Model description

Pod used to train the model:

1. Open pod and choose connect.
2. Use Jupyter lab web application to use terminal and upload own photos.
3. Inside workspace, execute below to setup ai-tools.

git clone https://github.com/ostris/ai-toolkit.git
cd ai-toolkit
git submodule update --init --recursive
python3 -m venv venv
source venv/bin/activate
# .\venv\Scripts\activate on windows
# install torch first
pip3 install torch
pip3 install -r requirements.txt

1. Since flex-dev is a gated model, you need to provide huggingface token for authentication. Track and visualize all the pieces of your machine learning pipeline, from datasets to production machine learning models. Wandb would be needed only if you wanted to log and track the training weights and losses

huggingface-cli login
pip install wandb
wandb login

1. Incase you are geeting with low ram error, try giving below command,

1. Create a folder named "lora_me" inside ai-tools, and drag and drop your images and prompt files inside the folder.
2. Upload the "train_lora_flux_24gb.yaml" file inside ai-tools root.
3. Start the training by executing below,

Trigger words

Generating images using your LoRa

git clone https://github.com/comfyanonymous/ComfyUI
cd ComfyUI
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt
python main.py

Other links

Download model