sanjaymarreddi / emotion-investigator

An Exciting Deep Learning-based Flask web app that predicts the Facial Expressions of users and also does Graphical Visualization of the Expressions.

Hello 👋 Sanjay here !

About Me ⚡

• Currently working as an Electronic Trading Developer at Arctic Lake Technology based in London, UK.
• Worked as a Software Development Engineer Intern at Amazon UK, Intuit, and Blaze AI.
• Enjoys building low-latency full-stack systems at the intersection of Finance, Trading, and AI.
• Graduated with a Bachelor's Degree in Computer Science from IIT Goa with a grade of 9.38/10.
• Cleared interviews and received offers from top tech companies such as Amazon UK, Flow Traders Amsterdam, Intuit, Blaze AI, Morgan Stanley, and Mathworks.

---

• Contributed to a few open-source organizations like Scikit-Learn & Fury.
• Worked as an ML intern at a few startups & as a Research Intern at IIT Goa.
• Mentored students in open-source programs like Dev Script - WOC, LetsUpgrade - BWLU.
• Enjoys contributing to open-source projects and doing Competitive Programming.
• An active Member of Google DSC IIT GOA - Conducted several Live Coding Workshops, Competitions, etc.

---

• Skills
  • Languages: C++, C, Python, Java, JavaScript, TypeScript, Haskell, Prolog.
  • Tech Stack: React, Express, SQL, Snowflake, AWS, Flask, Django, OpenCV, Keras, RASA, Git, GitHub.
  • Interests: Quantitative Development, Algo Trading, High-Frequency Trading, Artificial Intelligence.
  • Other skills: GTest, CMake, MultiThreading, Data Structures and Algorithms, Low level and High-Level System Design, Computer Science fundamentals, and Object-Oriented Programming.

emotion-investigator's Issues

need https?

pls refer this video:

https://drive.google.com/file/d/1ansX6bU-w_nvntOAjWZqka3AiO8u0nNi/view?usp=drivesdk

next step will be????

Open an Innovative Issue on your Own ⁉ & try fixing it !!

Use your Creativity & Open up a New Issue relating to our Project & Fix it

Resolve the Bug when Adding Contributors via BOT

Try to resolve the Bug which is creating MERGE CONFLICTS when this bot is used.

Improve the Validation Accuracy of the Model

You can try to improve the accuracy of the model by :

• Tuning the Hyper Parameters like :
  Learning Rate, Epochs, Batch Size, Number of Layers in CNN, Number of filters ,Size of filters, Value in Dropout Layers, Optimizers
• Using different Datasets.

Think of Practical Applications and How to apply our project there ? 💡

Here are some popular Practical Applications of Facial Expression Recognition :

• During Interviews
• Students activeness during Online Education
• Real-time monitoring of physically disabled patients.

STEPS to Proceed :

1. Once you think of an Application, Try to figure out what additional Tech Stack is required to implement that.
2. If you are already familiar with the Tech Stack, You can start working on it.
3. If not and you are Interested in Learning those, We can provide resources or you can also explore online & the mentors will resolve queries.
4. Finally, Open up a new Issue describing the Application and Solution and Start CONTRIBUTING !!!

Suggesting Activity👨‍💻🏃‍♂️ to the user based on Emotion Predicted !

As there are multiple possibilities, First, You can restrict yourselves to some set of actions that can be suggested to user like

• Watch a Comedy Movie
• Go for a Walk
• Have a Talk with your family
• Share your happiness with your friends
• Play your Favorite Game
• Do Coding :)

Suggesting Music🎵 based on Facial Emotion predicted

Once the Model detects the Emotion of the User, It should provide an option to the user to listen Music that suits his expression !

CODE OF CONDUCT file

Hi @SanjayMarreddi
I found that this project repository is not having a code of conduct file in this repository ,
kindly assign me this issue and I would like to contribute a well formatted code of conduct file as you know code of conduct is a necessary file which the contributors need to follow so that everyone follows a decorum

Emotion classification using keras and building streamlit app for deployment!!

🚀 Feature

Emotion classification using keras and building streamlit app for future deployment.

Have you read the Contributing Guidelines on Pull Requests?

Yes

Motivation

The work that we build should be seen by others. Building streamlit app and deploying will ideally help !!

Build a Facial Recognition & Classifier Model 👨🏻‍🤝‍👨🏻👩‍🎓👩‍🏫👮‍♂️

• First of all, Fix your Set of Possible Persons in the Images
• Ex : You can take Possible Persons as Avengers or Popular Film Stars
• Then after you train the Model, It should be able to detect and recognize the Characters in the given Image

Add Contributing.md

I would like to add Contributing.md with the following features.

