def test_wrong_dimensions():
with pytest.raises(ValueError) as err_msg:
fig = px.scatter(px.data.tips(), x="tip", y=[1, 2, 3])
assert "All arguments should have the same length." in str(
err_msg.value)
# the order matters
with pytest.raises(ValueError) as err_msg:
fig = px.scatter(px.data.tips(), x=[1, 2, 3], y="tip")
assert "All arguments should have the same length." in str(
err_msg.value)
with pytest.raises(ValueError):
fig = px.scatter(px.data.tips(), x=px.data.iris().index, y="tip")
def test_multiindex_raise_error():
index = pd.MultiIndex.from_product([[1, 2, 3], ["a", "b"]],
names=["first", "second"])
df = pd.DataFrame(np.random.random((6, 3)),
index=index,
columns=["A", "B", "C"])
# This is ok
fig = px.scatter(df, x="A", y="B")
with pytest.raises(TypeError) as err_msg:
fig = px.scatter(df, x=df.index, y="B")
assert "pandas MultiIndex is not supported by plotly express" in str(
err_msg.value)
def test_scatter():
iris = px.data.iris()
fig = px.scatter(iris, x="sepal_width", y="sepal_length")
assert fig.data[0].type == "scatter"
assert np.all(fig.data[0].x == iris.sepal_width)
assert np.all(fig.data[0].y == iris.sepal_length)
# test defaults
assert fig.data[0].mode == "markers"
def test_arrayattrable_numpy():
tips = px.data.tips()
fig = px.scatter(tips,
x="total_bill",
y="tip",
hover_data=[np.random.random(tips.shape[0])])
assert (fig.data[0]["hovertemplate"] ==
"total_bill=%{x}<br>tip=%{y}<br>hover_data_0=%{customdata[0]}")
tips = px.data.tips()
fig = px.scatter(
tips,
x="total_bill",
y="tip",
hover_data=[np.random.random(tips.shape[0])],
labels={"hover_data_0": "suppl"},
)
assert (fig.data[0]["hovertemplate"] ==
"total_bill=%{x}<br>tip=%{y}<br>suppl=%{customdata[0]}")
def test_int_col_names():
# DataFrame with int column names
lengths = pd.DataFrame(np.random.random(100))
fig = px.histogram(lengths, x=0)
assert np.all(np.array(lengths).flatten() == fig.data[0].x)
# Numpy array
ar = np.arange(100).reshape((10, 10))
fig = px.scatter(ar, x=2, y=8)
assert np.all(fig.data[0].x == ar[:, 2])
def test_repeated_name():
iris = px.data.iris()
fig = px.scatter(
iris,
x="sepal_width",
y="sepal_length",
hover_data=["petal_length", "petal_width", "species_id"],
custom_data=["species_id", "species"],
)
assert fig.data[0].customdata.shape[1] == 4
def test_arguments_not_modified():
iris = px.data.iris()
petal_length = iris.petal_length
hover_data = [iris.sepal_length]
fig = px.scatter(iris,
x=petal_length,
y="petal_width",
hover_data=hover_data)
assert iris.petal_length.equals(petal_length)
assert iris.sepal_length.equals(hover_data[0])
def test_with_index():
tips = px.data.tips()
fig = px.scatter(tips, x=tips.index, y="total_bill")
assert fig.data[0]["hovertemplate"] == "index=%{x}<br>total_bill=%{y}"
fig = px.scatter(tips, x=tips.index, y=tips.total_bill)
assert fig.data[0]["hovertemplate"] == "index=%{x}<br>total_bill=%{y}"
fig = px.scatter(tips,
x=tips.index,
y=tips.total_bill,
labels={"index": "number"})
assert fig.data[0]["hovertemplate"] == "number=%{x}<br>total_bill=%{y}"
# We do not allow "x=index"
with pytest.raises(ValueError) as err_msg:
fig = px.scatter(tips, x="index", y="total_bill")
assert "To use the index, pass it in directly as `df.index`." in str(
err_msg.value)
tips = px.data.tips()
tips.index.name = "item"
fig = px.scatter(tips, x=tips.index, y="total_bill")
assert fig.data[0]["hovertemplate"] == "item=%{x}<br>total_bill=%{y}"
def test_custom_data_scatter():
iris = px.data.iris()
# No hover, no custom data
fig = px.scatter(iris, x="sepal_width", y="sepal_length", color="species")
assert fig.data[0].customdata is None
# Hover, no custom data
fig = px.scatter(
