Draw experiments independent samples of size sample_size from an $\mathrm{Exponential}(1)$ distribution, which is strongly right-skewed and nothing like a normal. Return the list of sample means, one per experiment.

Plotted, those means look normal even though the thing they came from does not. That is the central limit theorem, and it is why a z-test on a mean works for data that is not itself normal.

Seed np.random.seed(seed) inside solve, immediately before drawing.

Input

sample_size = 30
experiments = 5
seed = 0

Output

[1.2094295641207409, 0.8167041779847283, 0.8325754884745505, 1.0290426928963081, 1.1916098935280082]