pygauss.analysis module

class pygauss.analysis.Analysis(folderpath='', server=None, username=None, passwrd=None, folder_obj=None, headers=[])

Bases: object

a class to analyse multiple computations

Parameters:

• folderpath (str) – the folder directory storing the files to be analysed
• server (str) – the name of the server storing the files to be analysed
• username (str) – the username to connect to the server
• passwrd (str) – server password, if not present it will be asked for during initialisation
• headers (list) – the variable categories for each computation

add_basic_properties(props=['basis', 'nbasis', 'optimised', 'conformer'])

adds columns giving info of basic run properties

add_mol_property(name, method, *args, **kwargs)

compute molecule property for all rows and create a data column

Parameters:

• name (str) – what to name the data column
• method (str) – what molecule method to call
• *args –

  arguments to pass to the molecule method

• **kwargs –

  keyword arguments to pass to the molecule method

add_mol_property_subset(name, method, rows=[], filters={}, args=[], kwargs={}, relative_to_rows=[])

compute molecule property for a subset of rows and create/add-to data column

Parameters:

• name (str or list of strings) – name for output column (multiple if method outputs more than one value)
• method (str) – what molecule method to call
• rows (list) – what molecule rows to calculate the property for
• filters (dict) – filter for selecting molecules to calculate the property for
• args (list) – the arguments to pass to the molecule method
• kwargs (dict) – the keyword arguments to pass to the molecule method
• relative_to_rows (list of ints) – compute values relative to the summated value(s) of molecule at the rows listed

add_run(identifiers={}, init_fname=None, opt_fname=None, freq_fname=None, nbo_fname=None, alignto=[], atom_groups={}, add_if_error=False, folder_obj=None)

add single Gaussian run input/outputs

add_runs(headers=[], values=[], init_pattern=None, opt_pattern=None, freq_pattern=None, nbo_pattern=None, add_if_error=False, alignto=[], atom_groups={}, ipython_print=False, folder_obj=None)

add multiple Gaussian run inputs/outputs

calc_kmean_groups(category_column, category_name, groups, columns=[], rows=[], filters={})

calculate the kmeans grouping of rows

The KMeans algorithm clusters data by trying to separate samples in n groups of equal variance, minimizing a criterion known as the inertia or within-cluster sum-of-squares. This algorithm requires the number of clusters to be specified. It scales well to large number of samples and has been used across a large range of application areas in many different fields.

copy()

count_runs()

get number of runs held in analysis

folder

The folder for gaussian runs

get_basic_property(prop, *args, **kwargs)

returns a series of a basic run property or nan if it is not available

Parameters:prop (str) – can be ‘basis’, ‘nbasis’, ‘optimised’, ‘opt_error’ or ‘conformer’

get_folder()

get_ids(variable_names, variable_lists)

return ids of a list of unique computations

get_molecule(row)

get molecule object coresponding to particular row

get_table(rows=[], columns=[], filters={}, precision=4, head=False, mol=False, row_index=[], column_index=[], as_image=False, na_rep='-', font_size=None, width=None, height=None, unconfined=False)

return pandas table of requested data in requested format

rows : integer or list of integers

select row ids

columns : string/integer or list of strings/integers

select column names/positions

filters : dict

filter for rows with certain value(s) in specific columns

precision : int

decimal precision of displayed values

head : int

return only first n rows

mol : bool

include column containing the molecule objects

row_index : string or list of strings

columns to use as new index

column_index : list of strings

srings to place in to higher order column indexs

as_image : bool

output the table as an image (used pygauss.utils.df_to_img)

na_rep : str

how to represent empty (nan) cells (if outputting image)

width, height, unconfined : int, int, bool

args for IPy Image

Returns:df – a table of data Return type:pandas.DataFrame

plot_mol_graphs(gtype='energy', share_plot=False, max_cols=1, padding=(1, 1), tick_rotation=0, rows=[], filters={}, sort_columns=[], info_columns=[], info_incl_id=False, letter_prefix='', start_letter='A', grid=True, sharex=True, sharey=True, legend_size=10, color_scheme='jet', eunits='eV', per_energy=1.0, lbound=None, ubound=None, color_homo='g', color_lumo='r', homo_lumo_lines=True, homo_lumo_values=True, band_gap_value=True)

get a set of data plots for each molecule

Parameters:

