santosjorge / cufflinks Goto Github PK

View Code? Open in Web Editor NEW

3.0K 3.0K 674.0 17.44 MB

Productivity Tools for Plotly + Pandas

License: MIT License

Python 0.44% Jupyter Notebook 99.56%

cufflinks's People

Contributors

Stargazers

Watchers

Forkers

drjmcauliffe michaelgerace thalesians chriddyp jeffatennis nehalecky eliaprns tijptjik yw288 immortalin jmportilla jmhehir rlugojr mistertooth skennedy-gpsw shivam6294 udhai17 sethmenghi lizhihao1990 chaoyue0307 lcota marvely jdion prakashraaz carlespoles srinyuta ozans tristanboudreault pangacademy guajasdev pmuller spark-lin veghbernadett daliati aprilwebster apetresc kannes danielchalef torough cailiang9 gnu-user totalgood qbonis yukinjie omarberg birajparikh16 jayden11 seanzhou1023 sumonkantidey bambyg gurjeetseh unsteadynightmare ranjiththavamaniraj exzod evetricia lucainnocenti paurichardson ccybnumath mistaka0s baudryj zarppy datamics ketigid punch872 carollei926 yueleng aisuluomar iambigmomma jaeyangp jaymichael78 u1and0 ajayshewale www3838438 nvempati letristanb muhammedalaa yongho pvirenj zoe0lin sycheniris marlowe97 cneiderer merishnasuwal cooperok github-hongweizhang peeedge neosun100 big-o darshanpatel1992 dandiestsquare1 edulongo7 bigdatazhangsir jpoles1 loveplay1983 latisrox gjeusel csharphai sngupta07 dhruvjain2000 hbcbh1999

cufflinks's Issues

Python 3 bug in colors.py

Python 3 upgrades unicode to str and string to bytes. I believe the fix is as follows:

in ~/anaconda/lib/python3.6/site-packages/cufflinks/colors.py

if type(scale) in (bytes,str):
		scale=get_scales(scale)
	else:
		if type(scale)!=list:
			raise Exception("scale needs to be either a scale name or list of colors")
		
	cs=[[1.0*c/(len(scale)-1),scale[c]] for c in range(len(scale))]
	cs.sort()
	return cs

what do the colors represent in the bubble chart?

import pandas as pd
df = pd.read_csv('http://www.stat.ubc.ca/~jenny/notOcto/STAT545A/examples/gapminder/data/gapminderDataFiveYear.txt', sep='\t')
df2007 = df[df.year==2007]

df2007.iplot(kind='bubble', x='gdpPercap', y='lifeExp', size='pop', text='country', filename='cufflinks/simple-bubble-chart')

I think I expected all of these points to have the same color

kind='scatter3d' or kind='bubble3d' not working when connecting to dataframe

import cufflinks as cf
from plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot
init_notebook_mode(connected=True)
cf.go_offline()

df3 = pd.DataFrame({'x':list(x_pca3[:,0]),
                    'y':list(x_pca3[:,1]),
                    'z':list(x_pca3[:,2])
                  })

df3.iplot(kind='scatter3d')

Error message:

---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
/usr/local/lib/python3.6/site-packages/pandas/core/indexes/base.py in get_loc(self, key, method, tolerance)
   2392             try:
-> 2393                 return self._engine.get_loc(key)
   2394             except KeyError:

pandas/_libs/index.pyx in pandas._libs.index.IndexEngine.get_loc (pandas/_libs/index.c:5239)()

pandas/_libs/index.pyx in pandas._libs.index.IndexEngine.get_loc (pandas/_libs/index.c:5085)()

pandas/_libs/hashtable_class_helper.pxi in pandas._libs.hashtable.PyObjectHashTable.get_item (pandas/_libs/hashtable.c:20405)()

pandas/_libs/hashtable_class_helper.pxi in pandas._libs.hashtable.PyObjectHashTable.get_item (pandas/_libs/hashtable.c:20359)()

KeyError: 12

During handling of the above exception, another exception occurred:

KeyError                                  Traceback (most recent call last)
<ipython-input-45-c3f461ea156a> in <module>()
----> 1 df3.iplot(kind='bubble3d')

