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("mozart_piano_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("The Annotated Mozart Sonatas: Score, Harmony, and Cadence version v2.3")
print(f"Datapackage '{package.package_name}' @ {git_tag}")
print(f"dimcat version {dc.__version__}\n")
D

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

The Annotated Mozart Sonatas: Score, Harmony, and Cadence version v2.3
Datapackage 'mozart_piano_sonatas' @ v2.3
dimcat version 3.4.0

Dataset
=======
{'inputs': {'basepath': None,
            'packages': {'mozart_piano_sonatas': ["'mozart_piano_sonatas.measures' "
                                                  '(MuseScoreFacetName.MuseScoreMeasures)',
                                                  "'mozart_piano_sonatas.notes' (MuseScoreFacetName.MuseScoreNotes)",
                                                  "'mozart_piano_sonatas.expanded' "
                                                  '(MuseScoreFacetName.MuseScoreHarmonies)',
                                                  "'mozart_piano_sonatas.chords' (MuseScoreFacetName.MuseScoreChords)",
                                                  "'mozart_piano_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	annotators	reviewers	composed_start
corpus
mozart_piano_sonatas	K279-1	{1: '4/4'}	{1: 0}	100	100	400.0	200	200	800.0	()	767.0	2031	1441	0	0	251	C	Uli Kneisel	Johannes Hentschel, Markus Neuwirth	1775

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()

104774 notes over 54 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
mozart_piano_sonatas	K279-1	0	1	1	0	0	0.25	1/16	0	0	4/4	2	1	<NA>	11	<NA>	48	C3	1/16	3	1	<NA>	C	0
		1	1	1	0	0	1.00	1/4	0	0	4/4	1	1	<NA>	0	<NA>	64	E4	1/4	4	1	<NA>	E	4
		2	1	1	0	0	1.00	1/4	0	0	4/4	1	1	<NA>	0	<NA>	67	G4	1/4	4	1	<NA>	G	1
		3	1	1	0	0	1.00	1/4	0	0	4/4	1	1	<NA>	0	<NA>	72	C5	1/4	5	1	<NA>	C	0
		4	1	1	1/4	1/4	0.25	1/16	1/16	1/16	4/4	2	1	<NA>	12	<NA>	60	C4	1/16	4	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	Mozart Piano Sonatas	29
1	Mozart Piano Sonatas	29
2	Mozart Piano Sonatas	30
3	Mozart Piano Sonatas	31
4	Mozart Piano Sonatas	31
...	...	...
2906	Mozart Piano Sonatas	88
2907	Mozart Piano Sonatas	89
2908	Mozart Piano Sonatas	89
2909	Mozart Piano Sonatas	89
2910	Mozart Piano Sonatas	89

2911 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	Mozart Piano Sonatas	-9
1	Mozart Piano Sonatas	-8
2	Mozart Piano Sonatas	-8
3	Mozart Piano Sonatas	-8
4	Mozart Piano Sonatas	-8
...	...	...
139507	Mozart Piano Sonatas	12
139508	Mozart Piano Sonatas	12
139509	Mozart Piano Sonatas	12
139510	Mozart Piano Sonatas	12
139511	Mozart Piano Sonatas	13

139512 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
mozart_piano_sonatas	K279-1	0	1	1	0	0	0.25	1/16	0	0	4/4	2	1	<NA>	11	<NA>	48	C3	1/16	3	1	<NA>	C	0	Mozart Piano Sonatas
		1	1	1	0	0	1.00	1/4	0	0	4/4	1	1	<NA>	0	<NA>	64	E4	1/4	4	1	<NA>	E	4	Mozart Piano Sonatas
		2	1	1	0	0	1.00	1/4	0	0	4/4	1	1	<NA>	0	<NA>	67	G4	1/4	4	1	<NA>	G	1	Mozart Piano Sonatas
		3	1	1	0	0	1.00	1/4	0	0	4/4	1	1	<NA>	0	<NA>	72	C5	1/4	5	1	<NA>	C	0	Mozart Piano Sonatas
		4	1	1	1/4	1/4	0.25	1/16	1/16	1/16	4/4	2	1	<NA>	12	<NA>	60	C4	1/16	4	1	<NA>	C	0	Mozart Piano Sonatas
	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
	K576-3	2415	188	188	375	375	0.50	1/8	1/4	1/4	2/4	1	1	<NA>	2276	<NA>	69	A4	1/8	4	1	<NA>	A	3	Mozart Piano Sonatas
		2416	188	188	375	375	0.50	1/8	1/4	1/4	2/4	1	1	<NA>	2276	<NA>	78	F#5	1/8	5	1	<NA>	F#	6	Mozart Piano Sonatas
		2417	189	189	376	376	1.00	1/4	0	0	2/4	2	1	<NA>	2280	<NA>	50	D3	1/4	3	1	<NA>	D	2	Mozart Piano Sonatas
		2418	189	189	376	376	1.00	1/4	0	0	2/4	1	1	<NA>	2279	<NA>	66	F#4	1/4	4	1	<NA>	F#	6	Mozart Piano Sonatas
		2419	189	189	376	376	1.00	1/4	0	0	2/4	1	1	<NA>	2279	<NA>	74	D5	1/4	5	1	<NA>	D	2	Mozart Piano Sonatas

104774 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: 39241.697916666664 / 52317.291666666664 = 0.7500712798110886'

Notes and staves

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

Distribution of notes over staves:

	counts	%
staff
1	56974	54.38
2	47800	45.62