IdeaSpark 🚀

Built With

Open in Streamlit

Welcome to IdeaSparkAi, a collection of Streamlit apps designed to spark your creativity and aid in idea generation. These apps are part of the INVOKE Analytics project.

IdeaSparkAi is an innovative app designed to help creatives generate compelling ads for social media platforms such as Facebook, Instagram, LinkedIn, and more. With its advanced AI technology, IdeaSparkAi takes the guesswork out of ad creation by providing users with a wealth of ideas and inspiration. Whether you're looking to boost engagement, increase conversions, or simply stand out from the crowd, IdeaSparkAi is the perfect tool for any creative looking to take their social media advertising to the next level. 💡

  • Project name: IdeaSpark - Sparking Ideas through Streamlit
  • Objective: Spark idea to Invoke's creative and marketing team to produce compelling visuals, speed-up ideation process
  • Document generated format: PNG
  • Document format upload: JPEG, PNG
  • Service provider: Streamlit Apps

To Do's

  • Redeploy code on Streamlit Community Cloud with Invoke-Solutions repo instead of Invoke-Analytics (to account for Org Name change).
  • Refactor code into Classes/functions (updating the API is a chore since the way it is initally written is messy).
  • Update n parameter so model produces more than 1 image at a time when OpenAI updates it's documentation (for now, dall-e-3 only produces 1 photo at a time, very annoying!).
  • Fine-tuning for digital marketing purposes so images are more suited for Instagram/Facebook ads (optional as may be super expensive!).

Table of Contents

Introduction

IdeaSparkAi is a repository that hosts various Streamlit apps developed by INVOKE Analytics. Each app is designed to assist you in brainstorming, idea generation, and creative thinking. Whether you're looking to spark your imagination, organize your thoughts, or explore new concepts, these apps can be a valuable resource.

Installation

To run this Streamlit app locally on your machine, follow these steps (we recommend using Python 3.10):

  1. Clone this repository to your local machine: git clone https://github.com/INVOKE-Analytics/IdeaSparkAi.git
  2. Navigate to the repository's directory: cd IdeaSparkAi
  3. Create a virtual environment: python -m venv .venv
  4. Activate your virtual environment: source venv/Scripts/activate (for Git Bash users)
  5. Install the required dependencies (ensure you have Python and pip installed): pip install -r requirements.txt
  6. To launch the application, run the following in the terminal: streamlit run 1_Open_Ended.py (you can use the tab key as well to autocomplete file names e.g., type "1_🥳_Open_Ended.py")

Streamlit Apps

Welcome to a treasure trove of Streamlit wonder! Behold the lineup of dazzling apps within this realm:

Page 1: Open Ended

Description: Set your creativity free with an open canvas for prompts and ideas.

Usage: Simply type in your prompt, choose an image size, and let the magic unfold!

Page 2: Ads Spark

Description: Ignite your ad campaigns with this dynamic tool that generates visual references for your advertising brainstorm sessions.

Usage: Follow the Decoris.io guide, fill in the blanks, and let inspiration light up your screen.

Page 3: Image Variation

Description: Unleash the magic of image variation by effortlessly creating diverse versions of your visuals.

Usage: All the steps are elegantly explained on the page itself.

Page 4: Stock Image

Description: Create captivating stock images that captivate the hearts of your audience.

Usage: The page holds your hand through the entire process.

Page 5: Going Wild

Description: Dive headfirst into a wild sea of prompt engineering, and watch as creativity flows like a river.

Usage: Cut loose, break the mold, and let your imagination run wild!

Description: Going wild with the prompt engineering Usage: Get out of your box, prompt in details what you need.

Usage

Each Streamlit app in this repository serves a specific purpose. To use a particular app, navigate to its directory and run the following command: streamlit run 1_Open_Ended.py

License

This project is licensed under the MIT License.

Contributor


Afif Mansor

LinkedIn