You will be evaluated using a proximity scoring function. The evaluation metric contains a circular region surrounding the motor. If your x,y coordinate prediction is within 20 pixels of the original truth (x,y), you will be given full credit for that image. There is a linear dropoff depending on how far your guess is from the full-credit scoring region, and a threshold radius of 40 pixels at which you are given a score of 0. 
If the truth label is "no motor visible" and the prediction label is "no motor visible" a score of 1 is received. If the truth label is "no motor visible" and the prediction label is "motor visible" a score of 0 is received.
If the truth label is "motor visible" and the prediction label is "no motor visible" a score of 0 is received.

Scores for all the images will be averaged to one final culminating score.

##Submission Format
You should submit a csv file with the following as column headers, in this order:
- `image_name`: name of corresponding image
- `x`: x coordinate of the location of the flagellar motor
- `y`: y coordinate of the location of the flagellar motor
- `motor_visible`: this column should contain a 1 or a 0. 1 represents a motor present and 0 the absence of a motor
For example:
```
image_name,x,y,motor_visible
0001.png,50,100,1 
```
This indicates that image Hylemonella_20009 has a motor present at the coordinates (50, 100).
An example of an image containing no motor:
```
0002.png,-1.0,-1.0,0
```
This indicates that image Hylemonella_20009 has no motor present.