iris,
x="sepal_width",
y="sepal_length",
color="species",
hover_data=["petal_length", "petal_width"],
)
for data in fig.data:
assert np.all(np.in1d(data.customdata[:, 1], iris.petal_width))
# Hover and custom data, no repeated arguments
fig = px.scatter(
iris,
x="sepal_width",
y="sepal_length",
hover_data=["petal_length", "petal_width"],
custom_data=["species_id", "species"],
)
assert np.all(fig.data[0].customdata[:, 0] == iris.species_id)
assert fig.data[0].customdata.shape[1] == 4
# Hover and custom data, with repeated arguments
fig = px.scatter(
iris,
x="sepal_width",
y="sepal_length",
hover_data=["petal_length", "petal_width", "species_id"],
custom_data=["species_id", "species"],
)
assert np.all(fig.data[0].customdata[:, 0] == iris.species_id)
assert fig.data[0].customdata.shape[1] == 4
assert (
fig.data[0].hovertemplate
== "sepal_width=%{x}<br>sepal_length=%{y}<br>petal_length=%{customdata[2]}<br>petal_width=%{customdata[3]}<br>species_id=%{customdata[0]}"
)
def plot_resultatsSummary(self, with_std=False):
obj = self.obj
if not with_std:
e = (obj.resultatsSummary(withStd=F).T.iplot(
kind="scatter", mode="markers", asFigure=T).update_config(
plotlyServerURL="").update_layout(width=600))
else:
g = (obj.resultatsSummary(
withStd=T,
withStdCol=T).T.rename_axis(index="model").reset_index())
dff = (g.dropCols(["Tr(std)", "Val(std)"]) >> df.gather(
"Var", "Values", ["Tr", "Val"])).assign(
ci=pd.concat([g.loc[:, "Tr(std)"], g.loc[:, "Val(std)"]],
ignore_index=T))
e = (px.scatter(dff,
x="model",
y="Values",
color="Var",
error_y="ci"))
return e
def test_several_dataframes():
df = pd.DataFrame(dict(x=[0, 1], y=[1, 10], z=[0.1, 0.8]))
df2 = pd.DataFrame(dict(time=[23, 26], money=[100, 200]))
fig = px.scatter(df, x="z", y=df2.money, size="x")
assert fig.data[0].hovertemplate == "z=%{x}<br>y=%{y}<br>x=%{marker.size}"
fig = px.scatter(df2, x=df.z, y=df2.money, size=df.z)
assert fig.data[
0].hovertemplate == "x=%{x}<br>money=%{y}<br>size=%{marker.size}"
# Name conflict
with pytest.raises(NameError) as err_msg:
fig = px.scatter(df, x="z", y=df2.money, size="y")
assert "A name conflict was encountered for argument y" in str(
err_msg.value)
with pytest.raises(NameError) as err_msg:
fig = px.scatter(df, x="z", y=df2.money, size=df.y)
assert "A name conflict was encountered for argument y" in str(
err_msg.value)
# No conflict when the dataframe is not given, fields are used
df = pd.DataFrame(dict(x=[0, 1], y=[3, 4]))
df2 = pd.DataFrame(dict(x=[3, 5], y=[23, 24]))
fig = px.scatter(x=df.y, y=df2.y)
assert np.all(fig.data[0].x == np.array([3, 4]))
assert np.all(fig.data[0].y == np.array([23, 24]))
assert fig.data[0].hovertemplate == "x=%{x}<br>y=%{y}"
df = pd.DataFrame(dict(x=[0, 1], y=[3, 4]))
df2 = pd.DataFrame(dict(x=[3, 5], y=[23, 24]))
df3 = pd.DataFrame(dict(y=[0.1, 0.2]))
fig = px.scatter(x=df.y, y=df2.y, size=df3.y)
assert np.all(fig.data[0].x == np.array([3, 4]))
assert np.all(fig.data[0].y == np.array([23, 24]))
assert fig.data[
0].hovertemplate == "x=%{x}<br>y=%{y}<br>size=%{marker.size}"
df = pd.DataFrame(dict(x=[0, 1], y=[3, 4]))
df2 = pd.DataFrame(dict(x=[3, 5], y=[23, 24]))
df3 = pd.DataFrame(dict(y=[0.1, 0.2]))
fig = px.scatter(x=df.y, y=df2.y, hover_data=[df3.y])
assert np.all(fig.data[0].x == np.array([3, 4]))
assert np.all(fig.data[0].y == np.array([23, 24]))
assert (fig.data[0].hovertemplate ==
"x=%{x}<br>y=%{y}<br>hover_data_0=%{customdata[0]}")
def test_px_templates():
import plotly_study.io as pio
import plotly_study.graph_objects as go
tips = px.data.tips()
# use the normal defaults
fig = px.scatter()
