Using conda, in your target env:
python -mpip install matplotlib
import numpy as np
import matplotlib.pyplot as plt
def plot_metrics(x, y1, y2):
idx = np.arange(len(x))
w = 0.4
plt.figure(figsize=(12, 10))
plt.bar(idx, y1, w, label='Series 1')
plt.bar(idx + w, y2, w, label='Series 2')
plt.ylabel('Count')
plt.title('My Title')
plt.xticks(idx + w / 2, x, rotation="vertical")
plt.legend(loc='best')
plt.show()
dates = ['2019-01', '2019-02', '2019-03', '2019-04', '2019-05']
series_a = [3, 4, 2, 5, 6]
seriea_b = [3, 7, 9, 14, 20]
plot_metrics(dates, series_a, seriea_b)