- repo with code of Qualifier Data Science Challenge Round of Hindustan Unilever Hackathon.
- Team name: null_shredders:
- Rank: 38 | Score: 92.017971
- Machine Learning technique used: XGBoost with GridSearch
Competition Official Website: Click Here
Machine Learning Challenge Objective of the problem: The objective of the problem is to predict values “current price” attribute from the given features of the Test data. The predictions are to be written to a CSV file along with ID which is the unique identifier for each tuple. Please view the sample submission file to understand how the submission file is to be written. Please upload the submission file to get a score.
Description of files:
V. id: Unique vehicle id
On road old: On road price of the vehicle when purchased from showroom in rupees
New on road: new on road price in rupees
Years: Vehicles age
km: total distance covered by vehicle in km
Rating: Overall rating of the new vehicle out of 5
Condition: current condition of the vehicle out of 10(note :- higher the number better the condition is)
Economy: current fuel economy of the vehicle per liter.
Top speed: current top speed of the vehicle indicated by dyno test.
Hp: horse power of the engine indicated by dyno test.
Torque: torque of the engine indicated by dyno test.
Current price: predicted price of the vehicle .
Evaluation Criteria: Normalised root of MSE . All values would be normalised to 100. Any member of a team may make a submission and 15 submissions are allowed per team. All teams must submit their projects to the code submission stage.
Evaluation Algorithm: Root Mean Square Error (RMSE) normalization_constant 100000