/usr/local/lib/python3.6/site-packages/cufflinks/plotlytools.py in _iplot(self, data, layout, filename, sharing, kind, title, xTitle, yTitle, zTitle, theme, colors, colorscale, fill, width, dash, mode, symbol, size, barmode, sortbars, bargap, bargroupgap, bins, histnorm, histfunc, orientation, boxpoints, annotations, keys, bestfit, bestfit_colors, mean, mean_colors, categories, x, y, z, text, gridcolor, zerolinecolor, margin, labels, values, secondary_y, secondary_y_title, subplots, shape, error_x, error_y, error_type, locations, lon, lat, asFrame, asDates, asFigure, asImage, dimensions, asPlot, asUrl, online, **kwargs)
    852                                 df['index']=keys
    853                                 if kind=='bubble3d':
--> 854                                         rg=self[size].values
    855                                         size=[int(100*(float(_)-rg.min())/(rg.max()-rg.min()))+12 for _ in rg]
    856                                 else:

/usr/local/lib/python3.6/site-packages/pandas/core/frame.py in __getitem__(self, key)
   2060             return self._getitem_multilevel(key)
   2061         else:
-> 2062             return self._getitem_column(key)
   2063 
   2064     def _getitem_column(self, key):

/usr/local/lib/python3.6/site-packages/pandas/core/frame.py in _getitem_column(self, key)
   2067         # get column
   2068         if self.columns.is_unique:
-> 2069             return self._get_item_cache(key)
   2070 
   2071         # duplicate columns & possible reduce dimensionality

/usr/local/lib/python3.6/site-packages/pandas/core/generic.py in _get_item_cache(self, item)
   1532         res = cache.get(item)
   1533         if res is None:
-> 1534             values = self._data.get(item)
   1535             res = self._box_item_values(item, values)
   1536             cache[item] = res

/usr/local/lib/python3.6/site-packages/pandas/core/internals.py in get(self, item, fastpath)
   3588 
   3589             if not isnull(item):
-> 3590                 loc = self.items.get_loc(item)
   3591             else:
   3592                 indexer = np.arange(len(self.items))[isnull(self.items)]

/usr/local/lib/python3.6/site-packages/pandas/core/indexes/base.py in get_loc(self, key, method, tolerance)
   2393                 return self._engine.get_loc(key)
   2394             except KeyError:
-> 2395                 return self._engine.get_loc(self._maybe_cast_indexer(key))
   2396 
   2397         indexer = self.get_indexer([key], method=method, tolerance=tolerance)

pandas/_libs/index.pyx in pandas._libs.index.IndexEngine.get_loc (pandas/_libs/index.c:5239)()

pandas/_libs/index.pyx in pandas._libs.index.IndexEngine.get_loc (pandas/_libs/index.c:5085)()

pandas/_libs/hashtable_class_helper.pxi in pandas._libs.hashtable.PyObjectHashTable.get_item (pandas/_libs/hashtable.c:20405)()

pandas/_libs/hashtable_class_helper.pxi in pandas._libs.hashtable.PyObjectHashTable.get_item (pandas/_libs/hashtable.c:20359)()

KeyError: 12

ideas for grouping

I think grouping could still be a lot easier. Here is an example with the Plotly syntax:

import pandas as pd
df = pd.read_csv('http://www.stat.ubc.ca/~jenny/notOcto/STAT545A/examples/gapminder/data/gapminderDataFiveYear.txt', sep='\t')


py.iplot([
            {
                    'x': df[df['year']==year]['gdpPercap'],
                    'y': df[df['year']==year]['lifeExp'],
                    'name': year, 'mode': 'markers',
            } for year in [1952, 1982, 2007]
], filename='cufflinks/scatter-group-by')

with a groupBy and groupVals key that segments by traces, this could be a on-liner with:

df.iplot(x='gdpPercap', y='lifeExp', groupBy='year', groupVals=[1952, 1982, 2007])

where groupVals is pd.unique(df[groupBy]) by default.

This could be extended to subplots really nicely too, where you could swap groupBy with rows and/or cols to split the traces out into multiple subplots.

cufflinks go_offline() and latex

Hi,

When I use go_offline() my latex formula on ipython stay on the left and are not centered. Any idea why ?

Thanks

xerr and yerr key support

Doesn't look like xerr or yerr keys are supported yet. Looking at this example:
http://pandas.pydata.org/pandas-docs/stable/visualization.html#visualization-errorbars

https://plot.ly/python/error-bars/

"Continuous error bars" could be a cool option:
http://help.plot.ly/continuous-filled-error-bars/

Stacked area chart tooltips show cumulative y value instead of trace value

Cufflinks features a stacked area plot to complement the stacked bar chart (which is awesome). However, the tooltips on the stacked are show the cumulative height along the y axis rather than the value of the data represented by that trace at that x value. Cufflinks should implement the stacked area with original values as addressed in the plotly docs as the default option. This would be consistent with the tooltips in other charting tools that support tooltips like Excel and ChartIO. The existing behavior can be retained as an option if it is useful.

