Feat: filter none numeric scores

recopytex/datalib/on_score_column.py

@@ -10,7 +10,7 @@ def score_to_mark(x, score_max, rounding=lambda x: round(x, 2)):
     if the item is leveled then the score is multiply by the score_rate
     otherwise it copies the score
 
-    :param x: dictionnary with "is_leveled", "score" and "score_rate" keys
+    :param x: dictionnary with "is_leveled", "score"  (need to be number) and "score_rate" keys
     :param score_max:
     :param rounding: rounding mark function
     :return: the mark
@@ -125,6 +125,10 @@ def score_to_level(x, level_max=3):
     return int(ceil(x["score"] / x["score_rate"] * level_max))
 
 
+def score_to_numeric_score(x, score_config):
+    pass
+
+
 # -----------------------------
 # Reglages pour 'vim'
 # vim:set autoindent expandtab tabstop=4 shiftwidth=4:

recopytex/datalib/on_score_dataframe.py

@@ -2,6 +2,7 @@
 # encoding: utf-8
 
 from .on_score_column import score_to_mark, score_to_level
+import pandas as pd
 
 
 def compute_marks(df, score_max, rounding=lambda x: round(x, 2)):
@@ -9,7 +10,7 @@ def compute_marks(df, score_max, rounding=lambda x: round(x, 2)):
 
     apply score_to_mark to each row
 
-    :param df: DataFrame with "score", "is_leveled" and "score_rate" columns.
+    :param df: DataFrame with "score" (need to be number), "is_leveled" and "score_rate" columns.
 
     >>> import pandas as pd
     >>> d = {"Eleve":["E1"]*6 + ["E2"]*6,
@@ -113,12 +114,12 @@ def compute_normalized(df, rounding=lambda x: round(x, 2)):
     >>> d = {"Eleve":["E1"]*6 + ["E2"]*6,
     ...    "score_rate":[1]*2+[2]*2+[2]*2 + [1]*2+[2]*2+[2]*2,
     ...    "is_leveled":[0]*4+[1]*2 + [0]*4+[1]*2,
-    ...    "score":[1, 0.33, 2, 1.5, 1, 3,   0.666, 1, 1.5, 1, 2, 3],
+    ...    "score":[0, 0.33, 2, 1.5, 1, 3,   0.666, 1, 1.5, 1, 2, 3],
     ...    }
     >>> df = pd.DataFrame(d)
     >>> df["mark"] = compute_marks(df, 3)
     >>> compute_normalized(df)
-    0     1.00
+    0     0.00
     1     0.33
     2     1.00
     3     0.75
@@ -135,6 +136,60 @@ def compute_normalized(df, rounding=lambda x: round(x, 2)):
     return rounding(df["mark"] / df["score_rate"])
 
 
+def filter_none_score(df, score_config):
+    """Filter rows where scores have None numeric values
+
+    :example:
+
+    >>> import pandas as pd
+    >>> d = {"Eleve":["E1"]*7,
+    ...    "score_rate": [1]*7,
+    ...    "is_leveled":[0]+[1]*6,
+    ...    "score":[0.33, "", ".", "a", 1, 2, 3],
+    ...    }
+    >>> score_config = {
+    ...   'BAD': {'value': 0, 'numeric_value': 0},
+    ...   'FEW': {'value': 1, 'numeric_value': 1},
+    ...   'NEARLY': {'value': 2, 'numeric_value': 2},
+    ...   'GOOD': {'value': 3, 'numeric_value': 3},
+    ...   'NOTFILLED': {'value': '', 'numeric_value': 'None'},
+    ...   'NOANSWER': {'value': '.', 'numeric_value': 0},
+    ...   'ABS': {'value': 'a', 'numeric_value': 'None'}
+    ...    }
+    >>> df = pd.DataFrame(d)
+    >>> filter_none_score(df, score_config)
+      Eleve  score_rate  is_leveled score
+    0    E1           1           0  0.33
+    2    E1           1           1     .
+    4    E1           1           1     1
+    5    E1           1           1     2
+    6    E1           1           1     3
+    """
+    not_leveled_df = df[df["is_leveled"] != 1]
+    leveled_df = df[df["is_leveled"] == 1]
+
+    not_none_values = [
+        v["value"]
+        for v in score_config.values()
+        if str(v["numeric_value"]).lower() != "none"
+    ]
+    filtered_leveled_df = leveled_df[leveled_df["score"].isin(not_none_values)]
+
+    return pd.concat([not_leveled_df, filtered_leveled_df])
+
+
+def score_to_numeric_score(df, score_config):
+    """Transform a score to the corresponding numeric value
+
+    >>> d = {"Eleve":["E1"]*6 + ["E2"]*6,
+    ...    "score_rate":[1]*2+[2]*2+[2]*2 + [1]*2+[2]*2+[2]*2,
+    ...    "is_leveled":[0]*4+[1]*2 + [0]*4+[1]*2,
+    ...    "score":[0, 0.33, 2, 1.5, 1, 3,   0.666, 1, 1.5, 1, 2, 3],
+    ...    }
+    """
+    pass
+
+
 # -----------------------------
 # Reglages pour 'vim'
 # vim:set autoindent expandtab tabstop=4 shiftwidth=4: