Notes

Show imports Hide imports

import os

import dimcat as dc
import ms3
import pandas as pd
import plotly.express as px
from dimcat import filters, plotting

import utils

pd.set_option("display.max_rows", 1000)
pd.set_option("display.max_columns", 500)

Show source Hide source

RESULTS_PATH = os.path.abspath(os.path.join(utils.OUTPUT_FOLDER, "notes_stats"))
os.makedirs(RESULTS_PATH, exist_ok=True)


def make_output_path(
    filename: str,
    extension=None,
    path=RESULTS_PATH,
) -> str:
    return utils.make_output_path(filename=filename, extension=extension, path=path)


def save_figure_as(
    fig, filename, formats=("png", "pdf"), directory=RESULTS_PATH, **kwargs
):
    if formats is not None:
        for fmt in formats:
            plotting.write_image(fig, filename, directory, format=fmt, **kwargs)
    else:
        plotting.write_image(fig, filename, directory, **kwargs)

Loading data

Show source Hide source

D = utils.get_dataset("wf_bach_sonatas", corpus_release="v2.3")
package = D.inputs.get_package()
package_info = package._package.custom
git_tag = package_info.get("git_tag")
utils.print_heading("Data and software versions")
print("Wilhelm Friedemann Bach – Piano Sonatas version v2.3")
print(f"Datapackage '{package.package_name}' @ {git_tag}")
print(f"dimcat version {dc.__version__}\n")
D

Data and software versions
--------------------------

Wilhelm Friedemann Bach – Piano Sonatas version v2.3
Datapackage 'wf_bach_sonatas' @ v2.3
dimcat version 3.4.0

Dataset
=======
{'inputs': {'basepath': None,
            'packages': {'wf_bach_sonatas': ["'wf_bach_sonatas.measures' (MuseScoreFacetName.MuseScoreMeasures)",
                                             "'wf_bach_sonatas.notes' (MuseScoreFacetName.MuseScoreNotes)",
                                             "'wf_bach_sonatas.expanded' (MuseScoreFacetName.MuseScoreHarmonies)",
                                             "'wf_bach_sonatas.chords' (MuseScoreFacetName.MuseScoreChords)",
                                             "'wf_bach_sonatas.metadata' (FeatureName.Metadata)"]}},
 'outputs': {'basepath': None, 'packages': {}},
 'pipeline': []}

Metadata

filtered_D = filters.HasHarmonyLabelsFilter(keep_values=[True]).process(D)

all_metadata = filtered_D.get_metadata()
all_metadata.reset_index(level=1).groupby(level=0).nth(0).iloc[:, :20]

	piece	TimeSig	KeySig	last_mc	last_mn	length_qb	last_mc_unfolded	last_mn_unfolded	length_qb_unfolded	volta_mcs	all_notes_qb	n_onsets	n_onset_positions	guitar_chord_count	form_label_count	label_count	annotated_key	harmony_version	annotators	reviewers
corpus
wf_bach_sonatas	F001_n08a	{1: '4/4'}	{1: 0}	63	63	252.0	126	126	504.0	()	602.75	1186	727	0	0	204	C	2.3.0	Christos Giannopoulos (1.0.0), Davor Krkljus (...	DK, AN

chronological_order = utils.chronological_corpus_order(all_metadata)
corpus_colors = dict(zip(chronological_order, utils.CORPUS_COLOR_SCALE))

notes_feature = filtered_D.get_feature("notes")
all_notes = notes_feature.df
print(f"{len(all_notes.index)} notes over {len(all_notes.groupby(level=[0,1]))} files.")
all_notes.head()

8925 notes over 9 files.

			mc	mn	quarterbeats	quarterbeats_all_endings	duration_qb	duration	mc_onset	mn_onset	timesig	staff	voice	volta	chord_id	gracenote	midi	name	nominal_duration	octave	scalar	tied	tpc_name	tpc
corpus	piece	i
wf_bach_sonatas	F001_n08a	0	1	1	0	0	1.00	1/4	0	0	4/4	2	1	<NA>	15	<NA>	36	C2	1/4	2	1	<NA>	C	0
		1	1	1	0	0	1.00	1/4	0	0	4/4	2	1	<NA>	15	<NA>	40	E2	1/4	2	1	<NA>	E	4
		2	1	1	0	0	1.00	1/4	0	0	4/4	2	1	<NA>	15	<NA>	43	G2	1/4	2	1	<NA>	G	1
		3	1	1	0	0	1.00	1/4	0	0	4/4	2	1	<NA>	15	<NA>	48	C3	1/4	3	1	<NA>	C	0
		4	1	1	1/4	1/4	0.25	1/16	1/16	1/16	4/4	1	1	<NA>	0	<NA>	72	C5	1/16	5	1	<NA>	C	0