Stacked Bar in plotly + cufflinks

Stacked Area in plotly + cufflinks

Stacked Area in Excel

support for logy and logx keys in df.iplot()

http://pandas.pydata.org/pandas-docs/version/0.15.0/visualization.html#scales

df.iplot(x='Date',kind='scatter',logy=True)

Support for grouping in line/scatter plots

Would it make sense to add support for grouping for line and scatter plots (and maybe others)?

If so, it might be nice to use df.groupby('column_A').iplot(...) API for that, as well as for the existing grouping feature available for bar plots.

Violin plot ?

Does anyone think violin plot useful ?
https://plot.ly/python/violin-plot/

Sharing="Secret" Gives "404 Error" in Cufflinks

I'm using Python3 and configured Cufflinks to use cf.set_config_file(sharing="secret", offline=False).

However, when the iframe is embedded in an HTML document, I receive the error:

404 Plot twist! There's nothing here...

But the same plot directly implemented with Python without Cufflinks displays fine in HTML. The plot only displays if I'm logged in to Plotly in the same browser I'm using to view the HTML. I'm thinking this is the behaviour for sharing="private"

Is my cf.set_config_file() incomplete?

go_offline in iPython Notebook Breaks LaTeX Renderings

When I run this in a code cell:
import cufflinks as cf
cf.go_offline()

Then this in a markdown cell:
$$
X=\left{ \begin{array}{rl}
null &\mbox{ if $S+F=0$}
\frac{A}{B+C} &\mbox{ if $S+F>0$}
\end{array} \right.
$$

I end up with this as a result:
$$ SR=\left{ \begin{array}{rl} null &\mbox{ if $S+F=0$} \ \frac{S}{S+F} &\mbox{ if $S+F>0$} \end{array} \right. $$

'Line' Not Allowed in Histogram

I'm following the cufflinks tutorials on plotly word for word and anytime I try to make a histogram, including dataframe scatter matrices, I get the following error:

PlotlyDictKeyError: 'line' is not allowed in 'histogram'

Path To Error: ['line']

Valid attributes for 'histogram' at path [] under parents []:

    ['dx', 'orientation', 'stream', 'zsrc', 'ysrc', 'xbins', 'xsrc',
    'visible', 'marker', 'y0', 'tsrc', 'uid', 'showlegend', 'error_x',
    'error_y', 'rsrc', 'xaxis', 'text', 'bardir', 'type', 'opacity',
    'legendgroup', 'nbinsx', 'nbinsy', 'textsrc', 'histfunc', 'hoverinfo',
    'x0', 'ybins', 'autobiny', 'autobinx', 'name', 'yaxis', 'dy',
    'histnorm', 'r', 't', 'y', 'x', 'z']

I tried reinstalling cufflinks (successfully), to no avail. Any ideas?

This is the code I'm using:

import plotly.plotly as py
import cufflinks as cf
import pandas as pd
import numpy as np

df = pd.DataFrame(np.random.randn(1000, 4), columns=['a', 'b', 'c', 'd'])
df.scatter_matrix(filename='cufflinks/scatter-matrix', world_readable=True)

and...

df = pd.DataFrame({'a': np.random.randn(1000) + 1,
                   'b': np.random.randn(1000),
                   'c': np.random.randn(1000) - 1})
df.iplot(kind='histogram', filename='cufflinks/basic-histogram')

(Sorry if this is an inappropriate place to ask this question, if it is, please direct me to somehwere more appropriate and I'll ask it there.)

secondary y-axis doesn't work with vline

VSpan argument incompatiable with plotly.graph_objs.Layout

Thanks for providing us with an awesome library!

When I try to call

import pandas as pd
import cufflinks as cf
import plotly.graph_objs as go

x.iplot(world_readable=False, filename='test/test',
                          vspan = [{
                            'x0': '2015-01-01', 'x1': '2015-12-31',
                            'color': 'teal', 'fill': True, 'opacity': .4}],
                          layout=go.Layout(title='title',
                            xaxis1=dict(title='date',tickfont=dict(size=18)),
                            yaxis1=dict(title='y',range=[0,100],tickfont=dict(size=18))))

It seems that the vspan argument is ignored. I have tried to use shapes as an argument with go.Layout, however its actually go.Scatter(layout) that can use the shapes parameter, not go.Layout. Any ideas on how to include both a vspan and a layout object?