• gtype (str) – the type of plot, energy = optimisation energies, freq = frequency analsis, dos = Densty of States,
• share_plot (bool) – whether to plot all data on the same or separate axes
• max_cols (int) – maximum columns on plots (share_plot=False only)
• padding (tuple) – padding between images (horizontally, vertically)
• tick_rotation (int) – rotation of x-axis labels
• rows (int or list) – index for the row of each molecule to plot (all plotted if empty)
• filters (dict) – {columns:values} to filter by
• sort_columns (list of str) – columns to sort by
• info_columns (list of str) – columns to use as info in caption
• info_incl_id (bool) – include molecule id number in labels
• letter_prefix (str) – prefix for labelling subplots (share_plot=False only)
• start_letter (str) – starting (capital) letter for labelling subplots (share_plot=False only)
• grid (bool) – whether to include a grid in the axes
• sharex (bool) – whether to align x-axes (share_plot=False only)
• sharey (bool) – whether to align y-axes (share_plot=False only)
• legend_size (int) – the font size (in pts) for the legend
• color_scheme (str) – the scheme to use for each molecule (share_plot=True only) according to http://matplotlib.org/examples/color/colormaps_reference.html
• eunits (str) – the units of energy to use
• per_energy (float) – energy interval to group states by (DoS only)
• lbound (float) – lower bound energy (DoS only)
• ubound (float) – upper bound energy (DoS only)
• color_homo (matplotlib.colors) – color of homo in matplotlib format
• color_lumo (matplotlib.colors) – color of lumo in matplotlib.colors
• homo_lumo_lines (bool) – draw lines at HOMO and LUMO energies
• homo_lumo_values (bool) – annotate HOMO and LUMO lines with exact energy values
• band_gap_value (bool) – annotate inbetween HOMO and LUMO lines with band gap value

Returns:

• data (matplotlib.figure.Figure) – plotted frequency data
• caption (str) – A caption describing each subplot, given info_columns

plot_mol_images(mtype='optimised', max_cols=1, padding=(1, 1), sort_columns=[], info_columns=[], info_incl_id=False, label_size=20, letter_prefix='', start_letter='A', rows=[], filters={}, align_to=[], rotations=[[0.0, 0.0, 0.0]], gbonds=True, represent='ball_stick', zoom=1.0, width=500, height=500, axis_length=0, relative=False, minval=-1, maxval=1, highlight=[], frame_on=False, eunits='kJmol-1', sopt_min_energy=20.0, sopt_cutoff_energy=0.0, atom_groups=[], alpha=0.5, transparent=False, hbondwidth=5, no_hbonds=False)

show molecules in matplotlib table of axes

Parameters:

• mtype – ‘initial’, ‘optimised’, ‘nbo’, ‘highlight’, ‘highlight-initial’, ‘sopt’ or ‘hbond’
• max_cols (int) – maximum columns in plot
• padding (tuple) – padding between images (horizontally, vertically)
• sort_columns (list of str) – columns to sort by
• info_columns (list of str) – columns to use as info in caption
• info_incl_id (bool) – include molecule id number in caption
• label_size (int) – subplot label size (pts)
• letter_prefix (str) – prefix for labelling subplots
• start_letter (str) – starting (capital) letter for labelling subplots
• rows (int or list) – index for the row of each molecule to plot (all plotted if empty)
• filters (dict) – {columns:values} to filter by
• align_to ([int, int, int]) – align geometries to the plane containing these atoms
• rotations (list of [float, float, float]) – for each rotation set [x,y,z] an image will be produced
• gbonds (bool) – guess bonds between atoms (via distance)
• represent (str) – representation of molecule (‘none’, ‘wire’, ‘vdw’ or ‘ball_stick’)
• zoom (float) – zoom level of images
• width (int) – width of original images
• height (int) – height of original images (although width takes precedent)
• axis_length (float) – length of x,y,z axes in negative and positive directions
• relative (bool) – coloring of nbo atoms scaled to min/max values in atom set (for nbo mtype)
• minval (float) – coloring of nbo atoms scaled to absolute min (for nbo mtype)
• maxval (float) – coloring of nbo atoms scaled to absolute max (for nbo mtype)
• highlight (list of lists) – atom indxes to highlight (for highlight mtype)
• eunits (str) – the units of energy to return (for sopt/hbond mtype)
• sopt_min_energy (float) – minimum energy to show (for sopt/hbond mtype)
• sopt_cutoff_energy (float) – energy below which bonds will be dashed (for sopt mtype)
• alpha (float) – alpha color value of geometry (for sopt/hbond mtypes)
• transparent (bool) – whether atoms should be transparent (for sopt/hbond mtypes)
• hbondwidth (float) – width of lines depicting interaction (for hbond mtypes)
• atom_groups ([list or str, list or str]) – restrict interactions to between two lists (or identifiers) of atom indexes (for sopt/hbond mtypes)
• no_hbonds (bool) – whether to ignore H-Bonds in the calculation (for sopt only)
• frame_on (bool) – whether to show frame around each image

