Text to Image generation using Stable Diffusion || HuggingFace Tutorial Diffusers Library

When Dalle-2 was released, creation of images using prompts became very easy. It had been a revolutionary tool for artists and creators
However, Dalle-2 is a proprietary model and you would not want to spend dollars merely when you are testing prompts or even for simple use cases
Stable Diffusion, based on the diffusion model, is a FREE and OPEN Source model that helps us in generating images from text prompts and even transferring the styles from one image to another
In this video, we shall learn to generate images using the Generative AI Stable Diffusion model. We shall use the diffusers library of hugging face to help us in creating a pipeline for image generation using text prompts
Additionally, we shall explore the various diffusion pipelines for tasks such as text2image, image2image, text2music, etc. [read more here: huggingface.co...]
A diffusion pipeline comprises of 2 main components: the UNet Model and the schedulers. We shall discuss their functioning as well
If you find the content useful, make sure to give it a THUMBS UP :)
References:
fine tuning language models using AutoTrain: [ • Fine tuning Llama2 usi... ]
Stable Diffusion notebook: colab.research...
--------------------------------------------------------------------------------------------------------------------------------------
Why learn from datahat??
"Datahat is a bridge"
learners --- professionals by simplified data science
It provides a platform to learn, create and collaborate helping in the #career #transition to data science, machine learning and data analysis
Here are few interesting reads on our blog: / souravagarwal54321
Connect with us on linkedIn: / datahat-unfoldingmystery
Book a mentorship session: topmate.io/dat...
----------------------------------------------------------------------------------------------------------------------------------
Remember, "the greatest investment ever made is investment in self growth and learning"

[@SantinoGenta-m4y]

Bro i need help

[@datahat642]

sure, tell me....

[@SantinoGenta-m4y]

Hello , where i can contact you? I’m creating an Ai text to image app for ios, your help is gold

[@datahat642]

hey, we can connect on LinkedIn: www.linkedin.com/in/sourav-agarwal-saag?

[@Xyz282kd]

​@@datahat642hello brother
Ma custom Model ko kese use Karu is me ?? Pls reply

[@datahat642]

@@Xyz282kd agar aap Hugging Face pretrained models use karna chahte ho then change the model_id with the diffusion model of your choice. Alternatively, if you want to use your own fine-tuned model for the purpose, then refer here: discuss.huggingface.co/t/how-to-use-the-fine-tuned-model-for-actual-prediction-after-re-loading-it/21374/2

[@ngoc_pham_encity]

Can we upload a model and use it? Could you guide?

[@fahaamshawl9335]

Can we give examples to the model like how to generate images of what type? Similar to few shot prompting concept

[@datahat642]

Please check this reference: medium.com/intel-analytics-software/personalized-stable-diffusion-with-few-shot-fine-tuning-on-a-single-cpu-f01a3316b13

[@chinmaydeshpande5046]

Great video bro !!
really helped a lot ...

[@datahat642]

Thank you

[@lorenzojuancarlosd.2890]

thanks brother! do you have paper for this?

[@datahat642]

Unfortunately, we don't have any on this

[@NehaKothari-iz3hy]

Plz help me to know that how to ise custom data for text to image gener

[@datahat642]

Would like to hear the feedback from your recent interaction

[@daanidev]

how to freeze it in the tf lite model to integrate in android app?

[@datahat642]

Tensorflow lite does not directly support freezing the layers. Alternatively, you can quantize your model in tensorflow before converting to TFLite

[@franchementcestca5630]

can I create a discord bot by giving it the stable distribution code in my own server?

[@datahat642]

Yes you can... save the model & deploy it on your server

[@SantinoGenta-m4y]

Using this