Cheers,

Jack

Error when making a Histogram.

Hello,

I'm currently running Python 2.7 with Cufflinks 0.6.7 and end up with a PlotlyDictKeyError everytime I make a histogram. Even just running code from [here](https://plot.ly/ipython-notebooks/cufflinks/ gives me the same error.).

Error is:

PlotlyDictKeyError: 'line' is not allowed in 'histogram'

Path To Error: ['line']

Valid attributes for 'histogram' at path [] under parents []:

['dx', 'orientation', 'stream', 'zsrc', 'ysrc', 'xbins', 'xsrc',
'visible', 'marker', 'y0', 'tsrc', 'uid', 'showlegend', 'error_x',
'error_y', 'rsrc', 'xaxis', 'text', 'bardir', 'type', 'opacity',
'legendgroup', 'nbinsx', 'nbinsy', 'textsrc', 'histfunc', 'hoverinfo',
'x0', 'ybins', 'autobiny', 'autobinx', 'name', 'yaxis', 'dy',
'histnorm', 'r', 't', 'y', 'x', 'z']

Run <histogram-object>.help('attribute') on any of the above.
'' is the object at []

log scale

log scale is supported and works well.
The label is given with letter (k, µ,n, ...) but I was wondering is there is a way to use power instead (10-n) ?

Thanks

Unable to import cufflinks; importerror no module named 'plotlytools'

Hi,

When I try to import the cufflinks library into my Python environment I'm getting an ImportError.

My environment:
Windows 8.1 64bit
WinPython 64bit running Python 3.4.3

From what I can see all the files are structured correctly in the proper WinPython environment. Any ideas what could be causing the issue?

secondary y axis title?

I see there is an option to have a secondary y axis here: https://plot.ly/pandas/multiple-axes/
but there does not seem to be a way to add a title for the second axis, only the first one.

sharing key isn't recognized

world_readable has been updated to "sharing":
https://plot.ly/python/privacy/

code error?

[in]
cf.datagen.sinwave(10,.25).iplot(kind='surface',theme='solar',colorscale='brbg',title='Cufflinks - Surface Plot',margin=(0,0,0,0))

[out]

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-10-51640e31aa23> in <module>()
      1 cf.datagen.sinwave(10,.25).iplot(kind='surface',theme='solar',colorscale='brbg',title='Cufflinks - Surface Plot',
----> 2                                  margin=(0,0,0,0))

/usr/local/lib/python2.7/dist-packages/cufflinks/plotlytools.pyc in _iplot(self, data, layout, filename, sharing, kind, title, xTitle, yTitle, zTitle, theme, colors, colorscale, fill, width, dash, mode, symbol, size, barmode, sortbars, bargap, bargroupgap, bins, histnorm, histfunc, orientation, boxpoints, annotations, keys, bestfit, bestfit_colors, mean, mean_colors, categories, x, y, z, text, gridcolor, zerolinecolor, margin, labels, values, secondary_y, subplots, shape, error_x, error_y, error_type, asFrame, asDates, asFigure, asImage, dimensions, asPlot, asUrl, online, **kwargs)
    584         layout=tools.getLayout(theme=theme,xTitle=xTitle,yTitle=yTitle,zTitle=zTitle,title=title,barmode=barmode,
    585                                                                 bargap=bargap,bargroupgap=bargroupgap,annotations=annotations,gridcolor=gridcolor,
--> 586                                 zerolinecolor=zerolinecolor,margin=margin,is3d='3d' in kind,**l_kwargs)
    587 
    588         if not data:

/usr/local/lib/python2.7/dist-packages/cufflinks/tools.pyc in getLayout(theme, title, xTitle, yTitle, zTitle, barmode, bargap, bargroupgap, gridcolor, zerolinecolor, margin, annotations, is3d, **kwargs)
    165                         margin=margin
    166                 else:
--> 167                         margin=dict(list(zip(('l','r','b','t')),margin))
    168                 layout.update(margin=margin)
    169 

TypeError: list() takes at most 1 argument (2 given)

do you mean these?