assert fig.layout.template == pio.templates[pio.templates.default]
# respect changes to defaults
pio.templates.default = "seaborn"
fig = px.scatter()
assert fig.layout.template == pio.templates["seaborn"]
# special px-level defaults over pio defaults
pio.templates.default = "seaborn"
px.defaults.template = "ggplot2"
fig = px.scatter()
assert fig.layout.template == pio.templates["ggplot2"]
# accept names in args over pio and px defaults
fig = px.scatter(template="seaborn")
assert fig.layout.template == pio.templates["seaborn"]
# accept objects in args
fig = px.scatter(template={})
assert fig.layout.template == go.layout.Template(data_scatter=[{}])
# read colorway from the template
fig = px.scatter(
tips,
x="total_bill",
y="tip",
color="sex",
template=dict(layout_colorway=["red", "blue"]),
)
assert fig.data[0].marker.color == "red"
assert fig.data[1].marker.color == "blue"
# default colorway fallback
fig = px.scatter(tips, x="total_bill", y="tip", color="sex", template=dict())
assert fig.data[0].marker.color == px.colors.qualitative.D3[0]
assert fig.data[1].marker.color == px.colors.qualitative.D3[1]
# pio default template colorway fallback
pio.templates.default = "seaborn"
px.defaults.template = None
fig = px.scatter(tips, x="total_bill", y="tip", color="sex")
assert fig.data[0].marker.color == pio.templates["seaborn"].layout.colorway[0]
assert fig.data[1].marker.color == pio.templates["seaborn"].layout.colorway[1]
# pio default template colorway fallback
pio.templates.default = "seaborn"
px.defaults.template = "ggplot2"
fig = px.scatter(tips, x="total_bill", y="tip", color="sex")
assert fig.data[0].marker.color == pio.templates["ggplot2"].layout.colorway[0]
assert fig.data[1].marker.color == pio.templates["ggplot2"].layout.colorway[1]
# don't overwrite top margin when set in template
fig = px.scatter(title="yo")
assert fig.layout.margin.t is None
fig = px.scatter()
assert fig.layout.margin.t == 60
fig = px.scatter(template=dict(layout_margin_t=2))
assert fig.layout.margin.t is None
# don't force histogram gridlines when set in template
pio.templates.default = "none"
px.defaults.template = None
fig = px.scatter(
tips, x="total_bill", y="tip", marginal_x="histogram", marginal_y="histogram"
)
assert fig.layout.xaxis2.showgrid
assert fig.layout.xaxis3.showgrid
assert fig.layout.yaxis2.showgrid
assert fig.layout.yaxis3.showgrid
fig = px.scatter(
tips,
x="total_bill",
y="tip",
marginal_x="histogram",
marginal_y="histogram",
template=dict(layout_yaxis_showgrid=False),
)
assert fig.layout.xaxis2.showgrid
assert fig.layout.xaxis3.showgrid
assert fig.layout.yaxis2.showgrid is None
assert fig.layout.yaxis3.showgrid is None
fig = px.scatter(
tips,
x="total_bill",
y="tip",
marginal_x="histogram",
marginal_y="histogram",
template=dict(layout_xaxis_showgrid=False),
)
assert fig.layout.xaxis2.showgrid is None
assert fig.layout.xaxis3.showgrid is None
assert fig.layout.yaxis2.showgrid
assert fig.layout.yaxis3.showgrid
def test_wrong_dimensions_mixed_case():
with pytest.raises(ValueError) as err_msg:
df = pd.DataFrame(dict(time=[1, 2, 3], temperature=[20, 30, 25]))
fig = px.scatter(df, x="time", y="temperature", color=[1, 3, 9, 5])
assert "All arguments should have the same length." in str(
err_msg.value)
def test_numpy_labels():
fig = px.scatter(x=[1, 2, 3], y=[2, 3, 4],
labels={"x": "time"}) # other labels will be kw arguments
assert fig.data[0]["hovertemplate"] == "time=%{x}<br>y=%{y}"
def test_size_column():
df = px.data.tips()
fig = px.scatter(df, x=df["size"], y=df.tip)
assert fig.data[0].hovertemplate == "size=%{x}<br>tip=%{y}"
def test_data_frame_from_dict():
