Welcome to AI Blitz XIII! 🚀 | Starter Kit For This Challenge! 🛠| Baseline For This Challenge

Community Contribution Prizes 📓  |  Find Teammates 👯‍♀️

Discord AI Community 🎧

Overview

Emotions are an integral part of being a human being. We express our thoughts and feelings throughout our day through positive, negative, or neutral emotions. Our eyes quickly pick up on these emotions by looking at facial expressions, but can we train a model to do that?

🕵️ Problem Statement

For this Sentiment Classification puzzle, participants are given embedding generated from an input image of a face and your model needs to classification the expression in one of the three categories - negative, neutral and positive.  

Example,

💪 Getting Started

This puzzle is a classification problem and has similarities with the Emotion Detection problem from AI Blitz 9. The emotion Detection problem aims to teach a computer to distinguish between positive and negative emotions. There we were using sentences here we are using embedding from the images. Can you use the resources and tools of that problem to come up with a unique solution for this puzzle?

Our Starter Kit comes with the random prediction and our baseline uses Random Forest Classifier to classify the embedding in different emotions. You can check it out here.

💾 Dataset

In this dataset, instead of giving an image to classify the face sentiment, we give the face image embeddings in the dataset generated using popular deep learning models.
 
The dataset is split into 3 different sets - train, validation & test set. The training & validation set will be used in training your models and the predictions generated from the test set will be used to evaluate your model.   Each set contains 5000, 2000, and 3000 samples respectively.

The training and validation set contains a csv file. The csv file contains two columns as below -

• embeddings - The face image embeddings represented in the column is a list of 512 elements,
• label - The sentiment of the face. There are over 3 classes, negative, neutral, and positive.

• embeddings	label
  [0.24, 0.56 … ]	positive
  [064, 0.28 … ]	neutral

• The test set will contain only a zip file with 3k samples for evaluation.

📁 Files

Following files are available in the resources section:

• train.csv - ( 5k samples ) This CSV file contains the embeddings and labels for the training set.
• val.csv - ( 2k samples ) This zip file contains the embeddings and labels for the validation set.
• test.csv - This CSV file contains the embeddings of the testing set.  
• sample_submission.csv - This CSV file contains the format of your CSV file for submitting the results.

🚀  Submission

Learn to make your first submission using the starter kit 🚀

• Create a submission folder in your working directory.
• Use sample_submission.csv provided in the resources section and replace the label column values with your model predictions for the corresponding embeddings.
• Save the CSV in the submission folder as  submission.csv  
• Inside a submission directory, put the .ipynb notebook from which you trained the model and generate predictions and save it as notebook.ipynb.
• Zip the submission directory

Overall, this is what your submission directory should look like

submission
├── assets
│   └── submission.csv
└── original_notebook.ipynb

Make your first submission here 🚀 !!

🖊 Evaluation Criteria

During the evaluation, the F1 Score ( average=weighted ) as the primary score and the Accuracy Score as the secondary score will be used to test the efficiency of the model. 

📱 Contact

• Shubhamai