Adbot Ad Engagement Forecasting Challenge
'%3e%3cpath fill='%23001489' d='M0 0v6h9V0z'/%3e%3cpath fill='%23e03c31' d='M0 0v3h9V0z'/%3e%3cg stroke='%23fff' stroke-width='2'%3e%3cpath id='d' d='m0 0 4.5 3L0 6m4.5-3H9'/%3e%3cuse xlink:href='%23b' stroke='%23ffb81c' clip-path='url(%23c)'/%3e%3c/g%3e%3cuse xlink:href='%23d' fill='none' stroke='%23007749' stroke-width='1.2'/%3e%3c/g%3e%3c/svg%3e)
Helping South Africa
$500 USD
Completed (over 2 years ago)
Forecast
456 joined
112 active
Start
04 Apr 24
Close
19 May 24
Reveal
19 May 24
Sample submission file
Help Ā· 11 May 2024, 18:04 Ā· 7
Hey guys!!
How can i make predictions for the dates in the sample submission file.
Hi Maven,
A starter-notebook is available to help you make your first submission.
https://zindi.africa/competitions/adbot-ad-engagement-forecasting-challenge/discussions/20705
This is my first time attempting a time series problem, when i trained my model, the training set had 14 feature columns, is this how the training set is supposed to be?
In your notebook, your training set had just two columns; date and target, does it mean the other features were not used in training the model?
This is a forcasting challenge, so we don't know what the future features would be. We can only rely on the train dataset to build a predictive model, then pass in the dates to predict the number of clicks for the future.
Just to get you clear, basically we use the features in the dataset to train our model and we just need to extract the dates from the submission sample to predict the clicks?
Thank you for your patience
Yes.
When i do this I'll get a dataframe consisting of single column that contains the dates I'm predicting for. But when i try to predict i get a feature mismatch
Because originally the data for testing should contain the same features as the train. I am just confused