moved plots to utils

utils/plot-startup.py

@@ -0,0 +1,31 @@
+import matplotlib.pyplot as plt
+import seaborn as sns
+import random
+import pandas as pd
+
+# Genera i dati mock
+data = [{"tempo di esecuzione": round(random.gauss(23, 1), 3)} for _ in range(9)]
+df = pd.DataFrame(data)
+
+# Creazione del boxplot con grafico più largo e barra più stretta
+plt.figure(figsize=(10, 6))  # Grafico più largo
+ax = sns.boxplot(y=df["tempo di esecuzione"], width=0.3, flierprops={"marker": "o", "color": "red", "markersize": 8}, color="lightgreen")
+
+# Calcolo della media
+mean_value = df["tempo di esecuzione"].mean()
+
+# Aggiungi il testo della media alla base
+plt.text(0, 0, f'Media: {mean_value:.2f}', ha='center', va='bottom', fontsize=14, fontweight='bold', color="black")
+
+# Imposta i label con font più grande
+plt.ylabel("Downtime (s)", fontsize=14)
+plt.xticks([])  # Rimuove i tick sull'asse X
+plt.yticks(fontsize=12)
+
+# Mostra griglia leggera
+plt.grid(axis='y', linestyle='--', alpha=0.6)
+
+# Salva il plot
+plt.tight_layout()
+plt.savefig("benchmark/boxplot_failover.png")
+plt.show()

utils/plot_compared_metrics.py

@@ -0,0 +1,98 @@
+import seaborn as sns
+import matplotlib.pyplot as plt
+import yaml
+import pandas as pd
+import numpy as np
+import os
+
+# Function to read and parse a metric file
+def read_metrics(file_path, label):
+    with open(file_path, 'r') as file:
+        data = yaml.safe_load(file)
+    runs = data['runs']
+    
+    extracted_data = []
+    for run in runs:
+        if 'build' in run:
+            extracted_data.append({'Task': run['n_task'], 'Type': 'Build Time', 'Time': float(run['build']['components_build_time']), 'Source': label})
+
+        if 'code_gen' in run:
+            extracted_data.append({'Task': run['n_task'], 'Type': 'Generation Time', 'Time': float(run['code_gen']['gen_time']), 'Source': label})
+
+        if 'deploy' in run:
+            extracted_data.append({'Task': run['n_task'], 'Type': 'Deployment Time', 'Time': float(run['deploy']['components_deploy_time']), 'Source': label})
+
+        if 'time_total' in run:
+            extracted_data.append({'Task': run['n_task'], 'Type': 'Total Time', 'Time': float(run['time_total']), 'Source': label})
+
+    return extracted_data
+
+# Paths for the two metric files
+file_path_1 = 'res/metrics/metrics-parallel-nats.yaml'  # First metrics file
+file_path_2 = 'res/metrics/metrics-sequential.yaml'  # Second metrics file
+
+# Read and combine the data
+data1 = read_metrics(file_path_1, 'Esecuzione Parallelizzata')
+data2 = read_metrics(file_path_2, 'Esecuzione Sequenziale')
+
+df = pd.DataFrame(data1 + data2)
+
+# Ensure benchmark directory exists
+os.makedirs('benchmark', exist_ok=True)
+
+# Function to plot boxplot
+def plot_boxplot(metric, filename):
+    subset = df[df['Type'] == metric]
+    plt.figure(figsize=(10, 6))
+    sns.boxplot(x='Task', y='Time', hue='Source', data=subset, showfliers=False)
+
+    plt.ylabel('Time (seconds)')
+    plt.grid(True, linestyle='--', alpha=0.7)
+    plt.legend(title='Source')
+    plt.tight_layout()
+    plt.savefig(f'benchmark/{filename}')
+    plt.close()
+
+# Function to plot line plot with confidence intervals
+def plot_lineplot(metric, filename):
+    subset = df[df['Type'] == metric]
+    plt.figure(figsize=(10, 6))
+    sns.lineplot(x='Task', y='Time', hue='Source', data=subset, errorbar=('ci', 95), linewidth=2, marker='o')
+
+    plt.xlabel('Number of Tasks')
+    plt.ylabel('Time (seconds)')
+    plt.grid(True, linestyle='--', alpha=0.7)
+    plt.legend(title='Source')
+    plt.tight_layout()
+    plt.savefig(f'benchmark/{filename}')
+    plt.close()
+
+# Function to plot bar plot with confidence intervals
+def plot_barplot(metric, filename):
+    subset = df[df['Type'] == metric]
+    plt.figure(figsize=(10, 6))
+    
+    sns.barplot(x='Task', y='Time', hue='Source', data=subset, errorbar=('ci', 95), palette=['salmon', 'skyblue'])
+
+    plt.xlabel('Task', fontsize=14)  # Aumenta la dimensione del font dell'asse X
+    plt.ylabel('Time (seconds)', fontsize=14)  # Aumenta la dimensione del font dell'asse Y
+    plt.xticks(fontsize=12)  # Modifica la dimensione del font dei tick dell'asse X
+    plt.yticks(fontsize=12)  # Modifica la dimensione del font dei tick dell'asse Y
+    plt.legend(title='Source', title_fontsize=14, fontsize=12)  # Modifica il font della legenda
+
+    plt.grid(True, linestyle='--', alpha=0.7)
+    plt.tight_layout()
+    plt.savefig(f'benchmark/{filename}')
+    plt.close()
+
+
+# Generate plots for each metric
+metrics = ['Build Time', 'Generation Time', 'Deployment Time', 'Total Time']
+filenames = ['build_time', 'gen_time', 'deploy_time', 'total_time']
+
+for metric, filename in zip(metrics, filenames):
+    #plot_boxplot(metric, f'{filename}_boxplot.png')
+    #plot_lineplot(metric, f'{filename}_lineplot.png')
+    plot_barplot(metric, f'{filename}_paired_barplot.png')
+
+print('Comparison plots saved successfully in "benchmark" directory!')

