FenGraphs module

FenGraphs is a module that regroup functions that plot different plot types specific to the different FENICS boards using matplotlib.pyplot library and Bokeh interactive plotting library.

Functions

Fenics features Histograms

FenGraphs.FenHist()

func FenGraphs.FenHist(FenDataFrame,testTable,columns,selectedBoards,show=True,**kwargs) source

Function that plot an histogram of the provided Fenics boards IDs, separatly or grouped.

Parameters

FenDataFrame (pandas.DataFrame) :

FENICS dataframe with board informations.

testTable (pandas.DataFrame) :

FENICS Fast/Slow DataFrame with boards result values.

columns (tuple) :

Tuple of columns that are being plotted (MainCol,SubCol) for Fast channel. (Gain,MainCol,SubCol) for Slow channel.

selectedBoards (list) :

List of boards IDs to plot as histogram.

show (bool, optional) ,default=True :

Show histograms in the cell output. By default it is true, but for any reasons someone needs to get plots saved without being showed, this can be turned off, a path to save the histogram needs to be provided.

Other parameters

path (string, optional), default=False :

A save directory for the histogram if show is set to False.

separate (bool, optional), default=False :

Plot the given FENICS boards separatly True, or grouped in one histogram False.

sigma (integer, optional), default=False :

within how many sigma deviation values are accepted. Data values within the $mean \pm n\sigma$ are kept on the wanted column/DataFrame.

saveFormat (string, optional), default='jpeg' :

change the output format of the saved histogram : JPEG, PDF, SVG, PNG

       

Fenics features Correlations

FenGraphs.FenCorr()

func FenGraphs.FenCorr(FenDataFrame,testTable,col1,col2,selectedBoards,show=True,**kwargs) source

This function makes a correlation plot of two different FENICS features as a matplotlib.pyplot.plot(feature1,feature2) optimized for FENICS boards.

Parameters

FenDataFrame (pandas.DataFrame) :

FENICS dataframe with board informations.

testTable (pandas.DataFrame) :

FENICS Fast/Slow DataFrame with boards result values.

col1,col2 (tuples) :

Tuples of plotted correlated features. (MainCol,SubCol) for Fast channel. (Gain,MainCol,SubCol) for Slow channel.

selectedBoards (list) :

List of boards IDs to plot as histogram.

show (bool, optional) ,default=True :

Show plot in the cell output. By default it is true, but for any reasons someone needs to get plots saved without being showed, this can be turned off, a path to save the plot needs to be provided.

Other parameters

path (string, optional), default=False :

A save directory for the correlation plot if show is set to False.

separate (bool, optional), default=False :

Plot the given FENICS boards separatly True, or grouped in one correlation plot False.

sigma (integer, optional), default=False :

within how many sigma deviation values are accepted. Data values within the $mean \pm n\sigma$ are kept on the wanted column/DataFrame.

burns (bool,optional), default=False :

Use burntimes as a third axis with a color bar.

burns_filter (integer,optional), default=None :

A burn-time value in burn-hours at when you want a filter.

LR (bool,optional), default=False :

Linear regression of the plotted data, One linear regression for one board plotted.

saveFormat (string, optional), default='jpeg' :

change the output format of the saved correlation plot : JPEG, PDF, SVG, PNG

       

Fenics features burn-evolution

FenGraphs.FenBurnEvol()

func FenGraphs.FenBurnEvol(FenDataFrame,testTable,columns,selectedBoards,show=True,**kwargs) source

This function makes an evolution plot of a selected feature with the corresponding burn-time as a matplotlib.pyplot.plot(burn-time,feature) optimized for FENICS boards.

Parameters

FenDataFrame (pandas.DataFrame) :

FENICS dataframe with board informations.

testTable (pandas.DataFrame) :

FENICS Fast/Slow DataFrame with boards result values.

columns (tuples) :

Feature to plot its evolution with burn-time. (MainCol,SubCol) for Fast channel. (Gain,MainCol,SubCol) for Slow channel.

selectedBoards (list) :

List of boards IDs to plot as histogram.

show (bool, optional) ,default=True :

Show plot in the cell output. By default it is true, but for any reasons someone needs to get plots saved without being showed, this can be turned off, a path to save the plot needs to be provided.