def weight_notes(nl, group_col="midi", precise=True):
    summed_durations = nl.groupby(group_col).duration_qb.sum()
    shortest_duration = summed_durations[summed_durations > 0].min()
    summed_durations /= shortest_duration  # normalize such that the shortest duration results in 1 occurrence
    if not precise:
        # This simple trick reduces compute time but also precision:
        # The rationale is to have the smallest value be slightly larger than 0.5 because
        # if it was exactly 0.5 it would be rounded down by repeat_notes_according_to_weights()
        summed_durations /= 1.9999999
    return repeat_notes_according_to_weights(summed_durations)


def repeat_notes_according_to_weights(weights):
    try:
        counts = weights.round().astype(int)
    except Exception:
        return pd.Series(dtype=int)
    counts_reflecting_weights = []
    for pitch, count in counts.items():
        counts_reflecting_weights.extend([pitch] * count)
    return pd.Series(counts_reflecting_weights)

Ambitus

corpus_names = {
    corp: utils.get_corpus_display_name(corp) for corp in chronological_order
}
chronological_corpus_names = list(corpus_names.values())
corpus_name_colors = {
    corpus_names[corp]: color for corp, color in corpus_colors.items()
}
all_notes["corpus_name"] = all_notes.index.get_level_values(0).map(corpus_names)

grouped_notes = all_notes.groupby("corpus_name")
weighted_midi = pd.concat(
    [weight_notes(nl, "midi", precise=False) for _, nl in grouped_notes],
    keys=grouped_notes.groups.keys(),
).reset_index(level=0)
weighted_midi.columns = ["dataset", "midi"]
weighted_midi

	dataset	midi
0	WF Bach Sonatas	31
1	WF Bach Sonatas	33
2	WF Bach Sonatas	33
3	WF Bach Sonatas	35
4	WF Bach Sonatas	36
...	...	...
4433	WF Bach Sonatas	87
4434	WF Bach Sonatas	88
4435	WF Bach Sonatas	88
4436	WF Bach Sonatas	89
4437	WF Bach Sonatas	89

4438 rows × 2 columns

# fig = px.violin(weighted_midi,
#                 x='dataset',
#                 y='midi',
#                 color='dataset',
#                 title="Corpus-wise distribution over registers (ambitus)",
#                 box=True,
#                 labels=dict(
#                     dataset='',
#                     midi='distribution of pitches by duration'
#                 ),
#                 category_orders=dict(dataset=chronological_corpus_names),
#                 color_discrete_map=corpus_name_colors,
#                 width=1000, height=600,
#                )
# fig.update_traces(spanmode='hard') # do not extend beyond outliers
# fig.update_layout(**utils.STD_LAYOUT,
#                  showlegend=False)
# fig.update_yaxes(
#     tickmode= 'array',
#     tickvals= [12, 24, 36, 48, 60, 72, 84, 96],
#     ticktext = ["C0", "C1", "C2", "C3", "C4", "C5", "C6", "C7"],
# )
# fig.update_xaxes(tickangle=45)
# save_figure_as(fig, "ambitus_corpuswise_violins")
# fig.show()

Tonal Pitch Classes (TPC)

weighted_tpc = pd.concat(
    [weight_notes(nl, "tpc") for _, nl in grouped_notes],
    keys=grouped_notes.groups.keys(),
).reset_index(level=0)
weighted_tpc.columns = ["dataset", "tpc"]
weighted_tpc

	dataset	tpc
0	WF Bach Sonatas	-4
1	WF Bach Sonatas	-4
2	WF Bach Sonatas	-4
3	WF Bach Sonatas	-4
4	WF Bach Sonatas	-4
...	...	...
13307	WF Bach Sonatas	12
13308	WF Bach Sonatas	12
13309	WF Bach Sonatas	12
13310	WF Bach Sonatas	12
13311	WF Bach Sonatas	13

13312 rows × 2 columns

As violin plot

# fig = px.violin(weighted_tpc,
#                 x='dataset',
#                 y='tpc',
#                 color='dataset',
#                 title="Corpus-wise distribution over line of fifths (tonal pitch classes)",
#                 box=True,
#                 labels=dict(
#                     dataset='',
#                     tpc='distribution of tonal pitch classes by duration'
#                 ),
#                 category_orders=dict(dataset=chronological_corpus_names),
#                 color_discrete_map=corpus_name_colors,
#                 width=1000,
#                 height=600,
#                )
# fig.update_traces(spanmode='hard') # do not extend beyond outliers
# fig.update_layout(**utils.STD_LAYOUT,
#                  showlegend=False)
# fig.update_yaxes(
#     tickmode= 'array',
#     tickvals= [-12, -9, -6, -3, 0, 3, 6, 9, 12, 15, 18],
#     ticktext = ["Dbb", "Bbb", "Gb", "Eb", "C", "A", "F#", "D#", "B#", "G##", "E##"],
#     zerolinecolor='grey',
#     zeroline=True
# )
# fig.update_xaxes(tickangle=45)
# save_figure_as(fig, "pitch_class_distributions_corpuswise_violins")
# fig.show()

