plots: scatter with mean line

src/database/dbAnalyser.py

@@ -9,37 +9,64 @@ import json
 import pandas as pd
 import matplotlib.pyplot as plt
 
+colorLookup = {
+    "google": "r",
+    "leaflet": "b",
+    "mapbox_feature": "g",
+    "mapbox_marker": "y",
+    "mapgl": "m",
+    "pigeons": "c"
+}
 
 def readFile():
     with open("time.json") as file:
         timeList = json.load(file)
-    
+
     timeDf = pd.DataFrame(timeList)
+    print("number of tests:", timeDf.shape[0])
     timeDf["markers"] = pd.to_numeric(timeDf["markers"], errors="raise")
+    print(timeDf.value_counts(subset=["id"]))
+    print(timeDf.value_counts(subset=["id", "type"]))
     return timeDf
 
-def avgTimeMaps(df):
-    avgDf = df.groupby(["id", "markers"]).mean().reset_index()
-    avgDf = avgDf.pivot(index="markers", columns="id", values="time")
-    avgDf.plot()
-    plt.savefig("plots/mapsAvg.png")
+def timeMaps(df, plotTitle="plots/maps.png"):
+    fig, ax = plt.subplots(figsize=(20,12))
+    for mapId, mapDf in df.groupby(["id"]):
+        if mapId != "mapbox":
+            ax.scatter(mapDf["markers"], mapDf["time"], color=colorLookup.get(mapId))
+            mapAvgDf = mapDf.groupby(["markers"]).mean().reset_index()
+            ax.plot(mapAvgDf["markers"], mapAvgDf["time"], label=mapId, color=colorLookup.get(mapId))
+        else:
+            for markerType, typeDf in mapDf.groupby(["type"]):
+                mapTypeId = "mapbox_" + markerType
+                ax.scatter(typeDf["markers"], typeDf["time"], color=colorLookup.get(mapTypeId))
+                typeMapAvgDf = typeDf.groupby(["markers"]).mean().reset_index()
+                ax.plot(typeMapAvgDf["markers"], typeMapAvgDf["time"], label=mapTypeId, color=colorLookup.get(mapTypeId))
+
 
-def timeMaps(df):
-    fig, ax = plt.subplots(figsize=(10,4))
-    for key, grp in df.groupby(["id"]):
-        ax.scatter(grp["markers"], grp["time"], label=key)#
     ax.legend(title="Maps")
-    plt.savefig("plots/maps.png")
+    plt.savefig(plotTitle)
 
-def mapboxTypes(df):
+def mapboxTypes(df, plotTitle="plots/mapbox.png"):
     mapboxDf = df.dropna(subset=["type"])
-    fig, ax = plt.subplots(figsize=(10,4))
-    for key, grp in mapboxDf.groupby(["type"]):
-        ax.scatter(grp["markers"], grp["time"], label=key)#
+    fig, ax = plt.subplots(figsize=(20,12))
+    for markerType, typeDf in mapboxDf.groupby(["type"]):
+        mapId = "mapbox_" + markerType
+        ax.scatter(typeDf["markers"], typeDf["time"], label=mapId, color=colorLookup.get(mapId))
+
+        typeAvgDf = typeDf.groupby(["markers"]).mean().reset_index()
+        ax.plot(typeAvgDf["markers"], typeAvgDf["time"], color=colorLookup.get(mapId))
+
     ax.legend(title="Marker Type")
-    plt.savefig("plots/mapbox.png")
+    plt.savefig(plotTitle)
+
+def smallValues(df):
+    df = df.loc[df["markers"] < 300]
+    timeMaps(df, "plots/mapsSmall.png")
+    mapboxTypes(df, "plots/mapboxSmall.png")
+
 
 df = readFile()
-avgTimeMaps(df)
 timeMaps(df)
-mapboxTypes(df)
+mapboxTypes(df)
+smallValues(df)