Pragyan ML Challenge

With the passage of time, various companies in various sectors are set up. In order to thrive in your field and “live another day”, you have to be smart. Unleash the coder in you to outrun all other business giants. As a developer, create a program that analyzes the current data sets to predict the future.

Prize Money INR 30,000

Round-1 ended on 2^nd March 2020

NOTE: Emails have been sent to selected candidates.

Round-2: Starts 12^th March 2020

Round 1:

An online round to test the knowledge of participants in algorithms, machine learning, and aptitude.
Consists of 30 questions.
Participants are shortlisted on the basis of score and time taken.

Round 2:

This will be a single round online event for 2 days.
You will be provided with a database. Using this, you will have to make the program that is suitable for the given problem statement.
You have to submit this file containing the program having .csv extension.
File description:
3 files will be provided.

1) Training.csv → the training dataset

2) Test.csv → the test dataset

3) Sample Submission.csv → a sample submission file with the correct format.

The languages allowed are Python and R.
Registration of the event will close two days before the event ends.
Participants should work independently; sharing answers or collaborating with another participant to secure multiple spots at finals constitutes cheating.
No queries will be entertained apart from technical failures.
Results are not subject to discussion and the judges’ decision will be final.

Judging criteria:

Root Mean Squared Error (RMSE):

Submissions are scored on the root mean square error. RMSE is defined as:

Where, y that is the natural-log of the predicted value and y^(y dash) is the natural-log of the actual summed value plus one.

“Lesser the RMSE, more the points received.”

Sample problem statement:

You will be provided with the dataset of a company. Using this you will have to make a program to predict the capitalisation and credit score rating of the company. You have to submit the file with .csv extension. The participants will be ranked based on their accuracy. The participant whose answer is close to that of the actual answer will be declared as the winner.

FAQs:

1) In case of a tie, what will be the criteria for judgement?
In case of a tie in Bleu metric, time of submission will be taken into account.

2) Who can participate in this event?
It is open to all students.

3) Is it a team event?
No. It's an individual event.

Contacts:

1) Lakshmanan: 97904 32692
2) Madhavan: 73580 32649
3) Mandeep: 99944 79069

You must be logged in to fill this form. Click here to login.