Round 1: Completed #educational Weight: 25.0
2310
242
14
176

## 🕵️ Introduction

It’s not uncommon in high speed overtakes and turns for cars to go out of control and spin out of the track. Either caused by cars colliding with one another or poor grip between the tyre and track, these spins can change the position of the car.

In such a scenario, how do you locate and fix the car position? For this task, you’ll be given images of F1 cars and your AI model needs to output the rotation of the car. This multi-class classification problem will require you to identify what rotation, out of the four categories (left, right, front, back) is the car in? You can access the starter kit over here

## 💾 Dataset

The given dataset contains images showing different rotation of F1. Size of each image is 265*256 in jpg format. The images in train.zip and val.zip  have labels - front, back, left and right in their corrosponding csv files.  The labels for the images in test.zip needs to be predicted. One thing to note that the rotations are only from one point of reference.

## 📁 Files

Following files are available in the resources section:

• train.zip - (40000 samples) This zip file contains f1 images with images name corresponding to ImageID column of train.csv
• train.csv - (40000 samples) This csv file contains the ImageID column corresponding to train.zip and label column as the rotation of F1.
• val.zip - (4000 samples) This zip file contains f1 images with images name corresponding to ImageID column of val.csv
• val.csv - (4000 samples) This csv file contains the ImageID column corresponding to val.zip and label column as the rotation of F1.
• test.zip - (10000 samples) This zip file contains f1 images which will be used to evaluate the performance of the model

## 🚀 Submission

• Prepare a CSV containing ImageID and label column as the predicted rotation of F1 Car.
• The name of the above file should be submission.csv.
• Sample submission format available at sample_submission.csv in the resources section.

Make your first submission here 🚀 !!

## 🖊 Evaluation Criteria

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