Welcome! I'm assuming you came here to give this coding challenge a try, great! After all, you could be the one to take home the big prize... So, let's get right to it!
You can either:
-
Open the Colab example
OR
-
Clone this repo
git clone https://github.com/godatadriven/dropblox.git
and open
quick_start.ipynb✓
Check out the Quickstart example in the provided notebook on how to generate the solutions file.
The task is simple: given a rectangular field and a set of predefined blocks, fill the field with the blocks in a way to maximize your score.
How do you score high? That depends on the colors of the blocks you used, the rows you filled and the amount of empty tiles in the final field.
You are given a Field object field with field.height = field.width = 100. Initially it is empty, but you can drop blocks in the field which will change the array field.values. To make it visual, simply print your field and view the result in a text editor.
Blocks have a shape, a color and a unique ID. They are defined as instances of the class Block. The list of available blocks is defined as blocks in the provided notebook. Similar as with the field, we can visualize the blocks. The first block, B = blocks[0] with B.block_id = 0, looks like this:
🟧🟧🟧
🟧⬜️🟧
⬜️⬜️🟧
This is an orange block, so B.color = "O", and it has B.width = B.height = 3.
Blocks can be dropped into the field from top to bottom, and they stop falling until they hit another block or the bottom of the field. Blocks are dropped in the field as is, so they cannot be rotated. Also, a block can only be dropped once, but you don't have to use all of the provided blocks.
Okay, now we can drop this block in our field, but how do we decide where to drop it?
Each color has a number of points and a multiplication factor. These will determine your score, once you dropped some of blocks in the field. The values for each color are defined under rewards. For example, for orange, we have rewards.points["O"] = 5 and rewards.multiplication_factors["O"] = 3.
This means: for every orange tile in our final field, we get +5 points. So just by dropping the orange block from above we would already get +30 points.
With the multiplication factor, things can get crazy: if the entire row is filled with tiles of the same color, you get the values of those tiles multiplied by the corresponding factor.
Lastly, you can get minus points. If a row contains at least one colored tile, all empty tiles in that row count as -1 towards your final score.
That should help you decide which blocks to drop, and where!
The rewards are like follows:
| Emoji | Color code | Points | Multiplication factor |
|---|---|---|---|
| 🟦 | "B" | 100 | 0 |
| 🟨 | "Y" | 1 | 3 |
| 🟩 | "G" | 3 | 1 |
| 🟪 | "P" | 5 | 2 |
| 🟧 | "O" | 5 | 3 |
| 🟥 | "R" | 8 | 4 |
Your solution should be a .txt file where each line specifies which block (by block ID) to drop in the field, and at which x-coordinate. To clarify, the bottom left of the block will be dropped at the provided x-coordinate in the field.
For exmple, you could write solution.txt with:
322 54
12 4
...
which tells us to first drop the block with block_id = 322 at x-coordinate 54, and then drop the block with block_id = 12 at x-coordinate 4, etc.
Submit your solution .txt at: https://dropblox.azurewebsites.net/submit.
Note: loading the website may take a moment - it sleeps when not used. If loading fails, please try again.
Good luck! 💪🏻
https://dropblox.azurewebsites.net/leaderboard
This challenge is hosted by GoDataDriven for PyData Eindhoven 2022.