Collecting metadata allows you to easily visualize your key performance metrics as a time series graph or a scatter plot.
You can also generate more complex graphs inside your execution and save them as Valohai outputs.
import matplotlib.pyplot as plt
import numpy as np
import valohai
np.random.seed(19680801)
data = np.random.randn(2, 100)
fig, axs = plt.subplots(2, 2, figsize=(5, 5))
axs[0, 0].hist(data[0])
axs[1, 0].scatter(data[0], data[1])
axs[0, 1].plot(data[0], data[1])
axs[1, 1].hist2d(data[0], data[1])
save_path = '/valohai/outputs/myplot.png'
plt.savefig(save_path)
plt.show()
plt.close()