Skip to content

Plots

tinytopics.plot

plot_loss(losses, figsize=(10, 8), dpi=300, title='Loss curve', color_palette=None, output_file=None)

Plot the loss curve over training epochs.

Parameters:

Name Type Description Default
losses list

List of loss values for each epoch.

required
figsize tuple

Plot size. Default is (10, 8).

(10, 8)
dpi int

Plot resolution. Default is 300.

300
title str

Plot title. Default is "Loss curve".

'Loss curve'
color_palette list or matplotlib colormap

Custom color palette.

None
output_file str

File path to save the plot. If None, displays the plot.

None

plot_structure(L_matrix, normalize_rows=False, figsize=(12, 6), dpi=300, title='Structure Plot', color_palette=None, output_file=None)

Structure plot for visualizing document-topic distributions.

Parameters:

Name Type Description Default
L_matrix ndarray

Document-topic distribution matrix.

required
normalize_rows bool

If True, normalizes each row of L_matrix to sum to 1.

False
figsize tuple

Plot size. Default is (12, 6).

(12, 6)
dpi int

Plot resolution. Default is 300.

300
title str

Plot title.

'Structure Plot'
color_palette list or matplotlib colormap

Custom color palette.

None
output_file str

File path to save the plot. If None, displays the plot.

None

plot_top_terms(F_matrix, n_top_terms=10, term_names=None, figsize=(10, 8), dpi=300, title='Top Terms', color_palette=None, nrows=None, ncols=None, output_file=None)

Plot top terms for each topic in horizontal bar charts.

Parameters:

Name Type Description Default
F_matrix ndarray

Topic-term distribution matrix.

required
n_top_terms int

Number of top terms to display per topic. Default is 10.

10
term_names list

List of term names corresponding to indices.

None
figsize tuple

Plot size. Default is (10, 8).

(10, 8)
dpi int

Plot resolution. Default is 300.

300
title str

Plot title.

'Top Terms'
color_palette list or matplotlib colormap

Custom color palette.

None
nrows int

Number of rows in the subplot grid.

None
ncols int

Number of columns in the subplot grid.

None
output_file str

File path to save the plot. If None, displays the plot.

None