class DataFrameServiceTest(unittest.TestCase):
def setUp(self):
self.sut = DataFrameService()
def test_getWordCorrelations(self):
df = self.sut.getWordCorrelations("modi")
self.assertTrue(len(df.columns) > 1)
def test_getWordDates(self):
df = self.sut.getWordDates("coronavirus")
self.assertTrue(len(df.columns) , 1)
def test_getWordDatesMultiIndex(self):
df = self.sut.getWordDates("coronavirus")
idx = pd.IndexSlice
def _update_checklist(self, input_value):
word = input_value.lower()
df = DataFrameService().getWordDates(word)
options = [{"label": c, 'value': c} for c in df["channel_id"].unique()]
options.append({
"label": WordTrend.HTML_IDS.CHECKLIST_ALL,
'value': WordTrend.HTML_IDS.CHECKLIST_ALL
})
return options
def _filteredDf(self, checklist, word):
df = DataFrameService().getWordDates(word)
if df is None:
return None
if WordTrend.HTML_IDS.CHECKLIST_ALL in checklist:
# if "All" selected
checklist = df["channel_id"].unique()
df = df[df["channel_id"].isin(checklist)]
df["rolling"] = df.groupby([
"channel_id"
])["mean_prop"].transform(lambda x: x.rolling(14, 1).mean())
#df["rolling"] = df["mean_prop"]
df["date"] = pd.to_datetime(df["date"], format="%m_%d_%y")
return df
def _filteredDf(self, checklist, selected_words):
dfs = []
for word in selected_words:
df = DataFrameService().getWordDates(word)
df["word"] = word
dfs.append(df)
if len(dfs) == 0:
return None
df = pd.concat(dfs)
if WordProportion.HTML_IDS.CHECKLIST_ALL in checklist:
# if "All" selected
checklist = df["channel_id"].unique()
df = df[df["channel_id"].isin(checklist)]
df = df.groupby(["channel_id",
"word"]).agg(overall_prop=("mean_prop",
"mean")).reset_index()
return df
def _filteredDf(self, checklist, word):
# dfs = []
# for word in selected_words:
#
# df["word"] = word
# dfs.append(df)
#
# df = pd.concat(dfs)
df = DataFrameService().getWordCorrelations(word)
if WordProportion.HTML_IDS.CHECKLIST_ALL in checklist:
# if "All" selected
checklist = df["channel_id"].unique()
df = df[df["channel_id"].isin(checklist)]
df["abs_corr"] = np.abs(df["corr"])
df = df.groupby(["channel_id"]).apply(lambda x: x.sort_values(
["abs_corr"], ascending=False).head(10)).reset_index(drop=True)
df = df[df["index"] != word]
#print(df)
return df
def getDataFrameMethod(self):
return DataFrameService().getWordDates
def setUp(self):
self.sut = DataFrameService()
def setupOptions(self):
df = DataFrameService().getWordDates(self.default_word)
if df.shape[0] > 0:
self.colors = ColorPalette.mapRandomColors(df["channel_id"],
THEME.COLOR)