fig = px.scatter({"time": [0, 1], "money": [1, 2]}, x="time", y="money")
assert fig.data[0].hovertemplate == "time=%{x}<br>money=%{y}"
assert np.all(fig.data[0].x == [0, 1])
def plot(self,
x=None,
y=None,
color=None,
by=None,
addTargetAuto=False,
types=None,
update_layout=None,
roundVal=2,
targetMode=None,
*args,
**xargs):
datas = self.obj.get()
target = self.obj.y.name if self.obj.y is not None else ""
targetI = self.obj.get() >> df.pull(
target) if target is not None and target != "" else None
xIType, yIType, colorIType, byIType = None, None, None, None
x, y, color, by = _get_and_checks([x, y, color, by], datas,
[None, None, None, None, False])
if (all([
i is None for i in [x, y, color, by]
])): #or all([i is None for i in [x,y]])) and addTargetAuto == False
raise Exception("plot impossible")
if all([i is None for i in [x, y, color, by]]):
types = types if types is not None else "bar"
if types in ["hist", "histogram", "bar"]:
argsx = xargs
if not secureAddArgs(self.obj.viz.plot_class_balance, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in self.obj.viz.plot_class_balance"""
)
argsx = removeBadArgs(self.obj.viz.plot_class_balance,
argsx)
return self.obj.viz.plot_class_balance(
title=f"Counts of '{target}' target variable",
yTitle="Count",
**argsx)
raise NotImplementedError(f"types '{types}' not again available")
if targetMode is not None and targetMode in ["x", "y", "color", "by"]:
if targetMode == "by":
by = target
elif targetMode == "color":
color = target
elif targetMode == "x":
x = target
elif targetMode == "y":
y = target
#exec(f"{targetMode}='{target}'",locals(),globals())
if x is not None:
xI, xIType = _get_dtype_and_data(x, datas)
if y is not None:
yI, yIType = _get_dtype_and_data(y, datas)
if color is not None:
colorI, colorIType = _get_dtype_and_data(color, datas)
if by is not None:
byI, byIType = _get_dtype_and_data(by, datas)
if all([i is None for i in [y, color, by]
]) and x is not None and addTargetAuto == False:
if pd.api.types.is_numeric_dtype(xIType):
types = types if types is not None else "histogram"
if types in ["hist", "histogram"]:
argsx = xargs
if not secureAddArgs(pex.histogram, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in pex.histogram""")
argsx = removeBadArgs(pex.histogram, argsx)
return pex.histogram(datas, x=x, **argsx)
if types in ["bar"]:
argsx = xargs
if not secureAddArgs(pex.bar, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in pex.bar""")
argsx = removeBadArgs(pex.bar, argsx)
return pex.bar(datas, x=x, **argsxg)
if types in ["points", "scatter"]:
sortOk = xargs.pop("sort", False)
x_ = np.sort(xI) if sortOk else x
sort_ = "sorted" if sortOk else ""
title_ = f"{types} plot of '{x}' {sort_}"
argsx = xargs
if not secureAddArgs(pex.scatter, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in pex.scatter""")
argsx = removeBadArgs(pex.scatter, argsx)
return pex.scatter(datas, x=x_, title=title_,
**argsx).update_layout(
yaxis_title="Range", xaxis_title=x)
raise NotImplementedError(
f"types '{types}' not again available")
elif pd.api.types.is_categorical_dtype(xIType):
types = types if types is not None else "bar"
if types in ["hist", "histogram", "bar"]:
catOrder = xI.cat.categories
#print(catOrder)
#print(xI.value_counts())
def fnCount(*args, **xargs):
rep = xI.value_counts(
*args, **xargs).sort_index(level=catOrder)
#rep=rep.to_frame().rename_cols(["Count"])
return rep
#fnCount=lambda *args,**xargs: xI.value_counts(*args,**xargs).sort_index(level=catOrder).as_frame()
argsx = xargs
if not secureAddArgs(self.plot_bar_count, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in self.plot_bar_count""")
