Description

This is Text-to-Image Model Finetuned on top of Flux-Dev Model on a Dataset of Amul mascot girl Images

Amul Mascot girl - Lora-fp16-v2

Model trained with AI Toolkit by Ostris

Prompt
amul girl - playing cricket at beach

Prompt
amul girl - M. Karunanidhi eminant DMK leader, writer and Amul Butter.

Prompt
amul girl - When helmets were made compulsory in Bombay

Trigger words

You should use amul girl to trigger the image generation.

Download model

Weights for this model are available in Safetensors format.

Download them in the Files & versions tab.

Scripts for preprocessing , inferencing and finetuning the most popular amul mascot girl

Used flux 1.1 dev model with lora (low rank adaption) for finetuning text to image generation

preprocessing

https://github.com/sanjay7178/amul-mascot-girl-flux-t2i/blob/main/amul_mascot_girl_preprocess.ipynb

dataset

https://huggingface.co/datasets/sanjay7178/amul-mascot-girl

configuration

Install conda-forge or mamba-forge create virtual environment

conda create -n amul python=3.10 
conda activate amul     # change current env to amul

Linux:

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

Windows:

git clone https://github.com/ostris/ai-toolkit.git
cd ai-toolkit
git submodule update --init --recursive
python -m venv venv
.\venv\Scripts\activate
pip install torch torchvision --index-url https://download.pytorch.org/whl/cu121
pip install -r requirements.txt

finetuning

Before fine tuning change dataset path and some required vram hyperparameters according to your system requirements from config.yml in files

python run.py /<relative path>/config.yml

inference (gradio demo)

https://github.com/sanjay7178/amul-mascot-girl-flux-t2i/tree/main/lora-gradio-demo

sanjay7178
/

amul_mascot_girl_flux1_dev_lora_v2_fp16