:21:13 -07:00
## Video tutorials
### [Basic setup and getting started](https://www.youtube.com/watch?v=OgpJK8SUW3c)
Covers install, setup of base models, startning training, basic tweaking, and looking at your logs
### [Multiaspect and crop jitter](https://www.youtube.com/watch?v=0xswM8QYFD0)
Behind the scenes look at how the trainer handles multiaspect and crop jitter
2023-02-06 11:11:24 -07:00
### Companion tools repo
2023-01-17 10:44:18 -07:00
Make sure to check out the [tools repo ](https://github.com/victorchall/EveryDream ), it has a grab bag of scripts to help with your data curation prior to training. It has automatic bulk BLIP captioning for BLIP, script to web scrape based on Laion data files, script to rename generic pronouns to proper names or append artist tags to your captions, etc.
2022-12-17 20:32:48 -07:00
## Docs
[Setup and installation ](doc/SETUP.md )
2022-12-18 15:21:13 -07:00
[Download and setup base models ](doc/BASEMODELS.md )
2022-12-17 20:32:48 -07:00
[Data Preparation ](doc/DATA.md )
2023-01-23 14:28:23 -07:00
[Training ](doc/TRAINING.md ) - How to start training
2022-12-17 20:32:48 -07:00
2023-01-23 14:28:23 -07:00
[Basic Tweaking ](doc/TWEAKING.md ) - Important args to understand to get started
2022-12-17 20:59:03 -07:00
2023-01-23 14:28:23 -07:00
[Logging ](doc/LOGGING.md )
2022-12-17 22:43:25 -07:00
2023-01-23 14:28:23 -07:00
[Advanced Tweaking ](doc/ATWEAKING.md ) - More stuff to tweak once you are comfortable
2023-01-08 16:52:39 -07:00
2023-02-25 14:19:23 -07:00
[Advanced Optimizer Tweaking ](/doc/OPTIMIZER.md ) - Even more stuff to tweak if you are *very adventurous*
2023-01-23 14:28:23 -07:00
[Chaining training sessions ](doc/CHAINING.md ) - Modify training parameters by chaining training sessions together end to end
2023-01-07 14:59:51 -07:00
[Shuffling Tags ](doc/SHUFFLING_TAGS.md )
2023-01-23 14:28:23 -07:00
2023-02-06 11:11:24 -07:00
[Data Balancing ](doc/BALANCING.md ) - Includes my small treatise on model preservation with ground truth data
2023-02-08 14:11:09 -07:00
[Validation ](doc/VALIDATION.md ) - Use a validation split on your data to see when you are overfitting and tune hyperparameters
2023-02-18 12:49:30 -07:00
[Troubleshooting ](doc/TROUBLESHOOTING.md )
2023-02-06 11:11:24 -07:00
## Cloud
[Free tier Google Colab notebook ](https://colab.research.google.com/github/victorchall/EveryDream2trainer/blob/main/Train_Colab.ipynb )
2023-02-25 14:19:23 -07:00
[RunPod / Vast ](/doc/CLOUD_SETUP.md )
[Docker image link ](https://github.com/victorchall/EveryDream2trainer/pkgs/container/everydream2trainer )