Other parameters

path (string, optional), default=False :

A save directory for the plot if show is set to False.

separate (bool, optional), default=False :

Plot the given FENICS boards separatly True, or grouped in one plot False.

sigma (bool, optional), default=False :

Keep one point per burn-hour, by calculating a $Mean$ of the repeated values in one burn-time.

unique (integer, optional), default=False :

within how many sigma deviation values are accepted. Data values within the $mean \pm n\sigma$ are kept on the wanted column/DataFrame.

burns (bool,optional), default=False :

Use burntimes as a third axis with a color bar.

burns_filter (integer,optional), default=None :

A burn-time value in burn-hours at when you want a filter.

LR (bool,optional), default=False :

Linear regression of the plotted data, One linear regression for one board plotted.

saveFormat (string, optional), default='jpeg' :

change the output format of the saved plot : JPEG, PDF, SVG, PNG

       

Fenics features jitter

FenGraphs.FenSpread()

func FenGraphs.FenSpread(FenDataFrame,testTable,columns,selectedBoards,show=True,**kwargs) source

A jitter plot for FENICS features for each selected board, can be customized with a burning indicator with a color bar.

Parameters

FenDataFrame (pandas.DataFrame) :

FENICS dataframe with board informations.

testTable (pandas.DataFrame) :

FENICS Fast/Slow DataFrame with boards result values.

columns (tuples) :

Feature to plot its evolution with burn-time. (MainCol,SubCol) for Fast channel. (Gain,MainCol,SubCol) for Slow channel.

selectedBoards (list) :

List of boards IDs to plot as histogram.

show (bool, optional) ,default=True :

Show plot in the cell output. By default it is true, but for any reasons someone needs to get plots saved without being showed, this can be turned off, a path to save the plot needs to be provided.

Other parameters

path (string, optional), default=False :

A save directory for the plot if show is set to False.

sigma (bool, optional), default=False :

Keep one point per burn-hour, by calculating a $Mean$ of the repeated values in one burn-time.

burns (bool,optional), default=False :

Use burntimes as a third axis with a color bar.

saveFormat (string, optional), default='jpeg' :

change the output format of the saved correlation plot : JPEG, PDF, SVG, PNG

Classes

Fenics interactive plots framework

FenBokehGrapher

Class FenGraphs.FenBokehGrapher(x,y,*args,**kwargs) source

A class for creating interactive Bokeh graphs.

Attributes

x (list) :

The x-coordinate data for the graph.

y (list) :

The y-coordinate data for the graph. Size needs to correspond to the size of x.

*args (list, optional) :

Additional data lists to customize the data tool tips

colors (list, optional), default= list of zeros :

A third dimension using colors and (optional) sliders for filtering.

labels (list, optional), default= list of "1" :

A label field to identify each x and y in the plot and use (optional) filtering checkboxes.

source (bokeh.source object, optional), default= None :

A bokeh.DataSource object that is generated using bokeh.ColumnDataSource.

plot (bokeh.glyph object, optional), default= None :

A bokeh.glyph object that can be generated using FenBokehGrapher.create_plot().

tooltips (bokeh.tooltips object, optional), default= None :

A bokeh.tooltips object that handles the tooltips and naming of the data showed.

Methods

__init__(x,y,*args,**kwargs) source

FenBokehGrapher initialization of the object structure and attributes using the declared attributes.

create_plot(title,axis_xlabel,axis_ylabel) source

Create a bokeh.glyph object.

create_hist(title,axis_xlabel,axis_ylabel) source

Create a bokeh.histograms for lasso selection interaction.

create_slider(**kwargs) source

Create a bokeh slider object and the HTML DOM.

create_checkboxes(**kwargs) source

Create a bokeh checkboxes object with HTML DOM.

create_radiocheckbox(**kwargs) source

Create a bokeh radio group object with HTML DOM.

selection_callback() source

A javascript callback event when changes are done to filtering tools.

create_combo_filter(**kwargs) source

Create a combo filter with sliders and checkboxes.

filters_tool(filters = [('Combo')]) source

Function that merge declared filters in one filter bokeh layout.

plotter(plotType='scatter',filters=[('Combo')], title='',axis_xlabel='',axis_ylabel='',linear_fit = False,showGraph=True) source

Function that gather all Bokeh graphical objects and filters layout. and display them.

save_as_html(x,y,*args,**kwargs) source

Function that saves the plotter layout in an HTML file.