[in]
margin=(0,0,0,0)
dict(list(zip(('l','r','b','t')),margin))

[out]

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-17-de394516c4a5> in <module>()
      1 margin=(0,0,0,0)
----> 2 dict(list(zip(('l','r','b','t')),margin))

TypeError: list() takes at most 1 argument (2 given)

[in]
margin=(0,0,0,0)
dict(list(zip(('l','r','b','t'),margin)))

[out]

{'b': 0, 'l': 0, 'r': 0, 't': 0}

Can I add hline or vline on candle plot ?

Its ok to add hline in iplot when kind='line' , but fail when kind='candle'

df[-60:].iplot(kind='candle', hline=[2,4])

Could it be possible or how can I add hline on exist candle plot ?

Thanks

Objects as labels raise, as they aren't coerced to strings

If these were coerced with str(obj), they'd be easily JSON-able.

I'm using a pandas Period here as a convenient well known object; this isn't Period-specific.

pd.DataFrame({pd.Period('20150101'): [1,2,3]}).iplot()

TypeError                                 Traceback (most recent call last)
<ipython-input-14-cebfba5820d8> in <module>()
      1 import datetime
----> 2 pd.DataFrame({pd.Period('20150101'): [1,2,3]}).iplot()

/usr/local/lib/python2.7.11/lib/python2.7/site-packages/cufflinks/plotlytools.pyc in _iplot(self, data, layout, filename, sharing, kind, title, xTitle, yTitle, zTitle, theme, colors, colorscale, fill, width, dash, mode, symbol, size, barmode, sortbars, bargap, bargroupgap, bins, histnorm, histfunc, orientation, boxpoints, annotations, keys, bestfit, bestfit_colors, mean, mean_colors, categories, x, y, z, text, gridcolor, zerolinecolor, margin, labels, values, secondary_y, subplots, shape, error_x, error_y, error_type, asFrame, asDates, asFigure, asImage, dimensions, asPlot, asUrl, online, **kwargs)
    869                 return py.plot(figure,sharing=sharing,filename=filename,validate=validate,auto_open=False)
    870         else:
--> 871                 return iplot(figure,sharing=sharing,filename=filename,validate=validate,online=online)
    872 
    873 

/usr/local/lib/python2.7.11/lib/python2.7/site-packages/cufflinks/plotlytools.pyc in iplot(data_or_figure, validate, sharing, filename, online, **kwargs)
    970                 show_link = auth.get_config_file()['offline_show_link']
    971                 link_text = auth.get_config_file()['offline_link_text']
--> 972                 return offline.py_offline.iplot(data_or_figure,show_link=show_link,link_text=link_text)
    973         else:
    974                 if 'layout' in data_or_figure:

/usr/local/lib/python2.7.11/lib/python2.7/site-packages/plotly/offline/offline.pyc in iplot(figure_or_data, show_link, link_text, validate)
    187     plot_html, plotdivid, width, height = _plot_html(
    188         figure_or_data, show_link, link_text, validate,
--> 189         '100%', 525)
    190 
    191     display(HTML(plot_html))

/usr/local/lib/python2.7.11/lib/python2.7/site-packages/plotly/offline/offline.pyc in _plot_html(figure_or_data, show_link, link_text, validate, default_width, default_height)
     95 
     96     plotdivid = uuid.uuid4()
---> 97     jdata = json.dumps(figure.get('data', []), cls=utils.PlotlyJSONEncoder)
     98     jlayout = json.dumps(figure.get('layout', {}), cls=utils.PlotlyJSONEncoder)
     99 

/usr/local/lib/python2.7.11/lib/python2.7/json/__init__.pyc in dumps(obj, skipkeys, ensure_ascii, check_circular, allow_nan, cls, indent, separators, encoding, default, sort_keys, **kw)
    249         check_circular=check_circular, allow_nan=allow_nan, indent=indent,
    250         separators=separators, encoding=encoding, default=default,
--> 251         sort_keys=sort_keys, **kw).encode(obj)
    252 
    253 

/usr/local/lib/python2.7.11/lib/python2.7/site-packages/plotly/utils.pyc in encode(self, o)
    143 
    144         # this will raise errors in a normal-expected way
--> 145         encoded_o = super(PlotlyJSONEncoder, self).encode(o)
    146 
    147         # now:

/usr/local/lib/python2.7.11/lib/python2.7/json/encoder.pyc in encode(self, o)
    205         # exceptions aren't as detailed.  The list call should be roughly
    206         # equivalent to the PySequence_Fast that ''.join() would do.
--> 207         chunks = self.iterencode(o, _one_shot=True)
    208         if not isinstance(chunks, (list, tuple)):
    209             chunks = list(chunks)