1.We can add Github documentation links for
i). Getting started with Git and GitHub
ii). Forking a Repo
iiI). Cloning a Repo
iv). How to create a Pull Request
2.Also, we can give the links for git installation
3.We can also add Style Guides for Git Commit Messages
These things will help people new to opensource understand clearly on how to create an issue and how to create a PR.

I would like to add this in a folder and incorporate it in the readme.

Face Expression Recognition using EfficientNet

🚀 Feature

I would like to Develop and contribute a model which performs Facial Expression Recognition using EfficientNet B7 Architecture.

Have you read the Contributing Guidelines on Pull Requests?

Yes

Motivation

Face Expression Recognition is a booming domain in Computer Vision and since the proliferation of EfficientNets , it is widely used for Classification tasks.

Can i get the Train and Test Data used to emotions Facial Recognition Model

📚 Documentation

(A clear and concise description of what the issue is.)

Have you read the Contributing Guidelines on Pull Requests?

(Write your answer here.)
Train and Test Data Used to Train Enotions Recognition model

Build your own "Facial" Recognition Model 🙂

• It should detect whether you are present in a given Image or not
• If you are present in the Image, It should Identify you & Label it accordingly

Deploying the Application 🌎

Deploy the entire Web App to a Global Server so that everyone can access it via a Link/URL

You may use Heroku as a platform

Update Contributors Table in README.md File

📚 Documentation

Like this

Have you read the Contributing Guidelines on Pull Requests?

Yes

PR TEMPLATE

🚀 Feature

I can add a pr template , which will have few checklists and options like uploading screenshots and describing about the pr etc , which will give better understanding of pr for the maintainer
(A clear and concise description of what the feature is.)

Have you read the Contributing Guidelines on Pull Requests?

Yes
(Write your answer here.)

Motivation

Self Motivation
(Please outline the motivation for the proposal.)

Pitch

This is a great feature like issue template in this project , this will give a better idea about the pr and also make the work easy
(Please explain why this feature should be implemented and how it would be used.)

Update readme.md

ADD ISSUE_TEMPLATE

Hello there,
I would like to add an issue template for your repository. This template would have four issues namely bug, documentation, feature, proposal. Kindly assign me this issue as a part of DWOC.
THANK YOU

Logo

💥 Proposal

(A clear and concise description of what the proposal is.)

Have you read the Contributing Guidelines on Pull Requests?

We can add a beautiful logo in your project to look it more attractive .

Frontend of the project or Deployment of model

I would like to contribute to this project in frontend or deployment of model to heruku using flask?
let me know can i contribute it ? I am part of DevScript winter of code (dwoc)

AttributeError: module 'utils' has no attribute 'datasets'

🐛 Bug Report

I get this error when i tried to train it on another dataset.
How can i solve this error :

AttributeError Traceback (most recent call last)
in
----> 1 utils.datasets.fer.plot_example_images(plt).show()

AttributeError: module 'utils' has no attribute 'datasets'

Beautify and Improve the README

Try to incorporate the following features :

1. Add all the Appropriate README Labels
2. Reduce the Content and enhance it with Images
3. Add a section for the Project Admin & Contributors
4. Other things which you feel to be important

Feel free to look at the Popular Project Repo's and incorporate the above points accordingly to our Project !!

Add a GitHub action to auto-assign Mentors for every PR

I would like to add a GitHub action to auto-assign mentors for every PR. It sometimes becomes hectic for the mentors to take a look at all the Projects for the PRs made by the Contributors. This action would give a nice notification about reviewing the pr and make things pretty easy for the mentor. It would automatically ask for reviews by the mentors on every pr created.

Demo :

Integrate welcome bot

I can add a welcome bot config file having a proper message that will show up when any user will open up an issue or pull-request for the first time.
For reference, check out: https://github.com/apps/welcome

If you feel this is a great idea!
Please assign it to me.

Auto Updating Contributors table

🐛 Bug Report

As of now the Contributors table is not up to date.
Moreover, it is hard coded so needs to be updated manually each and every time.
I wanna add a Contributors table which would update automatically.

Have you read the Contributing Guidelines on Pull Requests?

Yes

ImportError: libGL.so.1: cannot open shared object file: No such file or directory

ImportError: libGL.so.1: cannot open shared object file: No such file or directory
Traceback:
File "/usr/local/lib/python3.8/dist-packages/streamlit/script_runner.py", line 338, in _run_script
    exec(code, module.__dict__)
File "/root/Emotion-Investigator/main.py", line 3, in <module>
    import cv2
File "/usr/local/lib/python3.8/dist-packages/cv2/__init__.py", line 5, in <module>
    from .cv2 import *

how solved?

Enhance the Frontend & Make it more Attractive 🔥

• Mention the proper Title, Headings
• Add Colorful Background & Animations
• Insert Buttons to go for Graphical Visualization instead of typing URL