import numpy as np
import matplotlib.pyplot as plt
from scipy.stats import gaussian_kde
[docs]
def scatter_plot(pred: np.ndarray = None, target: np.ndarray = None):
fig, ax = plt.subplots()
ax.scatter(x=pred, y=target, s=25, c='g', alpha=0.5, marker=".")
ax.set_title('Prediction VS Target')
ax.set_aspect('equal')
min_val, max_val = np.min([target, pred]), np.max([target, pred])
ax.plot([min_val, max_val], [min_val, max_val],
ls="--", linewidth=1, c='r')
plt.xlabel('Prediction', fontsize=15)
plt.ylabel('Target', fontsize=15)
plt.xticks(fontsize=15)
plt.yticks(fontsize=15)
return fig