Fix bivariate

digautoprofiler/profile_lib.py

@@ -252,6 +252,57 @@ def getTemporalMeta(colData: pd.Series):
         result = "ascending"
     elif colData.is_monotonic_decreasing:
         result = "descending"
+    else:
+        result = "noSort"
+    return {"sortedness": result}
+
+def getAggrData(dfName: pd.DataFrame, catColName: str, quantColName: str, aggrType: str="mean", n=10):
+    """ 
+    aggrType is one of ["mean", "sum", "count", "min", "max"] 
+    """
+    aggrData = dfName.groupby(catColName).agg({quantColName: aggrType})[quantColName].sort_values(ascending=False)[:n]
+    print(aggrData.to_json())
+
+
+def getTempAggrData(dfName: pd.DataFrame, tempColName: str, quantColName: str, aggrType: str="count"): 
+    """
+    timestep kwarg must be one of ["Y","M","W","D","H","T","S"]
+    """
+    # sturge's rule
+    binNum = min(round(1 + 3.322 * math.log10(len(dfName))),20)
+    offsetAliases = ['Y','M','W','D','H','T','S']
+    i = 0
+    previousLen = -float("inf")
+    currentLen = len(dfName[tempColName].dt.to_period(offsetAliases[i]).unique())
+    while i < len(offsetAliases) and currentLen < binNum and currentLen > previousLen:
+        previousLen = currentLen
+        currentLen = len(dfName[tempColName].dt.to_period(offsetAliases[i]).unique())
+        i += 1
+    if currentLen <= previousLen:
+        timestep = offsetAliases[max(i - 2,0)]
+    else:
+        timestep = offsetAliases[max(i - 1,0)]
+
+    indices = dfName[tempColName].dt.to_period(timestep).astype('string')
+    groups = pd.Series(dfName[quantColName].tolist(),index=indices).groupby(level=0)
+    if aggrType == "mean":
+        tempAggrData = groups.mean()
+    if aggrType == "count":
+        tempAggrData = groups.count()
+    if aggrType == "sum":
+        tempAggrData = groups.sum()
+    if aggrType == "min":
+        tempAggrData = groups.min()
+    if aggrType == "max":
+        tempAggrData = groups.max()
+    print(json.dumps({"data":tempAggrData.to_dict(), "timestep":timestep}))
+
+def getTemporalMeta(colData: pd.Series):
+    if colData.is_monotonic_increasing:
+        result = "ascending"
+    elif colData.is_monotonic_decreasing:
+        result = "descending"
+
     else:
         result = "noSort"
     vc = colData.value_counts()
@@ -282,7 +333,7 @@ def getTempAggrData(dfName: pd.DataFrame, tempColName: str, quantColName: str, a
     offsetAliases = ['Y','M','W','D','H','T','S']
     i = 0
     previousLen = -float("inf")
-    currentLen = 0
+    currentLen = len(dfName[tempColName].dt.to_period(offsetAliases[i]).unique())
     while i < len(offsetAliases) and currentLen < binNum and currentLen > previousLen:
         previousLen = currentLen
         currentLen = len(dfName[tempColName].dt.to_period(offsetAliases[i]).unique())

package.json

@@ -141,4 +141,4 @@
       ]
     }
   }
-}
+}

src/components/create-chart/DropdownMenu.svelte

@@ -15,7 +15,7 @@
     export let clickable: boolean = true;
     export let title = 'Select a column';
     export let selectedColumnName: string = undefined;
-
+    
     let displayGroups: string[];
 
     function determineGroup(column: IColTypeTuple) {