def create_line_plot(df):
""" create a mg line plot
Args:
df (pandas.DataFrame): data to plot
"""
fig = Figure("/mg/line_plot/", "mg_line_plot")
fig.graphics.transition_on_update(True)
fig.graphics.animate_on_load()
fig.layout.set_size(width=450, height=200)
fig.layout.set_margin(left=40, right=40)
return LineChart(df, fig, "Date", ["value"],
init_params={"Data": "Steps"}, timeseries=True)
def create_histogram(df):
""" create a mg line plot
Args:
df (pandas.DataFrame): data to plot
"""
fig = Figure("/mg/histogram/", "mg_histogram")
fig.layout.set_size(width=450, height=200)
fig.layout.set_margin(left=40, right=40)
fig.graphics.animate_on_load()
# Make a histogram with 20 bins
return Histogram(df, fig, "value", 20, init_params={"Data": "Steps"})
def create_scatterplot(df):
""" create a mg line plot
Args:
df (pandas.DataFrame): data to plot
"""
fig = Figure("/mg/scatter/", "mg_scatter")
fig.layout.set_size(width=450, height=200)
fig.layout.set_margin(left=40, right=40)
fig.graphics.animate_on_load()
init_params = {"Data": "Steps"}
def get_data():
y = request.args.get("Data", "Steps")
return jsonify(ScatterPlot.to_json(df, "Steps", y))
# Make a histogram with 20 bins
return ScatterPlot(df, fig, "Steps", "Distance",
init_params={}, route_func=get_data)
def create_line_plot(df, i, name, n, titel):
""" create a mg line plot
Args:
df (pandas.DataFrame): data to plot
"""
fig = Figure("/mg/line_plot" + str(i), "mg_line_plot")
fig.graphics.transition_on_update(True)
fig.graphics.animate_on_load()
fig.graphics.point_size(2.0)
fig.layout.set_size(width=450 + 10 * i, height=210)
fig.layout.set_margin(left=40, right=40)
fig.axes.set_min_y_from_data(True)
fig.axes.set_inflator(1.005)
fig.axes.set_xticks_count(n)
return LineChart(df=df,
figure=fig,
x="Year",
y=["value"],
init_params={"Data": name},
timeseries=True,
title=titel)
filter_style="''")
# Read in the data and stack it, so that we can filter on columns
df = pd.read_csv("fitbit_data.csv")
sf = df.set_index("Date").stack().reset_index()
sf = sf.rename(columns={"level_1": "Data", 0: "value"})
# Make a Button
cols = [c for c in df.columns if c != "Date"]
btn = SelectButton("Data", cols, "Data", "Steps")
# Make a FilterFrame and add the button to the UI
ui.add_filter(btn)
# Make a Figure, add some settings, make a line plot
fig = Figure("/mgchart/", "mychart")
fig.graphics.transition_on_update(True)
fig.graphics.animate_on_load()
fig.layout.set_size(width=450, height=200)
fig.layout.set_margin(left=40, right=40)
lc = LineChart(sf, fig, "Date", ["value"], init_params={"Data": "Steps"}, timeseries=True)
ui.add_chart(lc)
# Now make a FilterFrame for the histogram
hFig = Figure("/mghist/", "myhist")
hFig.layout.set_size(width=450, height=200)
hFig.layout.set_margin(left=40, right=40)
hFig.graphics.animate_on_load()
# Make a histogram with 20 bins
hc = Histogram(sf, hFig, "value", 20, init_params={"Data": "Steps"})
ui.add_chart(hc)
def uploaded_file(filename):
extension = get_extension(filename)
path = os.path.join(app.config['UPLOAD_FOLDER'], extension, filename)
txt_path = os.path.join(app.config['UPLOAD_FOLDER'], 'txt', filename)
if extension == "pdf":
txt_path += '.txt'
if not os.path.isfile(txt_path):
#Layout preservation crucial to preserve clues about tabular data
cmd = "pdftotext -layout %s %s" % (path, txt_path)
os.system(cmd)
min_columns = request.args.get('min_columns')
tables = return_tables(txt_path)
#Construct histogram
lines_per_page = 80
nr_data_rows = []
for b, t in tables.iteritems():
e = t['end_line']
#print b, e
for l in range(b, e):
page = l / lines_per_page
if len(nr_data_rows) <= page:
nr_data_rows += ([0] * (page - len(nr_data_rows) + 1))
nr_data_rows[page] += 1
dr = pd.DataFrame()
dr['value'] = nr_data_rows
dr['page'] = range(0, len(dr))
js_layout = filename + ".js"
ui_show = UILayout("FilterChart",
"../static/bower_components/pyxley/build/pyxley.js",
"component_id",
filter_style="''")
if filename in all_charts:
print "old/update ui", filename
path_to_fig = '/show/line/' + filename
#del all_charts[filename]
#hFig = Figure(path_to_fig, "line")
#bc = LineChart(dr, hFig, "page", ["page"], "Rows containing Data per Page")
elif True:
print "new ui", filename
# Make a Button
cols = ["page"]
btn = SelectButton("Data", cols, "Data", "Data Rows per Page")
# Make a FilterFrame and add the button to the UI
ui_show.add_filter(btn)
# Now make a FilterFrame for the histogram
path_to_fig = '/show/line/' + filename
hFig = Figure(path_to_fig, "line")
hFig.layout.set_size(width=1000, height=300)
hFig.layout.set_margin(left=80, right=80)
#hFig.graphics.animate_on_load()
bc = LineChart(dr, hFig, "page", ["value"],
"Rows containing Data per Page")
ui_show.add_chart(bc)
all_charts[filename] = bc
sb = ui_show.render_layout(app, "./static/ug/" + js_layout)
_scripts = ["ug/" + js_layout]
notices = ['Extraction Results for ' + filename, 'Ordered by lines']
dfs = (table_to_df(table).to_html() for table in tables.values())
headers = []
for t in tables.values():
if 'header' in t:
headers.append(t['header'])
else:
headers.append('-')
line_nrs = [('line %i-%i' % (t['begin_line'], t['end_line']))
for t in tables.values()]
#headers = ['aslkdfjas', ' alsdkfjasoedf']
return render_template('index.html',
title=TITLE + ' - ' + filename,
base_scripts=scripts,
page_scripts=_scripts,
css=css,
notices=notices,
tables=dfs,
headers=headers,
line_nrs=line_nrs)
for k, table in tables.iteritems():
df = table_to_df(table)
print k, ' !!! ', table['header'], ' -----------------'
print df.head()
#print dr
# Make a Button
cols = [c for c in df.columns if c != "Date"]
btn = SelectButton("Data", cols, "Data", "Steps")
# Make a FilterFrame and add the button to the UI
ui.add_filter(btn)
# Now make a FilterFrame for the histogram
hFig = Figure("/mghist/", "myhist")
hFig.layout.set_size(width=450, height=200)
hFig.layout.set_margin(left=40, right=40)
hFig.graphics.animate_on_load()
# Make a histogram with 20 bins
hc = Histogram(sf, hFig, "value", 20, init_params={"Data": "Steps"})
ui.add_chart(hc)
# Let's play with our input
df["Date"] = pd.to_datetime(df["Date"])
df["week"] = df["Date"].apply(lambda x: x.isocalendar()[1])
gf = df.groupby("week").agg({
"Date": [np.min, np.max],
"Steps": np.sum,
"Calories Burned": np.sum,
"Distance": np.sum