| **You can download the initial exercise file from [link](https://quera.org/contest/assignments/103149/download_problem_initial_project/356098/).** | | ------------------------------------------------------------------------------ | Sina is a student and also goes to work at the same time. With the arrival of the cold season and the phenomenon of temperature inversion, he checks the news every morning with stress. Sina is highly hoping that due to Tehran's air pollution, universities and offices will be closed, and his classes and exams will be held virtually so he can stay at home and get to his tasks more easily!  But checking the news is not fast enough, and Sina wants to always be one step ahead. He has decided to use meteorological data to build a powerful AI model to predict the air pollution status in the coming days. Your mission as a data scientist is to help Sina. Using an hourly dataset containing weather information, you must design a model based on machine learning or time series that can predict the concentration of particulate matter (AQI) as accurately as possible based on these features. | **Column Name** | **Description** | | ---------------- | -------------------------------------------------------------------------------- | | `No` | Unique row identifier (record number) | | `timestamp` | Time and date of record entry as a text string | | `AQI` | **(Target Variable):** Concentration of particulate matter less than 2.5 microns | | `DEWP` | Dew point (Celsius) | | `TEMP` | Temperature (Celsius) | | `TEMP_F` | Temperature (Fahrenheit) | | `PRES` | Air pressure (hPa) | | `sensor_voltage` | Measurement sensor voltage at the time of recording (millivolts) | | `cbwd` | Dominant wind direction | | `Iws` | Cumulated wind speed (meters per second) | | `Is` | Cumulated hours of snow | | `Ir` | Cumulated hours of rain | ### Evaluation Metric The judging system first calculates the Mean Absolute Error (_MAE_) between your predictions and the actual values in the system: $$MAE = \frac{1}{n} \sum_{i=1}^{n} \vert{}y_i - \hat{y}_i\vert{}$$ \*_Final Score Formula:_\* Due to the severe fluctuations of the pollution index, your final score on the leaderboard is calculated using a descending exponential function. This score will be a decimal number where the lower your error, the closer it gets to 100: $$Final\_Score = 100 \times e^{-MAE / 100}$$ > In this formula, the number 100 is a constant coefficient (C). With this evaluation metric: > > + If you build a perfect model and $MAE = 0$, your score will be **100**. > + The larger your error gets, your score will approach zero. ### Predictions Output Format Your model must make its predictions for all rows present in the test file (`test.csv`). In the initial notebook file, code has been provided that saves your model's output in the format of a file named `submission.csv`. This file must definitely include the following two columns: \* `No`: Row identifier (exactly the same as the test file) \* `AQI`: Your prediction for the pollution concentration (AQI) The table below shows an example of the output file: | **No** | **AQI** | | ------ | ------- | | 35065 | 21.0 | | 35066 | 63.0 | | 35067 | 75.0 | | 35068 | 10.0 | | 35069 | 99.0 | ### How to Submit Your Answer After running the result generator cell (the last cell of the notebook file), submit the generated `result.zip` file. This zip file will automatically include your notebook and the `submission.csv` file. Please note that you must save the changes made in the notebook using the `ctrl+s` shortcut before running the result generator cell; otherwise, your **score** might not be calculated correctly at the end of the competition. Also, if you are using Google Colab to run this notebook file, download the latest version of your notebook and place it inside the submission file instead of the previous notebook before submitting the `result.zip` file.