Updated Interestingness Tests (#133)

* add black

* update cars dataset and tests

* Delete old dataset

* Updated Interestingness Tests

Updated tests to use the newly updated cars dataset

* switch to local cars reference

Co-authored-by: Kunal Agarwal <kagarwal2@berkeley.edu>
Co-authored-by: Kunal Agarwal <32151899+westernguy2@users.noreply.github.com>
Co-authored-by: 19thyneb <thyne.boonmark@gmail.com>
Co-authored-by: Doris Lee <dorisjunglinlee@gmail.com>

tests/test_compiler.py

@@ -155,8 +155,7 @@ def test_sort_bar():
 
 
 def test_specified_vis_collection():
-    url = "https://github.com/lux-org/lux-datasets/blob/master/data/cars.csv?raw=true"
-    df = pd.read_csv(url)
+    df = pd.read_csv("lux/data/car.csv")
     df["Year"] = pd.to_datetime(
         df["Year"], format="%Y"
     )  # change pandas dtype for the column "Year" to datetype

tests/test_dates.py

@@ -93,7 +93,7 @@ def test_period_to_altair():
 
     exported_code = df.recommendation["Filter"][2].to_Altair()
 
-    assert "Year = 1971" in exported_code
+    assert "Year = 1972" in exported_code
 
 
 def test_refresh_inplace():

tests/test_interestingness.py

@@ -51,9 +51,9 @@ def test_interestingness_1_0_0():
             if int(vis._inferred_intent[2].value) == 8:
                 rank1 = f
             if int(vis._inferred_intent[2].value) == 6:
-                rank2 = f
-        if "1972" in str(df.recommendation["Filter"][f]._inferred_intent[2].value):
-            rank3 = f
+                rank3 = f
+        if "ford" in str(df.recommendation["Filter"][f]._inferred_intent[2].value):
+            rank2 = f
     assert rank1 < rank2 and rank1 < rank3 and rank2 < rank3
 
 
@@ -112,7 +112,7 @@ def test_interestingness_0_1_0():
             rank1 = f
         if str(df.recommendation["Filter"][f]._inferred_intent[2].value) == "Europe":
             rank2 = f
-        if "1971" in str(df.recommendation["Filter"][f]._inferred_intent[2].value):
+        if "1970" in str(df.recommendation["Filter"][f]._inferred_intent[2].value):
             rank3 = f
     assert rank1 < rank2 and rank1 < rank3 and rank2 < rank3
 
@@ -151,11 +151,11 @@ def test_interestingness_1_1_0():
         if len(vis.get_attr_by_attr_name("Cylinders")) > 0:
             if int(vis._inferred_intent[2].value) == 6:
                 rank1 = f
-            if int(vis._inferred_intent[2].value) == 5:
-                rank3 = f
+            if int(vis._inferred_intent[2].value) == 8:
+                rank2 = f
         if len(vis.get_attr_by_attr_name("Origin")) > 0:
             if str(vis._inferred_intent[2].value) == "Europe":
-                rank2 = f
+                rank3 = f
     assert rank1 < rank2 and rank1 < rank3 and rank2 < rank3
 
     # check that top recommended generalize graph score is not none
@@ -261,9 +261,9 @@ def test_interestingness_0_2_0():
     for f in range(0, len(df.recommendation["Filter"])):
         if "1973" in str(df.recommendation["Filter"][f]._inferred_intent[2].value):
             rank1 = f
-        if "1976" in str(df.recommendation["Filter"][f]._inferred_intent[2].value):
+        if "ford" in str(df.recommendation["Filter"][f]._inferred_intent[2].value):
             rank2 = f
-        if str(df.recommendation["Filter"][f]._inferred_intent[2].value) == "Europe":
+        if str(df.recommendation["Filter"][f]._inferred_intent[2].value) == "USA":
             rank3 = f
     assert rank1 < rank2 and rank1 < rank3 and rank2 < rank3
 

tests/test_maintainence.py

@@ -77,12 +77,12 @@ def test_recs_inplace_operation():
     df = pd.read_csv("lux/data/car.csv")
     df._repr_html_()
     assert df._recs_fresh == True, "Failed to maintain recommendation after display df"
-    assert len(df.recommendation["Occurrence"]) == 3
+    assert len(df.recommendation["Occurrence"]) == 4
     df.drop(columns=["Name"], inplace=True)
     assert "Name" not in df.columns, "Failed to perform `drop` operation in-place"
     assert (
         df._recs_fresh == False
     ), "Failed to maintain recommendation after in-place Pandas operation"
     df._repr_html_()
-    assert len(df.recommendation["Occurrence"]) == 2
+    assert len(df.recommendation["Occurrence"]) == 3
     assert df._recs_fresh == True, "Failed to maintain recommendation after display df"

tests/test_pandas_coverage.py

@@ -111,6 +111,7 @@ def test_rename3():
         "col7",
         "col8",
         "col9",
+        "col10",
     ]
     df._repr_html_()
     assert list(df.recommendation.keys()) == [
@@ -119,7 +120,7 @@ def test_rename3():
         "Occurrence",
         "Temporal",
     ]
-    assert len(df.cardinality) == 9
+    assert len(df.cardinality) == 10
     assert "col2" in list(df.cardinality.keys())
 