/usr/local/lib/python2.7.11/lib/python2.7/json/encoder.pyc in iterencode(self, o, _one_shot)
    268                 self.key_separator, self.item_separator, self.sort_keys,
    269                 self.skipkeys, _one_shot)
--> 270         return _iterencode(o, 0)
    271 
    272 def _make_iterencode(markers, _default, _encoder, _indent, _floatstr,

/usr/local/lib/python2.7.11/lib/python2.7/site-packages/plotly/utils.pyc in default(self, obj)
    209             except NotEncodable:
    210                 pass
--> 211         return json.JSONEncoder.default(self, obj)
    212 
    213     @staticmethod

/usr/local/lib/python2.7.11/lib/python2.7/json/encoder.pyc in default(self, o)
    182 
    183         """
--> 184         raise TypeError(repr(o) + " is not JSON serializable")
    185 
    186     def encode(self, o):

TypeError: Period('2015-01-01', 'D') is not JSON serializable

Update resample

how in pandas will be deprecated

pie chart subplots

pie chart subplots are created slightly differently than cartesion subplots - they use the domain property inside the traces. Here is an example: https://plot.ly/python/pie-charts/#pie-chart-subplots

would be cool if this were supported. my attempt was something along the lines of:

cf.tools.subplots([df.iplot(kind='pie', labels='a', values=c, asFigure=True) for c in ['a', 'b', 'c', 'd']])

`kind='scatter'` doesn't support `text` array

example:

import pandas as pd
df = pd.read_csv('http://www.stat.ubc.ca/~jenny/notOcto/STAT545A/examples/gapminder/data/gapminderDataFiveYear.txt', sep='\t')
df[df.year==2007].iplot(kind='scatter', mode='markers', x='gdpPercap', y='lifeExp', text='country')

In plotly, we support text on bar, scatter (+bubble, mode=markers and lines), scatter3d

dimensions arg not propogated to .getLayout method

Hi @santosjorge,

Thanks for the great library. Since plot.ly has been open sourced, I've been using cufflinks all the time to great fanfare. Kudos.

It seems that the dimensions argument in an .iplot call isn't passed through to the subsequent .getLayout method in generation of a plotly Layout object with defined height and widths. This prevents creating a Plotly visualization when working interactively (e.g., offline in Jupyter notebook). Is this by design / intentional?

Minimal working example, demonstrating the resultant Layout dict repr not containing height or width.

In [1]:  import cufflinks as cf
In [2]: cf.datagen.lines().iplot(asFigure=True, dimensions=(500, 500)).layout

{ 'legend': {'bgcolor': '#F5F6F9', 'font': {'color': '#4D5663'}},
 'paper_bgcolor': '#F5F6F9',
 'plot_bgcolor': '#F5F6F9',
 'titlefont': {'color': '#4D5663'},
 'xaxis1': {'gridcolor': '#E1E5ED',
  'showgrid': True,
  'tickfont': {'color': '#4D5663'},
  'title': '',
  'titlefont': {'color': '#4D5663'},
  'zerolinecolor': '#E1E5ED'},
 'yaxis1': {'gridcolor': '#E1E5ED',
  'showgrid': True,
  'tickfont': {'color': '#4D5663'},
  'title': '',
  'titlefont': {'color': '#4D5663'},
  'zerolinecolor': '#E1E5ED'}}

I have a quick fix for this (if it is indeed an issue), and will submit a PR shortly—thanks! :)

Add support for categories

It looks like categories are not supported.

The following fails with an exception:

from plotly.offline import init_notebook_mode, iplot
import pandas as pd
import cufflinks as cf
cf.go_offline()
init_notebook_mode()

df = pd.DataFrame({'x': [1,2,3,4], 'y': ['A', 'B', 'C', 'D']})
df['y'] = df['y'].astype('category')
df.set_index('y').iplot(kind='bar')
df.set_index('y').columns

And the following fails silently:

df = pd.DataFrame({'x': [1,2,3,4], 'y': ['A', 'B', 'C', 'D']})
df['y'] = df['y'].astype('category')
df.set_index('y').T.iplot(kind='bar')
df.set_index('y').T.columns

Categories in pandas are a great feature and I'd love to see them supported!

python 2 compatibility?

