Readme of these three datasets:
For example, in 'circle.csv', [Xcircle_X1,Xcircle_X2] are the features of each data points.
[Xcircle_X1[0],Xcircle_X2[0]] is the data point.

For each data sets, we are asking for:
1. draw the distribution of dataset
2. run the DBSCAN from sklearn and find the best parameter to cluster the dataset, and color each point by your cluster labels.
3. Implement your own DBSCAN, run it on these three datases and color them.