HSV color chart for vallejo paint

				
					import matplotlib.pyplot as plt
import matplotlib.colors as mcolors
import colorsys
from mpl_toolkits.mplot3d import Axes3D
import numpy as np
from sklearn.cluster import KMeans

# Step 0: Your list of hex colors
hex_colors = [
    "#A54336", "#DC5E35", "#F8C85E", "#F8DB5D", "#DDD1A4", "#BBB992", "#9A835F",
    "#99624B", "#805C3F", "#7F6355", "#544231", "#3B2C29", "#232A2C", "#3C3431",
    "#402C33", "#372743", "#5A8B78", "#32665A", "#324A32", "#7C7B66", "#757C70",
    "#9AA0A0", "#DDDED7", "#FDF8F3", "#3B4D77", "#3A4372", "#3A2F4B", "#252527"
]

# Step 1: Convert hex colors to HSL
hsl_data = [colorsys.rgb_to_hls(*mcolors.hex2color(c)) for c in hex_colors]
hsl_array = np.array(hsl_data)  # shape: (N_colors, 3)
hue = hsl_array[:, 0]
lightness = hsl_array[:, 1]
saturation = hsl_array[:, 2]

# Step 2: Basic statistics
mean_hsl = np.mean(hsl_array, axis=0)
std_hsl = np.std(hsl_array, axis=0)

print("=== Basic Statistics ===")
print("Mean HSL:", mean_hsl)
print("Std HSL: ", std_hsl)
print()

# Step 3: Histograms of Hue, Saturation, Lightness
fig, axs = plt.subplots(1, 3, figsize=(12, 4))
labels = ["Hue", "Lightness", "Saturation"]
data = [hue, lightness, saturation]
for i in range(3):
    axs[i].hist(data[i], bins=10, color='gray', edgecolor='black')
    axs[i].set_title(f"{labels[i]} Distribution")
    axs[i].set_xlim(0, 1)
plt.suptitle("Histograms of HSL Components")
plt.tight_layout()
plt.show()

from sklearn.cluster import KMeans
import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D

# Step 4: KMeans clustering with visualization
hsl_array = np.array(hsl_data)

# Number of clusters
n_clusters = 4
kmeans = KMeans(n_clusters=n_clusters, n_init="auto", random_state=0)
labels = kmeans.fit_predict(hsl_array)
centers = kmeans.cluster_centers_

# Convert HSL centers to RGB and then to HEX
import colorsys
import matplotlib.colors as mcolors

center_hex_colors = []
for h, l, s in centers:
    r, g, b = colorsys.hls_to_rgb(h, l, s)
    center_hex_colors.append(mcolors.to_hex((r, g, b)))


# Create a colormap
cluster_colors = plt.cm.get_cmap("tab10", n_clusters)

# 3D plot
fig = plt.figure(figsize=(10, 7))
ax = fig.add_subplot(111, projection='3d')

for i in range(n_clusters):
    cluster_points = hsl_array[labels == i]
    ax.scatter(
        cluster_points[:, 0],  # Hue
        cluster_points[:, 2],  # Saturation
        cluster_points[:, 1],  # Lightness
        color=cluster_colors(i),
        label=f'Cluster {i}',
        s=80
    )

# Plot the cluster centers using their actual colors
for (h, l, s), hex_color in zip(centers, center_hex_colors):
    ax.scatter(h, s, l, color=hex_color, edgecolor='black', marker='o', s=200, linewidth=1.5)


# Axes and labels
ax.set_xlabel("Hue")
ax.set_ylabel("Saturation")
ax.set_zlabel("Lightness")
ax.set_title("KMeans Clustering in HSL Space")
ax.legend()

plt.tight_layout()
plt.show()


# Step 5: Find underrepresented regions using binning
bins = (5, 5, 5)  # (H, S, L) grid resolution
hist, edges = np.histogramdd(hsl_array, bins=bins, range=[[0, 1], [0, 1], [0, 1]])

print("=== Underrepresented Regions ===")
underrepresented = np.argwhere(hist == 0)
for idx in underrepresented:
    h_bin = f"{edges[0][idx[0]]:.2f}-{edges[0][idx[0]+1]:.2f}"
    s_bin = f"{edges[1][idx[1]]:.2f}-{edges[1][idx[1]+1]:.2f}"
    l_bin = f"{edges[2][idx[2]]:.2f}-{edges[2][idx[2]+1]:.2f}"
    print(f"No colors in bin: Hue={h_bin}, Saturation={s_bin}, Lightness={l_bin}")