Plots of neural sixgram models

Path	langsim.modules.local_lm.plot
Executable	yes

Produces various plots to help with analysing the results of training a neural_sixgram model.

Note that this used to be designed to support other model types, but I’m now cleaning up and only supporting neural_sixgram2.

Inputs

Name	Type(s)
model	KerasModelBuilderClass
vocabs	list of Dictionary
corpora	list of TarredCorpus<IntegerListsDocumentType>
frequencies	list of NumpyArray

Outputs

Name	Type(s)
output	PimlicoDatatype

Options

Name	Description	Type
distance	Distance metric to use	‘eucl’, ‘dot’, ‘cos’, ‘man’ or ‘sig_kern’
num_pairs	Number of most frequent character pairs to show on the chart (passed through the composition function to get their representation)	int
min_token_prop	Minimum frequency, as a proportion of tokens, that a character in the vocabulary must have to be shown in the charts	float
lang_names	(required) Comma-separated list of language IDs to use in output	comma-separated list of strings