Skip to content

Add plotting functionality of computing costs. #53

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
myxie wants to merge 1 commit into
base: main
Choose a base branch
from
Draft

Add plotting functionality of computing costs. #53

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
111 changes: 111 additions & 0 deletions skaworkflows/data/utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,111 @@
"""
Utility methods to evaluate SKAWorkflows data
"""

import numpy as np
import matplotlib.pyplot as plt
import matplotlib.patches as patches
import pandas as pd

from scipy.spatial import ConvexHull
from matplotlib.colors import LogNorm

from skaworkflows.common import SKALow, LOW_TOTAL_SIZING

from matplotlib import rcParams
rcParams["text.usetex"] = False
rcParams["font.family"] = "serif"
rcParams["font.size"] = 9.0


def heatmap_computing_permutations():
"""
Produce a heatmap of the different combinations of stations and
baselines for SKA Low HPSOs
"""
channels = 16384*2 #/2
df = pd.read_csv(LOW_TOTAL_SIZING)
df = df[['HPSO', 'Baseline', 'Stations', 'Total Batch [Pflop/s]', 'Channels']]
matrices = []
fig, axes = plt.subplots(figsize=(10/3, 2.5), dpi=300,nrows=1, ncols=1)

i = 0
for hpso in df['HPSO'].unique():
if hpso == 'hpso04a' or hpso=='hpso05a':
continue

df_hpso = df[(df['HPSO'] == hpso) & (df['Channels'] == channels)]
matrix=df_hpso.pivot_table(index='Baseline', columns='Stations', values='Total Batch [Pflop/s]',
aggfunc='sum')
matrices.append(matrix.to_numpy())

# ratio_df = pd.read_csv("/home/rwb/github/experiments/simulation_scheduling_experiments/hpso_permutation_ratios-4_64-0.5_128-0.3_256-0.2_alpha-0.20_pivot.csv",
# index_col=False)
# ratio_hpso_df = ratio_df[ratio_df['hpso']==hpso].drop(['hpso'], axis=1).replace(np.nan, 0)
# ratio_matrix = ratio_hpso_df.to_numpy()

mean_matrix = np.mean(np.stack(matrices, axis=0), axis=0)
im = axes.imshow(mean_matrix, cmap='viridis', origin='lower')#, norm=LogNorm())

axes.set_xticks(range(len(matrix.columns)), matrix.columns)
axes.set_xticklabels(matrix.columns)
axes.set_yticks(range(len(matrix.index)), matrix.index)
axes.set_yticklabels(matrix.index)
axes.set_title(f"Mean PFLOPs")
# Cells we want to highlight (row, col)

# For each cell, work out its square corners
# imshow aligns cells so (col-0.5, row-0.5) to (col+0.5, row+0.5)
def draw_edges(cells: list):
edges = set()
for r, c in cells:
# each edge is ((x1, y1), (x2, y2))
corners = [
(c - 0.5, r - 0.5), (c + 0.5, r - 0.5),
(c + 0.5, r + 0.5), (c - 0.5, r + 0.5)
]
square_edges = [
(corners[0], corners[1]), # bottom
(corners[1], corners[2]), # right
(corners[2], corners[3]), # top
(corners[3], corners[0]) # left
]
for e in square_edges:
# store edges in a normalized order so shared edges cancel
e_norm = tuple(sorted(e))
if e_norm in edges:
edges.remove(e_norm) # internal edge → remove
else:
edges.add(e_norm) # boundary edge → keep for now

return edges

cells = [(4, 1),(4,2), (3,3), (2,4), (3,2), (2, 3), (1, 4)]
edges = draw_edges(cells)
for (x1, y1), (x2, y2) in edges:
axes.plot([x1, x2], [y1, y2], color="yellow", linewidth=2)

cells = [(4, 4), (4, 3), (3, 4)]
edges = draw_edges(cells)
for (x1, y1), (x2, y2) in edges:
axes.plot([x1, x2], [y1, y2], color="red", linewidth=2)

axes.spines['right'].set_visible(False)
axes.spines['top'].set_visible(False)

axes.set_xlabel("# Stations")
axes.set_ylabel("Baseline (km)")
axes.tick_params(axis='x', labelrotation=45)
plt.subplots_adjust(left=0.3, bottom=0.1, right=0.85, top=1)
# from mpl_toolkits.axes_grid1 import make_axes_locatable
# divider = make_axes_locatable(axes)
# cax = divider.append_axes("right", size="5%", pad=0.05) # same height as ax
fig.colorbar(im, ax=axes, orientation='vertical', pad=0.1, label='PFLOPs', shrink=0.3)

plt.savefig("HeatmapPermutations.png",dpi=fig.dpi)
# print(arr)


if __name__ == '__main__':
heatmap_computing_permutations()