I'm wondering if your library is intended to be Python 2 compatible? Importing it yields a problem with the very first import here:

from . import date_tools

I've added

from __future__ import absolute_import

at the top of the init file and I was at least able to import the library and use cf.datagen.lines().

support for secondary_y in df.iplot()

This would be awesome too. Setting asFigure=True and adding the axis in Plotly's native syntax is a little clunky.
http://pandas.pydata.org/pandas-docs/version/0.15.0/visualization.html#plotting-on-a-secondary-y-axis

Licence file missing

There seems to be no licence file in the repo, is this correct?
In the setup.py file there's a MIT licence indication, so it would probably be best to add the file as well.

Trouble with TA Lib

I just installed this for the first time, and having trouble using it.

The error I'm getting is "no module named talib" when I import cufflinks.

I ran brew install talib. Are there other steps involved? Running pip install TA-Lib also fails on the gcc command (fatal error: 'ta-lib/ta_defs.h' file not found).

How can I get a hold of v0.10 in the meantime? I'd love to take this for a spin.

showing multiple small graphs in a notebook cell output

For example: assume each row can show 3 figures (maybe we can also adjust the number for very big displays), the output could be:

fig0 fig1 fig2
fig3 fig4 fig5
fig6 ...

when we click a figure, it can expand to 100% in width.

This is very important for dense experiment work, such as CNN (Deep learning). We can plot neuron weights for many neurons (a heatmap for each neuron) easily. And we don't lose global views.

We don't use subplot because it needs to determine the numbers first and we need to hold the handle of a panel. (We don't like too much intrusive code in the framework for assist purpose)

Do you think it can be integrated in cufflinks?

Categories support for box plots

Support:

df.iplot(kind='box', x='myX',y='myY',categories='myGroup')

Filename parameter for offline plots?

Can the filename option be passed as the local filename (HTML) to be stored on the user's computer when cufflinks is in offline mode?

MultiIndex support for chart grouping

Looking at this example in particular

http://pandas.pydata.org/pandas-docs/stable/visualization.html#visualization-errorbars

Offline Bubble Chart Using Cufflinks Shows All Bubbles in One Color

Hello,
According to this example:
https://plot.ly/ipython-notebooks/cufflinks/

I should have bubbles with different colors. I have created an issue at the plotly.py github repo, but I think this is a cufflinks issue.

'categories' with kind='bar'

I expected this to work, but I think 'categories' doesn't support bar charts yet.

Trouble with TA-lib

I tried this steps by @Netherdrake

install talib

RUN wget http://prdownloads.sourceforge.net/ta-lib/ta-lib-0.4.0-src.tar.gz
RUN tar -xzf ta-lib-0.4.0-src.tar.gz
WORKDIR /ta-lib
RUN ls
RUN ./configure
RUN make
RUN make install

RUN pip install TA-lib

Still getting this error

I am on python 3.6 mac osx 10.12.5

Javascript error in jupyter on cufflinks Import

I am using datepicker widget given in Ipywidgets example along with plotly, Whenever I import cufflinks for plotting I start to get following javascript error: TypeError: require is not a function

Error on asImage

if there is no img/ folder then there is an error

Requiring 'core' in python 3.4.3

Using newest Ananconda install with python 3.4.3. In notebook !pip install cufflinks works, but importing set_confog_file return erros requiring datetool. This install sucessfully also using !pip install, but importing set_config_file then requires core. Install of core returns: Could not find a version that satisfies the requirement core (from versions: )
No matching distribution found for core

Chart Dimensions config

Is there a way to set a default chart dimension? Currently, the charts default to a size that exceeds my cell output window so all charts have a horizontal scroll bar. I've gotten around it by using dimensions but was wondering if it was possible to have the dimensions set by default.

df = pd.DataFrame({
     'y1': [3, 1, 5], 
     'y2': [30, 50, 20]
}, index=[1, 2, 3])

df.iplot(kind={'y1': 'bar', 'y2': 'scatter'})

df.iplot(kind=['bar', 'scatter'])

this pattern could apply to any of the non-array figure properties, e.g.

df.iplot(mode=['markers', 'text'], marker=[{'color': 'blue}, {'color': 'red'}])

cannot import name date_tools

Have updated plotly and cufflinks, but am getting this weird error I've never seen before.