(all_notes)

			mc	mn	quarterbeats	quarterbeats_all_endings	duration_qb	duration	mc_onset	mn_onset	timesig	staff	voice	volta	chord_id	gracenote	midi	name	nominal_duration	octave	scalar	tied	tpc_name	tpc	corpus_name
corpus	piece	i
wf_bach_sonatas	F001_n08a	0	1	1	0	0	1.000000	1/4	0	0	4/4	2	1	<NA>	15	<NA>	36	C2	1/4	2	1	<NA>	C	0	WF Bach Sonatas
		1	1	1	0	0	1.000000	1/4	0	0	4/4	2	1	<NA>	15	<NA>	40	E2	1/4	2	1	<NA>	E	4	WF Bach Sonatas
		2	1	1	0	0	1.000000	1/4	0	0	4/4	2	1	<NA>	15	<NA>	43	G2	1/4	2	1	<NA>	G	1	WF Bach Sonatas
		3	1	1	0	0	1.000000	1/4	0	0	4/4	2	1	<NA>	15	<NA>	48	C3	1/4	3	1	<NA>	C	0	WF Bach Sonatas
		4	1	1	1/4	1/4	0.250000	1/16	1/16	1/16	4/4	1	1	<NA>	0	<NA>	72	C5	1/16	5	1	<NA>	C	0	WF Bach Sonatas
	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
	F003_n04c	1672	110	108	1292/3	1292/3	0.333333	1/12	5/12	5/12	2/2	2	1	<NA>	1669	<NA>	57	A3	1/8	3	2/3	<NA>	A	3	WF Bach Sonatas
		1673	110	108	1292/3	1292/3	0.333333	1/12	5/12	5/12	2/2	1	1	<NA>	1662	<NA>	61	C#4	1/8	4	2/3	<NA>	C#	7	WF Bach Sonatas
		1674	110	108	431	431	1.000000	1/4	1/2	1/2	2/2	2	2	<NA>	1672	<NA>	50	D3	1/4	3	1	-1	D	2	WF Bach Sonatas
		1675	110	108	431	431	1.000000	1/4	1/2	1/2	2/2	2	1	<NA>	1670	<NA>	54	F#3	1/4	3	1	<NA>	F#	6	WF Bach Sonatas
		1676	110	108	431	431	1.000000	1/4	1/2	1/2	2/2	1	1	<NA>	1663	<NA>	62	D4	1/4	4	1	<NA>	D	2	WF Bach Sonatas

8925 rows × 23 columns

width = 1400
height = 800

weighted_pitch_values = pd.concat(
    [
        weighted_midi.rename(columns={"midi": "value"}),
        weighted_tpc.rename(columns={"tpc": "value"}),
    ],
    keys=["MIDI pitch", "Tonal pitch class"],
    names=["distribution"],
).reset_index(level=[0, 1])

fig = plotting.make_violin_plot(
    weighted_pitch_values,
    x_col="dataset",
    y_col="value",
    color="dataset",
    facet_row="distribution",
    box=True,
    labels=dict(dataset="", tpc="distribution of tonal pitch classes by duration"),
    category_orders=dict(dataset=chronological_corpus_names),
    # color_discrete_map=corpus_name_colors,
    color_discrete_sequence=px.colors.qualitative.Dark24,
    traces_settings=dict(
        spanmode="hard",
        width=1,
        # scalemode='width'
    ),
    layout=dict(
        showlegend=False,
        margin=dict(
            t=0,
            b=0,
            l=0,
            r=0,
        ),
    ),
    x_axis=dict(
        # tickangle=45,
        tickfont_size=15
    ),
    y_axis=dict(
        tickmode="array",
        tickvals=[-12, -9, -6, -3, 0, 3, 6, 9, 12, 15, 24, 36, 48, 60, 72, 84, 96],
        ticktext=[
            "Dbb",
            "Bbb",
            "Gb",
            "Eb",
            "C",
            "A",
            "F#",
            "D#",
            "B#",
            "G##",
            "C1",
            "C2",
            "C3",
            "C4",
            "C5",
            "C6",
            "C7",
        ],
        zerolinecolor="grey",
        zeroline=True,
    ),
    width=width,
    height=height,
)
utils.realign_subplot_axes(fig, y_axes=dict(title_text=""))
save_figure_as(fig, "notes_violin", width=width, height=height)
fig

