Add Iris.

import os

import argparse

import time

import pickle

import random

import itertools

import numpy as np

import pandas as pd

from sklearn import datasets

import models

random.seed(123)

np.random.seed(456)

parser = argparse.ArgumentParser(description="Convergence of bottleneck NNGP to NNGP in log-likelihood.")

parser.add_argument("--depths", "-d", dest="depths", type=lambda s: [int(i) for i in s.split(",")], default="1,1", help="comma-separated depths of bottleneck components.")

args = parser.parse_args()

# load Iris data

X, y = datasets.load_iris(return_X_y=True)

# one-hot encode

Y = np.zeros((len(y), 3))

for i in range(3):

	Y[y==i, i] = 1

# standardize

X = (X - np.mean(X, axis=0, keepdims=True))/np.std(X, axis=0, keepdims=True)

Y = (Y - np.mean(Y, axis=0, keepdims=True))/np.std(Y, axis=0, keepdims=True)

n_samples = 10**3

widths = [1, 2, 3, 4, 5, 10, 50, 100, 500, np.inf]

depths = args.depths

# variance hyperparameters

v_b = 1.0

v_w = 1.0

v_n = 1e-4

liks = []

for width in widths:

	print("width:", width)

	ds = [sum(depths)+len(depths)-1] if width == np.inf else depths

	ws = [] if width == np.inf else [width]*(len(depths)-1)

	prior = models.bottleneck_nngp_prior(X, output_dims=Y.shape[1], v_b=v_b, v_w=v_w, v_n=v_n, depths=ds, widths=ws, bottleneck_activation=True)

	l = prior.log_likelihood(Y, n_samples=n_samples)/X.shape[0]

	liks.append(l)

liks = list(map(lambda x: np.round(x, 3), liks))

data = pd.DataFrame.from_dict({"width": widths, "lik": liks})[["width", "lik"]]

# write to CSV

os.makedirs("results/iris", exist_ok=True)

filename = "results/iris/depth"+"-".join(map(str, depths))+".csv"

data.to_csv(filename, index=False)

print("Done!")

 No newline at end of file