In [120]: import cufflinks
---------------------------------------------------------------------------
ImportError                               Traceback (most recent call last)
<ipython-input-120-727a8f30b326> in <module>()
----> 1 import cufflinks

/usr/local/lib/python2.7/site-packages/cufflinks/__init__.py in <module>()
      9 from __future__ import absolute_import
     10
---> 11 from . import date_tools
     12 from . import utils
     13 from . import datagen

ImportError: cannot import name date_tools

And cufflinks installed:

Requirement already up-to-date: cufflinks in /usr/local/lib/python2.7/site-packages
Requirement already up-to-date: pandas in /usr/local/lib/python2.7/site-packages (from cufflinks)
Requirement already up-to-date: plotly>=2.0.0 in /usr/local/lib/python2.7/site-packages (from cufflinks)
Requirement already up-to-date: colorlover>=0.2 in /usr/local/lib/python2.7/site-packages (from cufflinks)
Requirement already up-to-date: python-dateutil in /usr/local/lib/python2.7/site-packages (from pandas->cufflinks)
Requirement already up-to-date: numpy>=1.7.0 in /usr/local/lib/python2.7/site-packages (from pandas->cufflinks)
Requirement already up-to-date: pytz>=2011k in /usr/local/lib/python2.7/site-packages (from pandas->cufflinks)
Requirement already up-to-date: six in /Users/chase.schwalbach/Library/Python/2.7/lib/python/site-packages (from plotly>=2.0.0->cufflinks)
Requirement already up-to-date: nbformat>=4.2 in /usr/local/lib/python2.7/site-packages (from plotly>=2.0.0->cufflinks)
Requirement already up-to-date: requests in /usr/local/lib/python2.7/site-packages (from plotly>=2.0.0->cufflinks)
Requirement already up-to-date: decorator>=4.0.6 in /Users/chase.schwalbach/Library/Python/2.7/lib/python/site-packages (from plotly>=2.0.0->cufflinks)
Requirement already up-to-date: jsonschema!=2.5.0,>=2.4 in /usr/local/lib/python2.7/site-packages (from nbformat>=4.2->plotly>=2.0.0->cufflinks)
Requirement already up-to-date: ipython-genutils in /Users/chase.schwalbach/Library/Python/2.7/lib/python/site-packages (from nbformat>=4.2->plotly>=2.0.0->cufflinks)
Requirement already up-to-date: jupyter-core in /usr/local/lib/python2.7/site-packages (from nbformat>=4.2->plotly>=2.0.0->cufflinks)
Requirement already up-to-date: traitlets>=4.1 in /Users/chase.schwalbach/Library/Python/2.7/lib/python/site-packages (from nbformat>=4.2->plotly>=2.0.0->cufflinks)
Requirement already up-to-date: certifi>=2017.4.17 in /usr/local/lib/python2.7/site-packages (from requests->plotly>=2.0.0->cufflinks)
Requirement already up-to-date: chardet<3.1.0,>=3.0.2 in /usr/local/lib/python2.7/site-packages (from requests->plotly>=2.0.0->cufflinks)
Requirement already up-to-date: urllib3<1.22,>=1.21.1 in /usr/local/lib/python2.7/site-packages (from requests->plotly>=2.0.0->cufflinks)
Requirement already up-to-date: idna<2.6,>=2.5 in /usr/local/lib/python2.7/site-packages (from requests->plotly>=2.0.0->cufflinks)
Requirement already up-to-date: functools32; python_version == "2.7" in /usr/local/lib/python2.7/site-packages (from jsonschema!=2.5.0,>=2.4->nbformat>=4.2->plotly>=2.0.0->cufflinks)
Requirement already up-to-date: enum34; python_version == "2.7" in /Users/chase.schwalbach/Library/Python/2.7/lib/python/site-packages (from traitlets>=4.1->nbformat>=4.2->plotly>=2.0.0->cufflinks)

macOS Sierra
python 2.7.13

Annotations location

Accoring to http://nbviewer.jupyter.org/gist/santosjorge/f3b07b2be8094deea8c6

" Annotations can be added to the chart and these are automatically positioned correctly. "

Is it possible to set more attribution of the annotations like multiple-annotations in plotly ?
https://plot.ly/python/text-and-annotations/#multiple-annotations

Because the annotations can't positioned correctly sometimes

ability to set `'index'` as an `'x'` or `'y'`

Example:

df = pd.DataFrame(np.random.randn(1000, 2), columns=['B', 'C']).cumsum()
df.iplot(x='index', y='B')

Note that the plotly son encoder also parses index's, so this would be something like:

py.iplot([{'x': df.index, 'y': df.B}])

santosjorge / cufflinks Goto Github PK

cufflinks's People

Contributors

Stargazers

Watchers

Forkers

cufflinks's Issues

do you mean these?

install talib

Recommend Projects

Recommend Topics

Recommend Org