argsx = removeBadArgs(self.plot_bar_count, argsx)
return self.plot_bar_count(
fnCount,
lambda *args, **xargs: xI.value_counts(
).sort_index(level=catOrder).to_frame().rename_cols(
["count"]).sort_index(level=catOrder).reset_index(
).rename_cols([x, "Count"]),
ind="x",
plot_kwargs=dict(y="Count",
barmode="overlay",
category_orders=dict(
zip([x], [list(catOrder)]))),
cbName=x,
**argsx).update_layout(yaxis_title="Count",
xaxis_title=x,
title=f"Count by '{x}'")
if all([i is None for i in [y, color, by]
]) and x is not None and addTargetAuto == True or (
all([i is None for i in [y, by]]) and x is not None
and color is not None and addTargetAuto == False):
# print("icici")
color = target if color is None else color
colorI = targetI if ("colorI" not in globals()
or colorI is None) else colorI
if pd.api.types.is_numeric_dtype(xIType):
types = types if types is not None else "histogram"
if types in ["hist", "histogram"]:
argsx = xargs
if not secureAddArgs(pex.histogram, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in pex.histogram""")
argsx = removeBadArgs(pex.histogram, argsx)
return pex.histogram(datas, x=x, color=color, **argsx)
if types in ["bar"]:
argsx = xargs
if not secureAddArgs(pex.bar, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in pex.bar""")
argsx = removeBadArgs(pex.bar, argsx)
return pex.bar(datas, x=x, color=color, **argsxg)
if types in ["points", "scatter"]:
sortOk = xargs.pop("sort", False)
x_ = np.sort(xI) if sortOk else x
sort_ = "sorted" if sortOk else ""
title_ = f"{types} plot of '{x}' {sort_}"
argsx = xargs
if not secureAddArgs(pex.scatter, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in pex.scatter""")
argsx = removeBadArgs(pex.scatter, argsx)
return pex.scatter(datas,
x=x_,
title=title_,
color=color,
**argsx).update_layout(
yaxis_title="Range", xaxis_title=x)
raise NotImplementedError(
f"types '{types}' not again available")
elif pd.api.types.is_categorical_dtype(xIType):
types = types if types is not None else "bar"
if types in ["hist", "histogram", "bar"]:
catOrder = xI.cat.categories
#print(catOrder)
#print(xI.value_counts())
def fnCount(*args, **xargs):
rep = xI.value_counts(
*args, **xargs).sort_index(level=catOrder)
#rep=rep.to_frame().rename_cols(["Count"])
return rep
#fnCount=lambda *args,**xargs: xI.value_counts(*args,**xargs).sort_index(level=catOrder).as_frame()
argsx = xargs
if not secureAddArgs(self.plot_bar_count, xargs):
import warnings
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in self.plot_bar_count""")
argsx = removeBadArgs(self.plot_bar_count, argsx)
return pex.histogram(datas,
x=x,
color=color,
barmode="group")
# return self.plot_bar_count(fnCount,lambda *args,**xargs:xI.value_counts().sort_index(level=catOrder).to_frame().rename_cols(["count"]).sort_index(level=catOrder).reset_index().rename_cols([x,"Count"]),ind="x",plot_kwargs=dict(y="Count",barmode="overlay",category_orders=dict(zip([x],[list(catOrder)]))),cbName=x,**argsx).update_layout(yaxis_title="Count",xaxis_title=x,title=f"Count by '{x}'")
raise NotImplementedError(f"types '{types}' not again available")
if all([i is None for i in [x, color, by]
]) and y is not None and addTargetAuto == False:
if pd.api.types.is_numeric_dtype(yIType):
types = types if types is not None else "scatter"
if types in ["points", "scatter"]:
sortOk = xargs.pop("sort", False)
y_ = np.sort(yI) if sortOk else yI
sort_ = "sorted" if sortOk else ""
title_ = f"{types} plot of '{y}' {sort_}"
argsx = xargs
if not secureAddArgs(pex.scatter, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in pex.scatter""")
argsx = removeBadArgs(pex.scatter, argsx)
return pex.scatter(datas, y=y_, title=title_,