 
@@ -194,7 +195,7 @@ def test_query():
         "Occurrence",
         "Temporal",
     ]
-    assert len(new_df.cardinality) == 9
+    assert len(new_df.cardinality) == 10
 
 
 def test_pop():
@@ -208,7 +209,7 @@ def test_pop():
         "Occurrence",
         "Temporal",
     ]
-    assert len(df.cardinality) == 8
+    assert len(df.cardinality) == 9
 
 
 def test_transform():
@@ -217,7 +218,7 @@ def test_transform():
     new_df = df.iloc[:, 1:].groupby("Origin").transform(sum)
     new_df._repr_html_()
     assert list(new_df.recommendation.keys()) == ["Correlation", "Occurrence"]
-    assert len(new_df.cardinality) == 6
+    assert len(new_df.cardinality) == 7
 
 
 def test_get_group():
@@ -232,7 +233,7 @@ def test_get_group():
         "Occurrence",
         "Temporal",
     ]
-    assert len(new_df.cardinality) == 9
+    assert len(new_df.cardinality) == 10
 
 
 def test_applymap():
@@ -247,13 +248,11 @@ def test_applymap():
         "Occurrence",
         "Temporal",
     ]
-    assert len(df.cardinality) == 9
+    assert len(df.cardinality) == 10
 
 
 def test_strcat():
-    df = pd.read_csv(
-        "https://github.com/lux-org/lux-datasets/blob/master/data/cars.csv?raw=true"
-    )
+    df = pd.read_csv("lux/data/car.csv")
     df["Year"] = pd.to_datetime(df["Year"], format="%Y")
     df["combined"] = df["Origin"].str.cat(df["Brand"], sep=", ")
     df._repr_html_()
@@ -267,9 +266,7 @@ def test_strcat():
 
 
 def test_named_agg():
-    df = pd.read_csv(
-        "https://github.com/lux-org/lux-datasets/blob/master/data/cars.csv?raw=true"
-    )
+    df = pd.read_csv("lux/data/car.csv")
     df["Year"] = pd.to_datetime(df["Year"], format="%Y")
     new_df = df.groupby("Brand").agg(
         avg_weight=("Weight", "mean"),
@@ -292,7 +289,7 @@ def test_change_dtype():
         "Occurrence",
         "Temporal",
     ]
-    assert len(df.data_type_lookup) == 9
+    assert len(df.data_type_lookup) == 10
 
 
 def test_get_dummies():
@@ -306,7 +303,7 @@ def test_get_dummies():
         "Occurrence",
         "Temporal",
     ]
-    assert len(new_df.data_type_lookup) == 310
+    assert len(new_df.data_type_lookup) == 339
 
 
 def test_drop():
@@ -321,7 +318,7 @@ def test_drop():
         "Occurrence",
         "Temporal",
     ]
-    assert len(new_df2.cardinality) == 6
+    assert len(new_df2.cardinality) == 7
 
 
 def test_merge():
@@ -336,7 +333,7 @@ def test_merge():
         "Occurrence",
         "Temporal",
     ]  # TODO once bug is fixed
-    assert len(new_df2.cardinality) == 10
+    assert len(new_df2.cardinality) == 11
 
 
 def test_prefix():
@@ -350,14 +347,12 @@ def test_prefix():
         "Occurrence",
         "Temporal",
     ]
-    assert len(new_df.cardinality) == 9
+    assert len(new_df.cardinality) == 10
     assert new_df.cardinality["1_Name"] == 300
 
 
 def test_loc():
-    df = pd.read_csv(
-        "https://github.com/lux-org/lux-datasets/blob/master/data/cars.csv?raw=true"
-    )
+    df = pd.read_csv("lux/data/car.csv")
     df["Year"] = pd.to_datetime(df["Year"], format="%Y")
     new_df = df.loc[:, "Displacement":"Origin"]
     new_df._repr_html_()
@@ -388,9 +383,7 @@ def test_loc():
 
 
 def test_iloc():
-    df = pd.read_csv(
-        "https://github.com/lux-org/lux-datasets/blob/master/data/cars.csv?raw=true"
-    )
+    df = pd.read_csv("lux/data/car.csv")
     df["Year"] = pd.to_datetime(df["Year"], format="%Y")
     new_df = df.iloc[:, 3:9]
     new_df._repr_html_()
@@ -563,8 +556,7 @@ def test_value_counts():
 
 
 def test_str_replace():
-    url = "https://github.com/lux-org/lux-datasets/blob/master/data/cars.csv?raw=true"
-    df = pd.read_csv(url)
+    df = pd.read_csv("lux/data/car.csv")
     df._repr_html_()  # compute metadata
     assert df.cardinality is not None
     series = df["Brand"].str.replace("chevrolet", "chevy")