FenBokehGrapher.create_plot

ClassMethod FenBokehGrapher.create_plot(x,y,*args,**kwargs) source

Create a bokeh.glyph object. This function is called when plottype is set to "Scatter"

Parameters

title (string, optional) , default = '' :

Custom title for the plot

axis_xlabel (string, optional) , default = '' :

Custom label for the x axis.

axis_ylabel (string, optional) , default = '' :

Custom label for the Y axis.

FenBokehGrapher.create_hist

ClassMethod FenBokehGrapher.create_hist(title='',axis_xlabel='',axis_ylabel='') source

Create a double projection histograph object and a JavaScript event listner for Bokeh selection tools and a scatter plot. This function is called when plottype is set to "Histogram"

Parameters

title (string, optional) , default = '' :

Custom title for the plot

axis_xlabel (string, optional) , default = '' :

Custom label for the x axis.

axis_ylabel (string, optional) , default = '' :

Custom label for the Y axis.

FenBokehGrapher.create_slider

ClassMethod FenBokehGrapher.create_slider(**kwargs) source

Create a Slider filtering HTML dom object using Bokeh, and creates a custom JavaScript listner that filter the colors property of FenBokehGraph object. different parameters can be customized using the kwargs parameters.

Called when filters=[('Slider')]

Other parameters

min (integer, optional) , default = min(self.colors) :

default start value at the creation of the slider.

max (integer, optional) , default = max(self.colors) :

default end value at the creation of the slider.

valmin (integer, optional) , default = min(self.colors) :

minimum value of the slider.

valmax (integer, optional) , default = max(self.colors) :

maximum value of the slider.

step (float, optional) , default = 20.0 :

At what stepping values, the cursors of the sliders will move.

title (string, optional) , default = 'Filter' :

Title of the slider.

orientation (string, optional) , default = 'vertical' :

Orientation of the slider (vertical / horizontal).

height (integer, optional) , default = 300 :

The apparent size in the HTML file in pixels.

direction (string, optional) , default = 'rtl' :

Direction of the min-max/max-min values of the slider.

'rtl' : right to left

'ltr' : left to right

FenBokehGrapher.create_checkboxes

ClassMethod FenBokehGrapher.create_checkboxes(**kwargs) source

Create an HTML dom object of checkbox groups using Bokeh, and creates a custom JavaScript listner that filter points by their attibuted label in FenBokehGraph object. different parameters can be customized using the kwargs parameters.

Called when filters=[('Checkbox')]

Other parameters

labels_filter (list, optional) , default = pd.unique(FenGrapher.labels) :

List of labels that is wanted to be filtred (recommended when the number of labels is higher than 8)

It is also recommended to gather similar labels in one big label in a radiocheckbox. Example : Life-Time tests FENICS boards together, and FENICS NIEL/TID together.

FenBokehGrapher.create_radiocheckbox

ClassMethod FenBokehGrapher.create_radiocheckbox(**kwargs) source

Create an HTML dom object of radiocheckboxes using Bokeh, and creates a custom JavaScript listner that filter points by their attibuted label in FenBokehGraph object as groups comapred to the FenBokehGrapher.create_checkboxes(). a default "Select all" "Select none" are available, different parameters can be customized using the kwargs parameters.

Other parameters

radio_categories (list, optional) :

List of dictionaries each category to filter in one dictionary. The dictionary structure should be as : 'name' key for the radiocheck box name, and a 'index' key with a list of labels to filter at same time.

FenBokehGrapher.selection_callback

ClassMethod FenBokehGrapher.selection_callback() source

A callback that recalculate the slope according to the selected points with the selection tools of bokeh. It is called when Linear_regression is True.

Returns

a JavaScript event listner.

FenBokehGrapher.create_combo_filter

ClassMethod FenBokehGrapher.create_combo_filter(**kwargs) source

Create a slider and a checkbox filtering HTML dom with a common filtering JavaScript event listner to handle both filtering simultaneously.