**argsx).update_layout(
xaxis_title="Range", yaxis_title=y)
raise NotImplementedError(
f"when only y and not target and types '{types}' -> not again available"
)
if all([i is None for i in [x, y, by]
]) and color is not None and addTargetAuto == False:
#print("ici")
if pd.api.types.is_numeric_dtype(colorIType):
raise NotImplementedError(
f"when only color and types '{colorIType}' -> not again available"
)
elif pd.api.types.is_categorical_dtype(colorIType):
types = types if types is not None else "bar"
if types in ["bar"]:
argsx = xargs
if not secureAddArgs(pex.bar, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in pex.bar""")
argsx = removeBadArgs(pex.bar, argsx)
catOrder = colorI.cat.categories
fnCount = lambda *args, **xargs: colorI.value_counts(
*args, **xargs).sort_index(level=catOrder)
title_ = f"Bar plot of '{color}'"
rep = pex.bar(datas,
x=color,
color=color,
barmode="overlay",
category_orders=dict(
zip([color], [list(catOrder)])),
**argsx).update_layout(title=title_,
xaxis_title=color,
yaxis_title="Count")
for i, y, z in zip(rep.data, fnCount(),
fnCount(normalize=True) * 100):
i.text = str(y) + " (" + str(np.round(z,
roundVal)) + "%)"
i.hovertemplate = f"{i.name}<br>Count={y}"
return rep
if types in ["count"]:
argsx = xargs
if not secureAddArgs(self.plot_bar_count, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in self.plot_bar_count""")
argsx = removeBadArgs(self.plot_bar_count, argsx)
catOrder = colorI.cat.categories
fnCount = lambda *args, **xargs: colorI.value_counts(
*args, **xargs).sort_index(level=catOrder)
return self.plot_bar_count(fnCount,
plot_kwargs=dict(color=color),
**argsx).update_layout(
xaxis_title=color,
yaxis_title="Count")
if types in ["hist", "histogram"]:
argsx = xargs
if not secureAddArgs(pex.histogram, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in pex.histogram""")
argsx = removeBadArgs(pex.bar, argsx)
return pex.histogram(datas, x=color, color=color,
**argsx).update_layout(
xaxis_title=color,
yaxis_title="Count")
if all([i is None for i in [x, y, color]
]) and by is not None and addTargetAuto == False:
if pd.api.types.is_numeric_dtype(byIType):
raise NotImplementedError(
f"when only color and types '{byIType}' -> not again available"
)
elif pd.api.types.is_categorical_dtype(byIType):
types = types if types is not None else "hist"
ncols = xargs.pop("ncols", 4)
if y is not None:
pass
multiple = False
# if all([i is None for i in [y,by]]) and x is not None and color in [None,target] and addTargetAuto == True:
# if pd.api.types.is_numeric_dtype(xIType):
# types=types if types is not None else "hist"
# if types in ["hist","histogram"]:
# argsx=xargs
# if not secureAddArgs(pex.histogram,xargs):
# with showWarningsTmp:
# warnings.warn(f"""
# error in {xargs} not in pex.histogram""")
# argsx=removeBadArgs(pex.histogram,argsx)
# title_=f"Histogram of '{x}' colored by '{target}'"
# argsx["title"]=title_
# return pex.histogram(datas,x=x,color=target,**argsx)
# raise NotImplementedError(f"'{xIType}' -> not again available")
if all([
i is None for i in [y, by]
]) and x is not None and color is not None and addTargetAuto == False:
if pd.api.types.is_numeric_dtype(xIType):
types = types if types is not None else "hist"
if types in ["hist", "histogram"]:
argsx = xargs
if not secureAddArgs(pex.histogram, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in pex.histogram""")
argsx = removeBadArgs(pex.histogram, argsx)
if not pd.api.types.is_categorical_dtype(colorIType):
raise NotImplementedError(
f"color '{colorIType}' and x -> not again available"
)
title_ = f"Histogram of '{x}' colored by '{color}'"
