@@ -65,6 +65,71 @@ def compare(
6565 include_legend = True ,
6666 show_symetry = True ,
6767):
68+ """
69+ Generate the MCM
70+
71+ Parameters
72+ ----------
73+
74+ df_results : pandas DataFrame, the csv file containing results
75+ output_dir : str, default = './', the output directory for the results
76+ pdf_savename : str, default = None, the name of the saved file into pdf format, if None, will not
77+ be saved into this format
78+ png_savename : str, default = None, the name of the saved file into png format, if None, will not
79+ be saved into this format
80+ csv_savename : str, default = None, the name of the saved file into csv format, if None, will not
81+ be saved into this format
82+ tex_savename : str, default = None, the name of the saved file into tex format, if None, will not
83+ be saved into this format
84+ used_statistic : str, default = 'Score', one can imagine using error, time, memory etc. instead
85+ save_as_json : bool, default = True, whether or not to save the python analysis dict
86+ into a json file format
87+ plot_1v1_comparisons : bool, default = True, whether or not to plot the 1v1 scatter results
88+ order_WinTieLoss : str, default = 'higher', the order on considering a win or a loss
89+ for a given statistics
90+ include_ProbaWinTieLoss : bool, default = False, condition whether or not include
91+ the bayesian test of [1] for a probabilistic win tie loss count
92+ bayesian_rope : float, default = 0.01, the rope used in case include_ProbaWinTieLoss is True
93+ include_pvalue : bool, default = True, condition whether or not include a pvalue stats
94+ pvalue_test : str, default = 'wilcoxon', the statistical test to produce the pvalue stats.
95+ pvalue_correction : str, default = None, which correction to use for the pvalue significant test
96+ pvalue_threshold : float, default = 0.05, threshold for considering a comparison is significant
97+ or not. If pvalue < pvalue_threshhold -> comparison is significant.
98+ use_mean : str, default = 'mean-difference', the mean used to comapre two comparates.
99+ order_stats : str, default = 'average-statistic', the way to order the used_statistic, default
100+ setup orders by average statistic over all datasets
101+ order_better : str, default = 'decreasing', by which order to sort stats, from best to worse
102+ dataset_column : str, default = 'dataset_name', the name of the datasets column in the csv file
103+ precision : int, default = 4, the number of floating numbers after decimal point
104+ load_analysis : bool, default = False, if the analysis json file is already created before, the
105+ use can choose to load it
106+ row_comparates : list of str, default = None, a list of included row comparates, if None, all of
107+ the comparates in the study are placed in the rows.
108+ col_comparates : list of str, default = None, a list of included col comparates, if None, all of
109+ the comparates in the study are placed in the cols.
110+ excluded_row_comparates : list of str, default = None, a list of excluded row comparates,
111+ if None, no comparate in the study is excluded from the rows
112+ excluded_col_comparates : list of str, default = None, a list of excluded col comparates,
113+ if None, no comparate in the study is excluded from the cols
114+ colormap : str, default = 'coolwarm', the colormap used in matplotlib, if set to None,
115+ no color map is used and the heatmap is turned off, no colors will be seen
116+ fig_size : str ot tuple of two int (example : '7,10'), default = 'auto', the height and width of the figure,
117+ if 'auto', use get_fig_size function in utils.py. Note that the fig size values are in
118+ matplotlib units
119+ font_size : int, default = 17, the font size of text
120+ colorbar_orientation : str, default = 'vertical', in which orientation to show the colorbar
121+ either horizontal or vertical
122+ colorbar_value : str, default = 'mean-difference', the values for which the heat map colors
123+ are based on
124+ win_label : str, default = "r>c", the winning label to be set on the MCM
125+ tie_label : str, default = "r=c", the tie label to be set on the MCM
126+ loss_label : str, default = "r<c", the loss label to be set on the MCM
127+ include_legend : bool, default = True, whether or not to show the legend on the MCM
128+ show_symetry : bool, default = True, whether or not to show the symetrical part of the heatmap
129+
130+
131+ """
132+
68133 if isinstance (df_results , str ):
69134 # assuming its a path
70135 try :
@@ -148,33 +213,34 @@ def get_analysis(
148213 Parameters
149214 ----------
150215
151- df_results : pandas DataFrame, the csv file containing results
152- output_dir : str, default = './', the output directory for the results
153- used_statistic : str, default = 'Score', one can imagine using error, time, memory etc. instead
154- save_as_json : bool, default = True, whether or not to save the python analysis dict
155- into a json file format
156- plot_1v1_comparisons : bool, default = True, whether or not to plot the 1v1 scatter results
157-
158-
159- order_WinTieLoss : str, default = 'higher', the order on considering a win or a loss
160- for a given statistics
161- includeProbaWinTieLoss : bool, default = False, condition whether or not include
216+ df_results : pandas DataFrame, the csv file containing results
217+ output_dir : str, default = './', the output directory for the results
218+ used_statistic : str, default = 'Score', one can imagine using error, time, memory etc. instead
219+ save_as_json : bool, default = True, whether or not to save the python analysis dict
220+ into a json file format
221+ plot_1v1_comparisons : bool, default = True, whether or not to plot the 1v1 scatter results
222+ order_WinTieLoss : str, default = 'higher', the order on considering a win or a loss
223+ for a given statistics
224+ include_ProbaWinTieLoss : bool, default = False, condition whether or not include
162225 the bayesian test of [1] for a probabilistic win tie loss count
163- include_pvalue : bool, default = True, condition whether or not include a pvalue stats
164- pvalue_test : str, default = 'wilcoxon', the statistical test to produce the pvalue stats.