utils/plot_metrics.py

@@ -0,0 +1,113 @@
+import seaborn as sns
+import matplotlib.pyplot as plt
+import yaml
+import pandas as pd
+import numpy as np
+import os
+
+# Function to read and parse the file
+def read_metrics(file_path):
+    with open(file_path, 'r') as file:
+        data = yaml.safe_load(file)
+    return data['runs']
+
+# Read metrics from the file
+file_path = 'res/metrics-parallel.yaml'  # Replace with your file path
+runs = read_metrics(file_path)
+
+# Flatten data into a list of dictionaries
+data = []
+for run in runs:
+    if 'build' in run:
+        data.append({'Task': run['n_task'], 'Type': 'Build Time', 'Time': float(run['build']['components_build_time'])})
+
+    if 'code_gen' in run:
+        data.append({'Task': run['n_task'], 'Type': 'Generation Time', 'Time': float(run['code_gen']['gen_time'])})
+
+    if 'deploy' in run:
+        data.append({'Task': run['n_task'], 'Type': 'Deployment Time', 'Time': float(run['deploy']['components_deploy_time'])})
+    
+    if 'time_total' in run:
+        data.append({'Task': run['n_task'], 'Type': 'Total Time', 'Time': float(run['time_total'])})
+
+# Convert to DataFrame
+df = pd.DataFrame(data)
+
+# Make sure the benchmark directory exists
+os.makedirs('benchmark', exist_ok=True)
+
+# Function to plot boxplot
+def plot_boxplot(metric, filename, color):
+    subset = df[df['Type'] == metric]
+    plt.figure(figsize=(10, 6))
+    sns.boxplot(x='Task', y='Time', data=subset, color=color, showfliers=False)
+
+    # Label median values
+    medians = subset.groupby('Task')['Time'].median()
+    for i, task in enumerate(medians.index):
+        median_value = medians[task]
+        plt.text(i, median_value, f'{median_value:.3f}', ha='center', va='center', fontsize=10, color='white',
+                 bbox=dict(facecolor='black', alpha=0.6, boxstyle='round,pad=0.3'))
+
+    plt.ylabel('Time (seconds)')
+    plt.grid(True, linestyle='--', alpha=0.7)
+    plt.tight_layout()
+    plt.savefig(f'benchmark/{filename}')
+    plt.close()
+
+# Function to plot line plot with confidence intervals
+def plot_lineplot(metric, filename, color):
+    subset = df[df['Type'] == metric]
+    plt.figure(figsize=(10, 6))
+    sns.lineplot(x='Task', y='Time', data=subset, errorbar=('ci', 95), color=color, linewidth=2, marker='o')
+
+    min_task = subset['Task'].min()
+    max_task = subset['Task'].max()
+    all_tasks = np.arange(min_task, max_task + 1)
+    plt.xticks(all_tasks)
+
+    for i, row in subset.groupby('Task')['Time'].median().reset_index().iterrows():
+        plt.text(row['Task'], row['Time'], f'{row["Time"]:.3f}', ha='center', va='bottom', fontsize=10)
+
+    plt.xlabel('Number of Tasks')
+    plt.ylabel('Time (seconds)')
+    plt.grid(True, linestyle='--', alpha=0.7)
+    plt.tight_layout()
+    plt.savefig(f'benchmark/{filename}')
+    plt.close()
+
+# Function to plot bar plot with confidence intervals
+def plot_barplot(metric, filename, color):
+    subset = df[df['Type'] == metric]
+    plt.figure(figsize=(10, 6))
+    sns.barplot(x='Task', y='Time', data=subset, color=color, errorbar=('ci', 95))
+
+    means = subset.groupby('Task')['Time'].mean()
+    for i, task in enumerate(means.index):
+        mean_value = means[task]
+        plt.text(i, 0, f'{mean_value:.3f}', ha='center', va='bottom', fontsize=10, alpha=0.7)
+
+    plt.ylabel('Time (seconds)')
+    plt.grid(True, linestyle='--', alpha=0.7)
+    plt.tight_layout()
+    plt.savefig(f'benchmark/{filename}')
+    plt.close()
+
+# Plot each metric separately
+#plot_boxplot('Build Time', 'build_time_boxplot.png', 'skyblue')
+#plot_lineplot('Build Time', 'build_time_lineplot.png', 'skyblue')
+plot_barplot('Build Time', 'build_time_barplot.png', 'skyblue')
+
+#plot_boxplot('Generation Time', 'gen_time_boxplot.png', 'lightgreen')
+#plot_lineplot('Generation Time', 'gen_time_lineplot.png', 'lightgreen')
+plot_barplot('Generation Time', 'gen_time_barplot.png', 'lightgreen')
+
+#plot_boxplot('Deployment Time', 'deploy_time_boxplot.png', 'salmon')
+#plot_lineplot('Deployment Time', 'deploy_time_lineplot.png', 'salmon')
+plot_barplot('Deployment Time', 'deploy_time_barplot.png', 'salmon')
+
+#plot_boxplot('Total Time', 'total_time_boxplot.png', 'purple')
+#plot_lineplot('Total Time', 'total_time_lineplot.png', 'purple')
+plot_barplot('Total Time', 'total_time_barplot.png', 'orange')
+
+print('Plots saved successfully in "benchmark" directory!')