argsx["title"] = title_
catOrder = targetI.cat.categories
argsx["category_orders"] = dict(
zip([target, color],
[list(catOrder),
list(colorI.cat.categories)]))
return pex.histogram(datas, x=x, color=color, **argsx)
raise NotImplementedError(f"'{xIType}' -> not again available")
if all([
i is None for i in [by, y]
]) and x is not None and color is not None and addTargetAuto == True:
if pd.api.types.is_numeric_dtype(xIType):
types = types if types is not None else "hist"
if types in ["hist", "histogram"]:
argsx = xargs
if not pd.api.types.is_categorical_dtype(colorIType):
raise NotImplementedError(
f"color '{colorIType}' and x -> not again available"
)
nbins = argsx.pop("nbins", None)
by = target
byI = targetI
if nbins is not None:
if isinstance(nbins, Iterable) and not isinstance(
nbins, str) and len(nbins) > 1:
if not secureAddArgs(createHistMultiBins, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in createHistMultiBins"""
)
argsx = removeBadArgs(createHistMultiBins,
argsx)
title_ = f"Histogram of '{x}' by '{by}' colored by '{color}'"
#print(title_)
return createHistMultiBins(datas,
x,
color,
colorI,
by,
byI,
nbins,
title=title_,
**argsx)
else:
argsx["nbins"] = nbins
if not secureAddArgs(pex.histogram, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in pex.histogram""")
argsx = removeBadArgs(pex.histogram, argsx)
title_ = f"Histogram of '{x}' colored by '{color}' by '{by}'"
argsx["title"] = title_
catOrder = targetI.cat.categories
argsx["category_orders"] = dict(
zip([target, color, by], [
list(catOrder),
list(colorI.cat.categories),
list(byI.cat.categories)
]))
return pex.histogram(datas,
x=x,
color=color,
facet_col=by,
**argsx)
raise NotImplementedError(f"'{xIType}' -> not again available")
if all([
i is None for i in [color, y]
]) and x is not None and by is not None and addTargetAuto == True:
if pd.api.types.is_numeric_dtype(xIType):
types = types if types is not None else "hist"
if types in ["hist", "histogram"]:
argsx = xargs
nbins = argsx.pop("nbins", None)
if nbins is not None:
if isinstance(nbins, Iterable) and not isinstance(
nbins, str) and len(nbins) > 1:
if not secureAddArgs(createHistMultiBins, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in createHistMultiBins"""
)
argsx = removeBadArgs(createHistMultiBins,
argsx)
color = target
colorI = targetI
title_ = f"Histogram of '{x}' by '{by}' colored by '{color}'"
#print(title_)
return createHistMultiBins(datas,
x,
color,
colorI,
by,
byI,
nbins,
title=title_,
**argsx)
else:
argsx["nbins"] = nbins
if not secureAddArgs(pex.histogram, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in pex.histogram""")
argsx = removeBadArgs(pex.histogram, argsx)
title_ = f"Histogram of '{x}' by '{by}' colored by '{target}'"
argsx["title"] = title_
catOrder = targetI.cat.categories
argsx["category_orders"] = dict(
zip([target, by],
[list(catOrder),
list(byI.cat.categories)]))
return pex.histogram(datas,
x=x,
color=target,
facet_col=by,
**argsx)
raise NotImplementedError(f"'{xIType}' -> not again available")
if all([i is None for i in [x, y, by]
]) and color is not None and addTargetAuto == True:
# print(pd.api.types.is_categorical_dtype(colorIType))
if pd.api.types.is_numeric_dtype(colorIType):
raise NotImplementedError(
f"when only color and types '{colorIType}' -> not again available"
)
elif pd.api.types.is_categorical_dtype(colorIType):
types = types if types is not None else "bar"
if types in ["bar"]:
argsx = xargs
if not secureAddArgs(pex.bar, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in pex.bar""")
argsx = removeBadArgs(pex.bar, argsx)
catOrder = targetI.cat.categories
