MachineHack, in association with Wipro, is lining up a two-week-long hiring hackathon starting at 6.00 PM IST on 25 January, 2022. The challenge is to build AI/ML models to forecast the Global Horizontal Irradiance (GHI) from the given datasets. The winners stand a chance to bag cash prizes worth INR 3.5 lakh and get hired by Wipro.
Wipro Limited is a leading global information technology, consulting and business process services firm. The Fortune 500 company mobilises the power of cognitive computing, hyper-automation, robotics, cloud, analytics and emerging technologies to drive digital adoption of their clientele. Wipro is renowned for its comprehensive portfolio of services, strong commitment to sustainability and good corporate citizenship, and has over 220,000 employees serving clients across six continents.
Though a little late in the day, the world is waking up to the deleterious effect of fossil fuels on our environment. As the doomsday clock ticks away, human beings are turning to renewable energy to avert a possible apocalypse. Solar energy is projected to reach $223.3 billion by 2026, growing at a CAGR of 20.5% from 2019 to 2026. Fortunately, the sun is a well-spring of clean energy.
Along with being a global leader in artificial intelligence services from the latest reports of analysts like Forrester, IDC and Everest Group, Wipro has been rated as the second-best organization for data scientists to work in India in 2021 by Analytics India Magazine. The company has also been committed to reaching a Net-Zero Greenhouse Gas Emissions by 2040.
Taking the cue, Wipro, in association with MachineHack, has designed a forecasting challenge to optimise solar power generation using AI/ML models.
Let the challenge begin!
A solar power generation company wants to optimise solar power production and needs the prediction model to predict the Clearsky Global Horizontal Irradiance(GHI). The data is ten years at an interval of every 30 mins with the following data points:
[‘Year’, ‘Month’, ‘Day’, ‘Hour’, ‘Minute’, ‘Temperature’, ‘Clearsky DHI’, ‘Clearsky DNI’, ‘Clearsky GHI’, ‘Cloud Type’, ‘Dew Point’, ‘Fill Flag’, ‘Relative Humidity’, ‘Solar Zenith Angle’, ‘Pressure’, ‘Precipitable Water’, ‘Wind Direction’, ‘Wind Speed’]
MachineHackers are required to predict ‘Clearsky DHI’, ‘Clearsky DNI’,’Clearsky GHI’ values for a year for the company to get the maximum yield of the solar energy.
The hackathon is open to data scientists, machine learning practitioners, analytics professionals, and tech enthusiasts at all experience levels. Apart from a chance to win cash prizes, the participants also get the opportunity to improve their Global Leaderboard Rankings and become the ultimate MachineHack GrandMaster. What are you waiting for? The holy grail is just one tap away!
Hackathon start date – January 25, 2022
Hackathon end date – February 14, 2022, at 6:00 pm.
Click here to participate
Problem Statement:
The winners of the hackathon stand a chance to win cash prizes worth up to INR 3.5 lakh.
MachineHack has created a training dataset of 1,75,296 rows and 18 columns and a testing dataset of 17,520 rows and 15 columns. The hackathon demands a few pre-requisite skills like time series forecasting, multi-label prediction, and the ability to optimise “MSE” to generalise well on unseen data.
Datasets will go live on January 25, 2022
Click here to participate
Submission guidelines
- Sklearn models should support the predict() method to generate the predicted values.
- The participant should submit a .csv file with exactly 17,520 rows with 3 column [‘Clearsky DHI’, ‘Clearsky DNI’,’Clearsky GHI’]. The submission will return an Invalid Score if you have extra rows or columns.
- The file should have exactly 3 column.
Note: Do not shuffle the sequence of the test series.
If you are using pandas, use this submission code:
submission_df.to_csv(‘my_submission_file.csv’, index=False)
Evaluation criteria
The submission will be evaluated using the Mean Squared Error. One can use sklearn.metrics.mean_squared_error to calculate the same
The evaluation will be done using
- It will be based on the participants standing on the private leaderboard.
- The public leaderboard uses 30% of the provided test.csv dataset to evaluate.
- The private leaderboard uses 100% of the provided test.csv dataset to evaluate.
Prizes
Wipro’s Sustainability Machine Learning Challenge will select three (3) winners based on the given evaluation criteria. The cash prizes are as follows:
First Prize: INR 2,00,000
Second Prize: INR 1,00,000
Third Prize: INR 50,000
Note:
- Make sure your MachineHack profile is up to date with all the relevant information/details.
- Make sure you have gone through the ‘Rules’ section before participating.
- The participants will receive the prize money only if selected by Tredence and MachineHack.
Hackathon start date – January 25, 2022
Hackathon end date – February 14, 2022, at 6:00 pm.
Click here to participate
Attribute Description:
# Train: 1,75,296 rows x 18 columns
# Test: 17,520 rows x 15 columns
Data Attributes:
- ‘Year’,
- ‘Month’,
- ‘Day’,
- ‘Hour’,
- ‘Minute’,
- ‘Temperature’, 0C
- ‘Clearsky DHI’, w/m2
- ‘Clearsky DNI’, w/m2
- ‘Clearsky GHI’, w/m2
- ‘Cloud Type’,
- Cloud Type 0 Clear
- Cloud Type 1 Probably Clear
- Cloud Type 2 Fog
- Cloud Type 3 Water
- Cloud Type 4 Super-Cooled Water
- Cloud Type 5 Mixed
- Cloud Type 6 Opaque Ice
- Cloud Type 7 Cirrus
- Cloud Type 8 Overlapping
- Cloud Type 9 Overshooting
- Cloud Type 10 Unknown
- Cloud Type 11 Dust
- Cloud Type 12 Smoke
- Cloud Type 15 N/A
- ‘Dew Point’, C
- ‘Fill Flag’,
- Fill Flag 0 N/A
- Fill Flag 1 Missing Image
- Fill Flag 2 Low Irradiance
- Fill Flag 3 Exceeds Clearsky
- Fill Flag 4 Missing CLoud Properties
- Fill Flag 5 Rayleigh Violation
- Fill Flag any N/A
- ‘Relative Humidity’, %
- ‘Solar Zenith Angle’, Degree to calculate cos(θ)
- ‘Pressure’, mbar
- ‘Precipitable Water’, cm
- ‘Wind Direction’, Degrees
- ‘Wind Speed’ m/s
Skills:
- Time series forecasting
- Multi-label prediction
- Optimizing MSE
Winners Announcement
Final winners will be notified via email based on an aggregate score of their private leaderboard rankings.