Called when filters=[('Combo')]

Other parameters

min (integer, optional) , default = min(self.colors) :

default start value at the creation of the slider.

max (integer, optional) , default = max(self.colors) :

default end value at the creation of the slider.

valmin (integer, optional) , default = min(self.colors) :

minimum value of the slider.

valmax (integer, optional) , default = max(self.colors) :

maximum value of the slider.

step (float, optional) , default = 20.0 :

At what stepping values, the cursors of the sliders will move.

title (string, optional) , default = 'Filter' :

Title of the slider.

orientation (string, optional) , default = 'vertical' :

Orientation of the slider (vertical / horizontal).

height (integer, optional) , default = 300 :

The apparent size in the HTML file in pixels.

direction (string, optional) , default = 'rtl' :

Direction of the min-max/max-min values of the slider.

'rtl' : right to left

'ltr' : left to right

labels_filter (list, optional) , default = pd.unique(FenGrapher.labels) :

List of labels that is wanted to be filtred

radio_categories (list, optional) :

List of dictionaries each category to filter in one dictionary. The dictionary structure should be as : 'name' key for the radiocheck box name, and a 'index' key with a list of labels to filter at same time.

FenBokehGrapher.filters_tool

ClassMethod FenBokehGrapher.filters_tool(filters = [('Combo')]) source

Function that is responsible of the layout hiarchy of the different filter declared.

Parameters

filters (list, optional) , default = [('Combo')] :

List of tupples of filters to be added, a second tupple level can be added as a dictionary of kwargs to customize the filter.

Example

    ##Custom combo filter
    custom_filter = [('Combo',
    ##Custom checkbox groups
    dict(labels_filter=FEN1_ids,
    ##Custon radio categories kwarg argument
    radio_categories=[dict(name = 'FEN1 LTT',indexes=FEN1_ids),
                      dict(name = 'FEN1 FAILED',indexes=FENICS_fail),]))]

FenBokehGrapher.plotter

ClassMethod FenBokehGrapher.plotter(plotType='scatter',filters=[('Combo')],title='',axis_xlabel='',axis_ylabel='',linear_fit = False, showGraph=True) Is the function that plots and show the plot of FenBokehGrapher object. source

Parameters

plotType (string (case sensitive), optional) , default = 'scatter' :

Plot type, with or without projection histograms.

filters (list of tuples, optional) , default = [('Combo')] :

List of custom filters and their parametrization.

title (string, optional) , default = '' :

Custom title for the plot

axis_xlabel (string, optional) , default = '' :

Custom label for the x axis.

axis_ylabel (string, optional) , default = '' :

Custom label for the Y axis.

linear_fir (bool, optional) , default = False :

Add a linear fit to the resulting plot with a selection event listner using Bokeh selection tools.

showGraph (bool, optional) , default = True :

Option to select whatever you want a resulting graph in the output cell or not.

FenBokehGrapher.save_as_html

ClassMethod FenBokehGrapher.save_as_html(x,y,*args,**kwargs) source

A function that save the plot object and the filtring tools into an HTML file.

Parameters

path (string, optional) , default = '' :

Path to the output directory, by default it saves the plot in the current running environnement main directory.

filename (string, optional) , default = 'default.html' :

Custom output name of the HTML file.

Returns

An HTML file that is openable on any HTML/JS capable environnement, lighter weight than PNG (around 100ko)

   

Fenics t-SNE calcualtions

FenBokehTSNE

Class FenGraphs.FenBokehTSNE(data,**kwargs) source

A class that uses Scikit-learn t-SNE package.

Attributes

data(numpy array or pandas DataFrame):

The input data for t-SNE.

normed (bool, optional):

Specifies whether to normalize the input data. Default is False.

perplexity (integer, optional):

The perplexity parameter of t-SNE. It balances the attention given to local and global aspects of the data. Default is 59.

random_state (integer, optional):

The random seed for t-SNE. It ensures reproducibility of the results. Default is 80.

n_iter (integer, optional):

The maximum number of iterations for t-SNE optimization. Default is 59.

learning_rate (integer, optional):

The learning rate for t-SNE optimization. Higher values lead to faster convergence but may result in less accurate embeddings. Default is 59.

Methods

__init__(x,y,*args,**kwargs) source

FenBokehGrapher initialization of the object structure and attributes using the declared attributes. Do a norming operation if attribute is declared.

tsne() source

Initiate t-SNE calculation. Uses Scikit-learn t-SNE fit and calculaltions, results with two local attributes (self.tsne_x and self.tsne_y).