Optimized ML Model for Emotion Detection in Low-Resource Settings

Optimized ML Model for Emotion Detection in Low-Resource SettingsOptimized ML Model for Emotion Detection in Low-Resource Settings

The network for creativity

Join 1.25M professional creatives like you

Connect with clients, get discovered, and run your business 100% commission-free

Creatives on Contra have earned over $150M and we are just getting started

Back to feedPost

David Okosa

• Mar 4

Optimized machine learning model, works in extremely resource constrained environments while retaining good accuracy and recall. Detects 7 emotions. Utilizes NCNN (a high perfomance neural network framework), to achieve 79% accuracy @ 90fps.

Machine Learning Engineering C++PyTorch

Sanket Sabharwal, PhD

pro

• Apr 2

Looks like the contra community is liking stuff on machine learning, and I thought it would be interesting to share another article we put together about why machine learning models can underperform for... just no obvious reason.

Despite having clean data and tuned hyperparameters, the accuracy graph barely moves. The trick here is everything related to feature engineering. We wrote a piece titled “Feature engineering decides machine learning outcomes” to share our thoughts on this quasi-"mystic" exercise.

I call it quasi-"mystic" because it is part art and part science. You have to be able to see through the rubble of messy data and know what will really move the needle for you. Much like the jeweller in our cover. It breaks down the essentials, from understanding your data’s structure to choosing well-validated features. Read the full piece here: https://www.algorithmic.co/blogs/feature-engineering-decides-machine-learning-outcomes

Machine Learning Engineering AI Engineer softwareengineer

Daniel G Bright

max

• Apr 2

Wow, this is sick. the cover as well as the article

Asad Aziz

• Apr 1

🚀 SMS Spam Detection Project

I’m excited to share one of my machine learning projects — an SMS Spam Detection System 📩

🔍 The goal of this project is to automatically classify incoming messages as spam or ham (legitimate messages).

💡 Key Features:

Machine Learning-based model

Text preprocessing & feature extraction

Accurate spam classification

Simple and user-friendly interface using Streamlit

🛠 Technologies Used:

Python

Scikit-learn

Pandas & NumPy

Streamlit

🌐 The project is available on GitHub and also deployed using Streamlit.

📎 Project Link: https://sms-spam-detection-001.streamlit.app/

🙏 Feel free to check it out and share your feedback. Your suggestions will help me improve further!

#MachineLearning #Python #AI #DataScience #Streamlit #GitHub #Learning

AI Development AI Engineer Python scikit-learn Streamlit Machine Learning Engineering

Marco Mastroeni

• Apr 1

Nice work! It seems really interesting. Some time ago i did something similar but for the news online. i sued textblob, nltk and vader for sentiment analysis. Did you use streamlit for the categorization of the words?

Ismel Figueroa

max

• Apr 1

I got accepted into something rare 🫢 A network where human taste shapes AI, for real. Most AI tools are built to be fast, but a few are built to be good. That distinction matters more than we think. I just joined @Contra Labs, the frontier human data and evaluation lab helping make AI better for creativity by centering the one thing algorithms can't fake: (human judgment) thanks @Contra HQ