Returns:

• fig (matplotlib.figure.Figure) – A figure containing subplots for each molecule image
• caption (str) – A caption describing each subplot, given info_columns

plot_radviz_comparison(category_column, columns=[], rows=[], filters={}, point_size=30, **kwargs)

return plot axis of radviz graph

RadViz is a way of visualizing multi-variate data. It is based on a simple spring tension minimization algorithm. Basically you set up a bunch of points in a plane. In our case they are equally spaced on a unit circle. Each point represents a single attribute. You then pretend that each sample in the data set is attached to each of these points by a spring, the stiffness of which is proportional to the numerical value of that attribute (they are normalized to unit interval). The point in the plane, where our sample settles to (where the forces acting on our sample are at an equilibrium) is where a dot representing our sample will be drawn. Depending on which class that sample belongs it will be colored differently.

remove_columns(columns)

remove_non_conformers(cutoff=0.0)

removes runs with negative frequencies

remove_non_optimised()

removes runs that were not optimised

remove_rows(rows)

remove one or more rows of molecules

Parameters:rows (int or list of ints:) – the rows to remove

set_folder(folderpath='', server=None, username=None, passwrd=None)

yield_mol_images(rows=[], filters={}, mtype='optimised', sort_columns=[], align_to=[], rotations=[[0.0, 0.0, 0.0]], gbonds=True, represent='ball_stick', zoom=1.0, width=300, height=300, axis_length=0, relative=False, minval=-1, maxval=1, highlight=[], active=False, sopt_min_energy=20.0, sopt_cutoff_energy=0.0, atom_groups=[], alpha=0.5, transparent=False, hbondwidth=5, eunits='kJmol-1', no_hbonds=False, ipyimg=True)

yields molecules

Parameters:

• mtype – ‘initial’, ‘optimised’, ‘nbo’, ‘highlight’, ‘highlight-initial’, ‘sopt’ or ‘hbond’
• info_columns (list of str) – columns to use as info in caption
• max_cols (int) – maximum columns in plot
• label_size (int) – subplot label size (pts)
• start_letter (str) – starting (capital) letter for labelling subplots
• save_fname (str) – name of file, if you wish to save the plot to file
• rows (int or list) – index for the row of each molecule to plot (all plotted if empty)
• filters (dict) – {columns:values} to filter by
• sort_columns (list of str) – columns to sort by
• align_to ([int, int, int]) – align geometries to the plane containing these atoms
• rotations (list of [float, float, float]) – for each rotation set [x,y,z] an image will be produced
• gbonds (bool) – guess bonds between atoms (via distance)
• represent (str) – representation of molecule (‘none’, ‘wire’, ‘vdw’ or ‘ball_stick’)
• zoom (float) – zoom level of images
• width (int) – width of original images
• height (int) – height of original images (although width takes precedent)
• axis_length (float) – length of x,y,z axes in negative and positive directions
• relative (bool) – coloring of nbo atoms scaled to min/max values in atom set (for nbo mtype)
• minval (float) – coloring of nbo atoms scaled to absolute min (for nbo mtype)
• maxval (float) – coloring of nbo atoms scaled to absolute max (for nbo mtype)
• highlight (list of lists) – atom indxes to highlight (for highlight mtype)
• eunits (str) – the units of energy to return (for sopt/hbond mtype)
• sopt_min_energy (float) – minimum energy to show (for sopt/hbond mtype)
• sopt_cutoff_energy (float) – energy below which bonds will be dashed (for sopt mtype)
• alpha (float) – alpha color value of geometry (for highlight/sopt/hbond mtypes)
• transparent (bool) – whether atoms should be transparent (for highlight/sopt/hbond mtypes)
• hbondwidth (float) – width of lines depicting interaction (for hbond mtypes)
• atom_groups ([list or str, list or str]) – restrict interactions to between two lists (or identifiers) of atom indexes (for sopt/hbond mtypes)
• no_hbonds (bool) – whether to ignore H-Bonds in the calculation
• ipyimg (bool) – whether to return an IPython image, PIL image otherwise
• Yields –
• ------- –
• indx (int) – the row index of the molecule
• mol (IPython.display.Image or PIL.Image) – an image of the molecule in the format specified by ipyimg

pygauss.analysis.unpack_and_make_molecule(val_dict)