fnCount = lambda *args, **xargs: targetI.value_counts(
*args, **xargs).sort_index(level=catOrder)
title_ = f"Bar plot of '{target}'"
datag = pd.crosstab(targetI,
colorI).sort_index(level=catOrder)
#print(datag)
#rep=datag.iplot(kind="bar")
#print(catOrder)
#print(dict(zip([target,color],[list(catOrder),list(colorI.cat.categories)])))
rep = pex.histogram(
datas,
x=target,
color=color,
barmode="group",
category_orders=dict(
zip([target, color],
[list(catOrder),
list(colorI.cat.categories)])),
**argsx).update_traces(showlegend=T).update_layout(
showlegend=T,
title=title_,
xaxis_title=target,
yaxis_title="Count")
# for i,y,z in zip(rep.data,fnCount(),fnCount(normalize=True)*100):
# i.text=str(y)+" ("+str(np.round(z,roundVal))+"%)"
# i.hovertemplate=f"{i.name}<br>Count={y}"
for j in rep.data:
j.legendgroup
return rep
if types in ["count"]:
argsx = xargs
if not secureAddArgs(self.plot_bar_count, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in self.plot_bar_count""")
argsx = removeBadArgs(self.plot_bar_count, argsx)
catOrder = colorI.cat.categories
fnCount = lambda *args, **xargs: colorI.value_counts(
*args, **xargs).sort_index(level=catOrder)
return self.plot_bar_count(fnCount,
plot_kwargs=dict(color=color),
**argsx).update_layout(
xaxis_title=color,
yaxis_title="Count")
if types in ["hist", "histogram"]:
argsx = xargs
if not secureAddArgs(pex.histogram, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in pex.histogram""")
argsx = removeBadArgs(pex.bar, argsx)
return pex.histogram(datas, x=color, color=color,
**argsx).update_layout(
xaxis_title=color,
yaxis_title="Count")
if all([i is None for i in [y]]) and all(
[i is not None for i in [color, x, by]]) and addTargetAuto == True:
if pd.api.types.is_categorical_dtype(
colorIType) and pd.api.types.is_categorical_dtype(
xIType) and pd.api.types.is_categorical_dtype(byIType):
types = types if types is not None else "point"
if types in ["point", "pointplot"]:
argsx = xargs
if not secureAddArgs(self.plot_pointplot, xargs):
with showWarningsTmp:
warnings.warn(f"""
error in {xargs} not in self.plot_pointplot""")
argsx = removeBadArgs(self.plot_pointplot, argsx)
return self.plot_pointplot(x=x,
y=target,
color=color,
by=by,
**argsx)
raise NotImplementedError(
f"when only color and types '{colorIType}' -> not again available"
)
raise NotImplementedError(
f"when color:'{colorIType}' x:'{xIType}' y: '{yIType}' by: '{byIType}' addTargetAuto: '{addTargetAuto}' -> not again available"
)
def test_name_heuristics():
df = pd.DataFrame(dict(x=[0, 1], y=[3, 4], z=[0.1, 0.2]))
fig = px.scatter(df, x=df.y, y=df.x, size=df.y)
assert np.all(fig.data[0].x == np.array([3, 4]))
assert np.all(fig.data[0].y == np.array([0, 1]))
assert fig.data[0].hovertemplate == "y=%{marker.size}<br>x=%{y}"
def test_numpy():
fig = px.scatter(x=[1, 2, 3], y=[2, 3, 4], color=[1, 3, 9])
assert np.all(fig.data[0].x == np.array([1, 2, 3]))
assert np.all(fig.data[0].y == np.array([2, 3, 4]))
assert np.all(fig.data[0].marker.color == np.array([1, 3, 9]))
def test_wrong_column_name():
with pytest.raises(ValueError) as err_msg:
fig = px.scatter(px.data.tips(), x="bla", y="wrong")
assert "Value of 'x' is not the name of a column in 'data_frame'" in str(
err_msg.value)
def test_missing_data_frame():
with pytest.raises(ValueError) as err_msg:
fig = px.scatter(x="arg1", y="arg2")
assert "String or int arguments are only possible" in str(
err_msg.value)
def test_wrong_dimensions_of_array():
with pytest.raises(ValueError) as err_msg:
fig = px.scatter(x=[1, 2, 3], y=[2, 3, 4, 5])
assert "All arguments should have the same length." in str(
err_msg.value)