Skip to content

Latest commit

 

History

History
66 lines (55 loc) · 2.55 KB

Examples.md

File metadata and controls

66 lines (55 loc) · 2.55 KB
Raw

FAF (Frequently Asked Functions)

Get Overview over Corpus

You can get a Dataframe of the Corpus documents and source metadata:

import pandas as pd
import monodikit
corpus = monodikit.Corpus(dir)
document_meta = pd.DataFrame.from_records(corpus.document_metadata)
source_meta = pd.DataFrame.from_records(corpus.source_metadata)

So you can list all document metadata of genre "Tropus" or you can get all liturgical play (which are groups of documents that are identified by the liturgical_play_id)

# Show metadata of all tropes
print(document_meta.loc[document_meta["genre"] == "Tropus"]) 

# Show IDs of all liturgical plays (for example to build a sub corpora out of one or all of them)
print(document_meta.loc[document_meta["liturgical_play_id"] != None].value_counts())

Export first document in different formats

corpus = monodikit.Corpus(dir)
mei = corpus.documents[0].mei
volpiano = corpus.documents[0].volpiano
text_syllables = corpus.documents[0].syllable_text

Search in Corpus

With search_in_window you can search for a sequence of pitches or intervals within a certain window size. It is a complete match if len(sequence) == window size. The following code will give you every segment of six pitches that include [D4, E4, F4, E4, F4] in the same order.

results = monodikit.Search.search_in_window(corpus, ["D4", "E4", "F4", "E4", "F4"], 6)

You can do the same with intervals, just provide the information in the "preprocess" argument.

results = monodikit.Search.search_in_window(corpus, [1,1,-1,-1,0,0], 6,  preprocess="intervals")

You can now create a html file with the results:

with open("result.html", "w") as f:
    f.write(monodikit.Search.visualize_html(results))
image

or view it directly in a jupyter notebook with:

from IPython.display import HTML
HTML(monodikit.Search.visualize_html(results))

Multiple Sequence Alignment with MAFFT

You can also align the found chants (you have to install MAFFT locally)

mafft_input = monodikit.Search.to_mafft_input(results[0:10], context=2) # Align the first 10 search results
align = monodikit.Synopsis.run_mafft(mafft_input)
html = monodikit.Synopsis.visualize_alignments(align, highlight=[segment["ids"] for result in results for segment in result["segments"]], offset=2) # visualize and highligh notes that were in the search result window.