165- pvalue_correction : str, default = None, which correction to use for the pvalue significant test
166- pvalue_threshhold : float, default = 0.05, threshold for considering a comparison is significant
167- or not. If pvalue < pvalue_threshhold -> comparison is significant.
168- used_mean : str, default = 'mean-difference', the mean used to comapre two comparates.
169- bayesian_rope : float, default = 0.01, the rope used in case include_ProbaWinTieLoss is True
170- order_stats : str, default = 'average-statistic', the way to order the used_statistic, default
171- setup orders by average statistic over all datasets
172- order_better : str, default = 'decreasing', by which order to sort stats, from best to worse
173- dataset_column : str, default = 'dataset_name', the name of the datasets column in the csv file
226+ bayesian_rope : float, default = 0.01, the rope used in case include_ProbaWinTieLoss is True
227+ include_pvalue : bool, default = True, condition whether or not include a pvalue stats
228+ pvalue_test : str, default = 'wilcoxon', the statistical test to produce the pvalue stats.
229+ pvalue_correction : str, default = None, which correction to use for the pvalue significant test
230+ pvalue_threshhold : float, default = 0.05, threshold for considering a comparison is significant
231+ or not. If pvalue < pvalue_threshhold -> comparison is significant.
232+ use_mean : str, default = 'mean-difference', the mean used to comapre two comparates.
233+ order_stats : str, default = 'average-statistic', the way to order the used_statistic, default
234+ setup orders by average statistic over all datasets
235+ order_better : str, default = 'decreasing', by which order to sort stats, from best to worse
236+ dataset_column : str, default = 'dataset_name', the name of the datasets column in the csv file
237+ precision : int, default = 4, the number of floating numbers after decimal point
238+ load_analysis : bool, default = False, if the analysis json file is already created before, the
239+ use can choose to load it
174240
175241 Returns
176242 -------
177- analysis : python dictionary containing all extracted comparisons
243+ analysis : python dictionary containing all extracted comparisons
178244
179245 """
180246
@@ -304,23 +370,40 @@ def draw(
304370
305371 Parameters
306372 ----------
307- analysis : python dict, default = None, a python dictionary exrtracted using get_analysis function
308- output_dir : str, default = './', output directory for the results
309- load_analysis : bool, default = True, whether or not to load the analysis json file
310- colormap : str, default = 'coolwarm', the colormap used in matplotlib, if set to None,
373+ analysis : python dict, default = None, a python dictionary exrtracted using get_analysis function
374+ output_dir : str, default = './', output directory for the results
375+ pdf_savename : str, default = None, the name of the saved file into pdf format, if None, will not
376+ be saved into this format
377+ png_savename : str, default = None, the name of the saved file into png format, if None, will not
378+ be saved into this format
379+ csv_savename : str, default = None, the name of the saved file into csv format, if None, will not
380+ be saved into this format
381+ tex_savename : str, default = None, the name of the saved file into tex format, if None, will not
382+ be saved into this format
383+ row_comparates : list of str, default = None, a list of included row comparates, if None, all of
384+ the comparates in the study are placed in the rows.
385+ col_comparates : list of str, default = None, a list of included col comparates, if None, all of
386+ the comparates in the study are placed in the cols.
387+ excluded_row_comparates : list of str, default = None, a list of excluded row comparates,
388+ if None, no comparate in the study is excluded from the rows
389+ excluded_col_comparates : list of str, default = None, a list of excluded col comparates,
390+ if None, no comparate in the study is excluded from the cols
391+ precision : int, default = 4, the number of floating numbers after the decimal point
392+ colormap : str, default = 'coolwarm', the colormap used in matplotlib, if set to None,
311393 no color map is used and the heatmap is turned off, no colors will be seen
312- colorbar_value : str, default = 'mean-difference', the values for which the heat map colors
313- are based on
314- fig_size : str ot tuple of two int, default = 'auto', the height and width of the figure,
315- if 'auto', use get_fig_size function in utils.py
316- font_size : int, default = 17, the font size of text
317- pixels_per_clf_hieght : float, default = 3, the number of pixels used per comparate in height
318- inside each cell of the heatmap
319- pixels_per_clf_width : float, default = 3.5, the number of pixels used per comparate in width
320- inside each cell of the heatmap
321- show_symetry : bool, default = True, whether or not to show the symetrical part of the heatmap
322- colorbar_orientation : str, default = 'vertical', in which orientation to show the colorbar
394+ fig_size : str ot tuple of two int (example : '7,10'), default = 'auto', the height and width of the figure,
395+ if 'auto', use get_fig_size function in utils.py. Note that the fig size values are in
396+ matplotlib units
397+ font_size : int, default = 17, the font size of text
398+ colorbar_orientation : str, default = 'vertical', in which orientation to show the colorbar
323399 either horizontal or vertical
400+ colorbar_value : str, default = 'mean-difference', the values for which the heat map colors
401+ are based on
402+ win_label : str, default = "r>c", the winning label to be set on the MCM
403+ tie_label : str, default = "r=c", the tie label to be set on the MCM
404+ loss_label : str, default = "r<c", the loss label to be set on the MCM
405+ show_symetry : bool, default = True, whether or not to show the symetrical part of the heatmap
406+ include_legend : bool, default = True, whether or not to show the legend on the MCM
324407
325408 """
326409
0 commit comments