feat: plot PO5 MSD

flake.nix

@@ -53,6 +53,8 @@
                 numpy
                 pillow
                 matplotlib
+                seaborn
+                pandas
               ]
             ))
             # pipenv

gridding.def

@@ -10,7 +10,7 @@ Stage: build
   dotnet publish src/CLI -o dist
 
 Bootstrap: docker
-From: mcr.microsoft.com/dotnet/aspnet:8.0
+From: mcr.microsoft.com/dotnet/aspnet:9.0
 Stage: runtime
 
 %files from build

tools/get_speedup.py

@@ -0,0 +1,27 @@
+import sys
+
+import numpy as np
+
+if __name__ == "__main__":
+    if len(sys.argv) != 3:
+        print(
+            "Usage: python3 get_speedup.py <orig_timings.data> <comp_timings.data>",
+            file=sys.stderr,
+        )
+        sys.exit(1)
+
+    orig_timings_path = sys.argv[1]
+    comp_timings_path = sys.argv[2]
+
+    with open(orig_timings_path, "r") as f:
+        orig_timings = [float(line.strip()) for line in f if line.strip()]
+    with open(comp_timings_path, "r") as f:
+        comp_timings = [float(line.strip()) for line in f if line.strip()]
+
+    orig_mean = np.mean(orig_timings)
+    comp_mean = np.mean(comp_timings)
+
+    speedup = orig_mean / comp_mean
+
+    print(f"original latency: ~{orig_mean}ms, compressed latency: ~{comp_mean}")
+    print(f"speedup original/compressed: {speedup}")

tools/plot_MSD_comp.py

@@ -1,4 +1,3 @@
-import argparse
 import glob
 import os
 import re

tools/plot_PO_MSD.py

@@ -0,0 +1,81 @@
+import os
+import sys
+
+import matplotlib.pyplot as plt
+import pandas as pd
+import seaborn as sns
+
+# Define multiple experiments
+experiments = [
+    {  # Experiment 1
+        "strategies": [
+            {"prefix": "k-38-spd", "label": "Speed"},
+            {"prefix": "k-38-tmp", "label": "Temperature"},
+            {"prefix": "k-38-sal", "label": "Salinity"},
+        ],
+    },
+]
+
+# Plot setup
+plt.figure(figsize=(7, 5))
+
+
+def read_MSD(file_path):
+    with open(file_path) as f:
+        scores = [float(line.strip()) for line in f if line.strip()]
+    return scores
+
+
+def plot_strategies(strategies, res_dir):
+    data = []
+    for strat in strategies:
+        prefix = strat["prefix"]
+        label = strat["label"]
+
+        msd_file = os.path.join(res_dir, f"PO5-{prefix}-MSD.data")
+        msd = read_MSD(msd_file)
+
+        data.extend({"Visualized Variable": label, "Similarity Score": s} for s in msd)
+
+    df = pd.DataFrame(data)
+    sns.set_style("whitegrid")
+    sns.swarmplot(x="Visualized Variable", y="Similarity Score", data=df)
+    # sns.boxplot(
+    #     x="Visualized Variable",
+    #     y="Similarity Score",
+    #     data=df,
+    #     showcaps=False,
+    #     boxprops={"facecolor": "None"},
+    #     showfliers=False,
+    #     whiskerprops={"linewidth": 0},
+    # )
+
+
+if __name__ == "__main__":
+    if len(sys.argv) != 2:
+        print(
+            "Usage: python3 plot_PO_MSD.py <experiment>",
+            file=sys.stderr,
+        )
+        sys.exit(1)
+
+    exp_num = int(sys.argv[1])
+    if exp_num < 1 or exp_num > len(experiments):
+        print("No such experiment!", file=sys.stderr)
+        sys.exit(1)
+
+    res_dir = f"../data"
+
+    # Experiments are 1-indexed, while list are not...
+    exp = experiments[exp_num - 1]
+    strategies = exp["strategies"]
+
+    # Add scatter plots for strategies
+    plot_strategies(strategies, res_dir)
+
+    # Show plot
+    plt.title(f"Compression vs Similarity")
+    # plt.grid(axis="y", linestyle="-", alpha=0.7)
+    # plt.tight_layout()
+    plt.savefig(f"exp-{exp_num}-swarm-MSD.svg", format="svg")
+    plt.show()