Welcome to Attentive AI Internship Hiring Challenge!

*******

Important Update from the Organizer: "There was a mix up in the solution key from our end, now it has been updated"

*******

Attentive AI is hiring data science interns through this challenge. The aim of this challenge is to build an image classification model for a satellite image dataset consisting of 6 classes. Dataset consists of varied dimension satellite images. The dataset is divided into two folders named train(for training) and test (for prediction). The zip also contains two CSV files, the format of the train CSV file is self-explanatory.

You can use an ensemble of multiple models to improve results. The image size varies and you will have to find a strategy to train a model that will be giving a good accuracy. Apart from the test accuracy, the creativity of your solution (which includes augmentation techniques, ensemble method, handling image size variation, loss function) will also be evaluated by the reviewing team.

The 6 classes contained in the dataset are:

• Adhered
• Concrete
• Plastic & fabric
• Steel
• Shingle
• Ballasted

For the final submission format, please refer to "Submission Guidelines".

Here is a Quick Start Guide.

*******

Important Update from the Organizer: "There was a mix up in the solution key from our end, now it has been updated"

*******

The final submission must include:

output.csv (single file having results for "test" directory).

NOTE: You are also required to upload your .ipynb notebook in the "My Submissions" section.

The first row should correspond to the column names "0" and "1". The zip contains test_challenge.csv which is a sample submission file, the first column is the name of the image in the "test" folder and the second column right now filled with one of the classes should be replaced with the final prediction of your model which is one of the 6 classes.

Sample Output:

...

How to make a submission?

1. Click on "My Submission"
2. On the next page, click on "+ New Submission"
3. Upload your CSV on the next page and click on "Submit for Review"

Please note:

1. You must submit your CSV file by uploading the CSV in the "My Submissions" section of this challenge.
2. Your submission will be auto-graded and you will be able to see your results instantly.
3. If there is an error in the submission, your final score will be marked as 0 or a warning prompt of “Invalid Score” will be displayed.

You can use any model and training strategies, there is no limit to the number of models you can use for the ensemble.

Judgment

1. F1 Accuracy Score
2. If the scores are tied, the person reaching the score FIRST will get the better rank.

Rules

1. We are using an automatic anti-cheat system. Our system flags submission which has low trust scores. Manually modifying your submission CSV (via comparison of any form etc.), using image comparison techniques (pixel matching, file size matching, etc,), in any form, can lead to your disqualification without any notice.
2. The participants must use only the provided dataset for training.
3. The submission should be in a proper format as described by "Submission Guidelines".
4. Late submission will not be accepted beyond the provided deadline (Indian Standard Time).

NOTE: We may request the code files if there is any discrepancy in your score. Your score will be marked invalid or you can be disqualified if the request for code is not fulfilled by you.

Rewards:

1. The top 10% of candidates will be called for interview and x will be awarded an internship at attentive.ai
2. The participants with a successful submission will get a certificate of participation.
3. The top 3 participants will receive a permanent place in Dockship's "Hall of Fame".