inference_times_plots.py: code to plot mean time for inference of mod…

…els from tests/results

static/inference_times_plots.py

@@ -5,17 +5,35 @@
 
 if __name__ == "__main__":
     with open('../test/tests/results/conversion_times.json', 'r', encoding='utf-8') as fp:
-        data = json.load(fp)
+        data_conv = json.load(fp)
+
+    with open('../test/tests/results/inference_times.json', 'r', encoding='utf-8') as fp:
+        data_plan = json.load(fp)
+
+    inference_times_conv = [{'model': k, 'time': v['mean']} for k, v in data_conv.items()]
+    inference_times_plan = [{'model': k, 'time': v['mean']} for k, v in data_conv.items()]
 
-    inference_times_conv = [{'model': k, 'time': v['mean']} for k, v in data.items()]
     df_conv = pd.DataFrame(inference_times_conv)
+    df_plan = pd.DataFrame(inference_times_plan)
 
-    plt.figure(figsize=(12, 12))
+    fig, axes = plt.subplots(nrows=1, ncols=2, figsize=(16, 16))
+    fig.suptitle('Inference Times')
     sns.set(style="whitegrid")
-    sns.barplot(x='model', y='time', data=df_conv)
-    plt.xlabel('Model')
-    plt.ylabel('Mean Time')
-    plt.title('Mean Conversion Times per Model')
-    plt.xticks(rotation=45)
 
+    ax_plan = axes[0, 0]
+    sns.barplot(x='model', y='time', data=df_plan, ax=ax_plan)
+    ax_plan.set_title('Planning Times')
+    ax_plan.set_xlabel('Model')
+    ax_plan.set_ylabel('Time')
+    ax_plan.xticks(rotation=45)
+
+    ax_conv = axes[0, 1]
+    sns.barplot(x='model', y='time', data=df_conv, ax=ax_conv)
+    ax_plan.set_title('Conversion Times')
+    ax_plan.set_xlabel('Model')
+    ax_plan.set_ylabel('Time')
+    ax_plan.xticks(rotation=45)
+
+    plt.tight_layout()
+    plt.subplots_adjust(top=0.92)
     plt.show()