def render_vis(self, query_raw):
# type: (str) -> VegaLite
response = self.analyze_query(query_raw)
if len(response['visList']) == 0:
print("No best Viz; please try again.")
return VegaLite({})
return VegaLite(response['visList'][0]['vlSpec'])
def vgplot_bar(df, x, y):
if x is None:
if df.index.name is None:
df.index.name = 'index'
x = df.index.name
df = df.reset_index()
assert x in df.columns
if y is not None:
assert y in df.columns
df = df[[x, y]]
df = df.melt([x], var_name='variable', value_name='value')
D = {
"$schema": "https://vega.github.io/schema/vega-lite/v2.json",
"mark": "bar",
"encoding": {
"x": {
"field": x,
"type": "ordinal"
},
"y": {
"field": "value",
"type": "quantitative"
},
"color": {
"field": "variable",
"type": "nominal"
}
}
}
return VegaLite(D, data=df)
def build_student_feedback_histogram(height, width, data, questions, question):
return VegaLite({
"$schema": "https://vega.github.io/schema/vega-lite/v2.0.json",
# "title": """",
"height": height,
"width": width,
"mark": "bar",
"encoding": {
"x": {
"field": "answer",
"type": "nominal",
"sort": None,
"axis": { "title": "" }
},
"y": {
"field": "q{0}".format(question),
"type": "quantitative",
"axis": { "title": "count" }
},
"color": {
"field": "answer",
"type": "nominal",
"legend": None,
}
}
}, data)
def build_grade_histogram(height, width, data):
return VegaLite({
"$schema": "https://vega.github.io/schema/vega-lite/v2.0.json",
"title": "Module attainment histogram",
"height": height,
"width": width,
"layer": [
{
"mark": "bar",
"encoding": {
"x": {
"bin": { "step": 10 },
"field": "final_grade",
"type": "quantitative",
"axis": { "title": "Grade", "tickCount": 10 },
"scale": { "domain": [0, 100] }
},
"y": {
"aggregate": "count",
"type": "quantitative",
"axis": { "title": "Number of students" }
},
"color": {
"bin": { "step": 10 },
"field": "final_grade",
"type": "quantitative",
"legend": None
}
}
},
{
"mark": "rule",
"encoding": {
"x": {
"value": width/100 * 40
},
"size": {
"value": 2
},
"color": {
"value": "#ccc"
}
}
},
{
"mark": "rule",
"encoding": {
"x": {
"value": width/100 * 60
},
"size": {
"value": 2
},
"color": {
"value": "#ccc"
}
}
}
]
}, data)
def vegalite_graph(self, shape=(400, 300), simple=False):
from vega import VegaLite
import pandas
idx = np.linspace(0, self.fpr.size - 1, 500, dtype=np.uint)
return VegaLite(
{
"title": {
"text": "AUC = %.3f " % self.auc,
"anchor": "end"
},
"width": shape[0],
"height": shape[1],
"mark": {
"type": "area",
"line": True,
"point": True
},
"encoding": {
"y": {
"type": "quantitative",
"field": "tpr",
"title": "True Positive Rate"
},
"x": {
"type": "quantitative",
"field": "fpr",
"title": "False Positive Rate"
},
"tooltip": [{
"type": "quantitative",
"field": "tpr",
"title": "True Positive"
}, {
"type": "quantitative",
"field": "fpr",
"title": "False Positive"
}, {
"type": "quantitative",
"field": "thresholds",
"title": "Threshold"
}]
}
},
pandas.DataFrame(
dict(fpr=self.fpr[idx],
tpr=self.tpr[idx],
thresholds=self.thresholds[idx])))
def build_grade_comparison_plot(height, width, data):
return VegaLite({
"$schema": "https://vega.github.io/schema/vega-lite/v2.0.json",
"title": "Module performance comparison",
"height": height,
"width": width,
"layer": [
{
"mark": "circle",
"encoding": {
"x": {
"field": "average_grade",
"type": "quantitative",
"axis": { "title": "Average Grade" },
"scale": { "domain": [0, 100] }
},
"y": {
"field": "final_grade",
"typs": "quantitative",
"axis": { "title": "Module grade" },
"scale": { "domain": [0, 100] }
},
}
},
{
"mark": "rule",
"encoding": {
"x": {
"value": 0
},
"y": {
"value": height
},
"size": {
"value": 1
},
"color": {
"value": "red"
}
}
}
]
}, data)
def vgplot_scatter(df, x, y):
assert x in df.columns
assert y in df.columns
df = df[[x, y]]
D = {
"$schema": "https://vega.github.io/schema/vega-lite/v2.json",
"mark": "circle",
"encoding": {
"x": {
"field": x,
"type": "quantitative"
},
"y": {
"field": y,
"type": "quantitative"
},
}
}
return VegaLite(D, data=df)
def static_vis(self, **kwargs):
"""This function is called inplace of `vis` for reactive cells, whose participation in the event loop is different from the others.
Optional arguments (Midas will guess the rest)
mark -- "bar" | "circle" | "line"
x -- name of the column to be the x axis
x_type -- "ordinal" | "quantitative" | "temporal"
y -- name of the column to be the y axis
y_type -- "ordinal" | "quantitative" | "temporal"
sort -- "" | "x" | "y" | "-x" | "-y"
"""
spec = parse_encoding(kwargs, self)
if spec:
sanity_check_spec_with_data(spec, self)
# do show me
vg_spec = static_vega_gen(spec, self)
# print(vg_spec)
return VegaLite(vg_spec)
return None
def see(self, *fields, anchor=None, name=None):
if (name is not None):
viewname = '\"{0}\"'.format(name)
else:
viewname = Chart.DEFAULT_NAME
fields = list(fields)
if (not fields):
fields = self._recent_fields
elif (all([x[0] == '+' for x in fields])):
fields = [x[1:] for x in fields] + self._recent_fields
elif (all([x[0] == '-' for x in fields])):
fields = filter(lambda x : x not in [y[1:] for y in fields], self._recent_fields)
self._recent_fields = fields
partial, asp = self._query(fields, anchor, viewname)
# print('\n'.join(asp))
self._sol = draco(asp)
if (name is not None):
self._anchors[viewname] = anchor_spec(partial, self._sol.props[viewname], viewname)
return VegaLite(self._sol.as_vl(viewname), self._data)
def _ipython_display_(self):
from IPython.display import display
from vega import VegaLite
display(VegaLite(self.to_dict()))
def _ipython_display_(self):
"""Use the vega package to display in the classic Jupyter Notebook."""
from IPython.display import display
from vega import VegaLite
display(VegaLite(self.to_dict()))
def _get_render(self): vegalite = self._get_vegalite() return VegaLite(vegalite, self._data)
def __init__(self, *params, **args):
VegaLite.__init__(self, *params, **args)
self.ID = uuid.uuid4()
self.InitNeeded = True
def _get_render(self):
return VegaLite(self._get_vegalite(), self._data)
def render_vegalite(vis):
# Render a visualization using vegalite
VegaLite(vis.to_vl_obj()).display()