fig = plotting.make_box_plot(
    weighted_pitch_values,
    x_col="dataset",
    y_col="value",
    color="dataset",
    facet_row="distribution",
    # box=True,
    labels=dict(dataset="", tpc="distribution of tonal pitch classes by duration"),
    category_orders=dict(dataset=chronological_corpus_names),
    # color_discrete_map=corpus_name_colors,
    color_discrete_sequence=px.colors.qualitative.Light24,
    # traces_settings=dict(spanmode='hard'),
    layout=dict(showlegend=False, margin=dict(t=0)),
    x_axis=dict(tickangle=45, tickfont_size=15),
    y_axis=dict(
        tickmode="array",
        tickvals=[-12, -9, -6, -3, 0, 3, 6, 9, 12, 15, 24, 36, 48, 60, 72, 84, 96],
        ticktext=[
            "Dbb",
            "Bbb",
            "Gb",
            "Eb",
            "C",
            "A",
            "F#",
            "D#",
            "B#",
            "G##",
            "C1",
            "C2",
            "C3",
            "C4",
            "C5",
            "C6",
            "C7",
        ],
        zerolinecolor="grey",
        zeroline=True,
    ),
    width=width,
    height=height,
)
utils.realign_subplot_axes(fig, y_axes=True)
save_figure_as(fig, "notes_box", width=width, height=height)
fig

As bar plots

bar_data = all_notes.groupby("tpc").duration_qb.sum().reset_index()
x_values = list(range(bar_data.tpc.min(), bar_data.tpc.max() + 1))
x_names = ms3.fifths2name(x_values)
fig = px.bar(
    bar_data,
    x="tpc",
    y="duration_qb",
    labels=dict(tpc="Named pitch class", duration_qb="Duration in quarter notes"),
    color_discrete_sequence=utils.CORPUS_COLOR_SCALE,
    width=1000,
    height=300,
)
fig.update_layout(**utils.STD_LAYOUT)
fig.update_xaxes(
    zerolinecolor="grey",
    tickmode="array",
    tickvals=x_values,
    ticktext=x_names,
    dtick=1,
    ticks="outside",
    tickcolor="black",
    minor=dict(dtick=6, gridcolor="grey", showgrid=True),
)
save_figure_as(fig, "pitch_class_distribution_absolute_bars")
fig.show()

scatter_data = all_notes.groupby(["corpus_name", "tpc"]).duration_qb.sum().reset_index()
fig = px.bar(
    scatter_data,
    x="tpc",
    y="duration_qb",
    color="corpus_name",
    labels=dict(
        duration_qb="duration",
        tpc="named pitch class",
    ),
    category_orders=dict(dataset=chronological_corpus_names),
    color_discrete_map=corpus_name_colors,
    width=1000,
    height=500,
)
fig.update_layout(**utils.STD_LAYOUT)
fig.update_xaxes(
    zerolinecolor="grey",
    tickmode="array",
    tickvals=x_values,
    ticktext=x_names,
    dtick=1,
    ticks="outside",
    tickcolor="black",
    minor=dict(dtick=6, gridcolor="grey", showgrid=True),
)
save_figure_as(fig, "pitch_class_distribution_corpuswise_absolute_bars")
fig.show()

As scatter plots

fig = px.scatter(
    scatter_data,
    x="tpc",
    y="duration_qb",
    color="corpus_name",
    labels=dict(
        duration_qb="duration",
        tpc="named pitch class",
    ),
    category_orders=dict(dataset=chronological_corpus_names),
    color_discrete_map=corpus_name_colors,
    facet_col="corpus_name",
    facet_col_wrap=3,
    facet_col_spacing=0.03,
    width=1000,
    height=1000,
)
fig.update_traces(mode="lines+markers")
fig.for_each_annotation(lambda a: a.update(text=a.text.split("=")[-1]))
fig.update_layout(**utils.STD_LAYOUT, showlegend=False)
fig.update_xaxes(
    zerolinecolor="grey",
    tickmode="array",
    tickvals=[-12, -6, 0, 6, 12, 18],
    ticktext=["Dbb", "Gb", "C", "F#", "B#", "E##"],
    visible=True,
)
fig.update_yaxes(zeroline=False, matches=None, showticklabels=True)
save_figure_as(fig, "pitch_class_distribution_corpuswise_scatter")
fig.show()

no_accidental = bar_data[bar_data.tpc.between(-1, 5)].duration_qb.sum()
with_accidental = bar_data[~bar_data.tpc.between(-1, 5)].duration_qb.sum()

entire = no_accidental + with_accidental
(
    f"Fraction of note duration without accidental of the entire durations: {no_accidental} / {entire} = "
    f"{no_accidental / entire}"
)

'Fraction of note duration without accidental of the entire durations: 3432.4583333333335 / 4437.708333333334 = 0.7734754236890287'

Notes and staves

print("Distribution of notes over staves:")
utils.value_count_df(all_notes.staff)

Distribution of notes over staves:

	counts	%
staff
1	5548	62.16
2	3377	37.84