I assume you get something like feature matrix, where each points represents to a certain value. Something like this

First you can use unsupervised learning and have nice separation between the examples,

but if you need just a simple recognition algorithm,

you can use **logistic regression** or very simple and small **classification network**.

Both approaches are implemented in scikit-learn package and you can just import and use it.

Here is a simple example of logistic regression

```
from sklearn.datasets import load_iris
from sklearn.linear_model import LogisticRegression
X, y = load_iris(return_X_y=True)
clf = LogisticRegression(random_state=0, solver='lbfgs',
multi_class='multinomial').fit(X, y)
clf.predict(X[:2, :])
clf.predict_proba(X[:2, :])
clf.score(X, y)
```