def plot_selections_per_epoch(
data_file='results/update_grades_over_bach_chorales.csv',
plt_dir='plots/augmented-generation/'):
"""
plot number of selections each epoch
"""
thres = get_threshold(
data_file='experiments/ablations/reg_pe_no_oe/bach_grades.csv',
column='grade',
aggregate='75p',
)
data_dict = read_training_data(data_file=data_file, feature='grade')
picked = [
np.sum([1 for x in data if x < thres]) for data in data_dict.values()
]
plt.figure()
fig, ax = plt.subplots()
plt.style.use('seaborn-whitegrid')
ax.grid(False)
rects = plt.bar(range(1, len(picked) + 1), picked)
label_bars(rects)
plt.xlabel('Epoch')
plt.ylabel('Number of generations passing threshold')
plt.title('Number of Generations Passing Threshold in Each Epoch')
plt.savefig(
os.path.join(plt_dir, 'generations_passing_threshold_per_epoch.png'))
def plot_histogram_per_iteration(data_file,
feature='grade',
plt_dir='plots/augmented-generation/',
threshold=None):
"""
visualize model updates by plotting histogram for grade distribution at each iteration
"""
# read update data as dictionary
data_dict = read_training_data(data_file=data_file,
feature=feature,
threshold=threshold)
plt.figure(figsize=(20, 10))
plt.style.use('seaborn-whitegrid')
for it, data in data_dict.items():
plt.subplot(2, 5, it)
plt.hist(data, alpha=0.7)
plt.xlabel(feature)
plt.title(f'Iteration {it}')
plt.tight_layout(rect=[0, 0.03, 1, 0.95])
plt.suptitle(
f'{feature} Distribution of Generations at Each Iteration of Training',
fontsize=20)
ensure_dir(plt_dir)
plt.savefig(os.path.join(plt_dir, f'{feature}_update_dist.png'))
plt.close()
def main():
bach_data = list(pd.read_csv(f'{bach_dir}/grades.csv')['grade'])
# aug_gen_data = list(read_training_data(f'{aug_gen_dir}/grades.csv', feature='grade')[20])
# base_data = list(read_training_data(f'{base_dir}/grades.csv', feature='grade')[17])
baseline_data = list(
read_training_data(f'{baseline_dir}/grades.csv', feature='grade')[17])
aug_gen_df = pd.read_csv(f'{aug_gen_dir}/351_mocks/grades.csv')
aug_gen_data = aug_gen_df['grade']
base_df = pd.read_csv(f'{base_dir}/351_mocks/grades.csv')
base_data = base_df['grade']
base_data = [x for x in base_data if x < 50]
baseline_data = [x for x in baseline_data if x < 50]
data_dict = {
'Bach': bach_data,
'Aug-Gen\n' + r'($t=Q_3$ of Bach grades)': aug_gen_data,
'Baseline-none\n' + r'($t=-\infty$)': base_data,
'Baseline-all\n' + r'($t=\infty$)': baseline_data,
}
for model, data in data_dict.items():
print(model)
print(np.median(data))
print(np.std(data))
plot_violinplots(
data_dict=data_dict,
plt_title='Grade Distribution of Generations from Different Models',
plt_dir='plots/')
def plot_boxplot_per_epoch(
data_file='results/update_grades_over_bach_chorales.csv',
feature='grade',
plt_dir='plots/augmented-generation/',
threshold=None):
"""
Arguments
data_file: file containing upgrade grades
feature: feature of interest (either overall grade or a feature distance)
plt_dir: directory to save plots
threshold: lower threshold for inclusion
visualize model updates by plotting boxplot for grade distribution at each epoch
"""
# read update data as dictionary
data_dict = read_training_data(data_file=data_file, feature=feature)
# plot
plt.figure()
plt.style.use('seaborn-whitegrid')
plt.rc('xtick', labelsize=11)
plt.rc('ytick', labelsize=11)
plt.rc('axes', titlesize=13)
fig, ax = plt.subplots(figsize=(8, 5))
ax.xaxis.grid(False)
ax.boxplot(list(data_dict.values()))
ax.set_xticks([i + 1 for i in data_dict.keys()])
ax.set_xticklabels([str(i) for i in data_dict.keys()])
for label in ax.get_xaxis().get_ticklabels()[1::2]:
label.set_visible(False)
ylabel0 = ax.get_yaxis().get_ticklabels()[0]
ylabel0.set_visible(False)
plt.xlabel('Epoch')
plt.title(
f'{feature.capitalize()} Distribution of Generations During Aug-Gen Training'
)
plt.text(-2.2, 1, 'better')
plt.text(-2.2, 47, 'worse')
plt.ylabel(feature.capitalize())
plt.ylim([0, 49.15])
threshold = get_threshold(
data_file='experiments/ablations/reg_pe_no_oe/bach_grades.csv',
column='grade',
aggregate='75p',
)
plt.axhline(y=threshold,
color='steelblue',
linestyle='-.',
label=r'$Q_3$' + ' of Bach grades')
plt.legend(loc='upper right')
ensure_dir(plt_dir)
fig.tight_layout()
plt.savefig(os.path.join(plt_dir, f'{feature}_update_boxplots.png'))
def plot_median_grade_per_epoch(dir_dict, num_epochs):
median_dict = defaultdict(lambda: [0] * num_epochs)
for model_label, model_path in dir_dict.items():
data_dict = read_training_data(data_file=f'{model_path}/grades.csv',
feature='grade')
for epoch, grades in data_dict.items():
if epoch < num_epochs:
median_dict[model_label][epoch] = np.median(grades)
plt.figure()
plt.style.use('seaborn-whitegrid')
fig, ax = plt.subplots()
ax.grid(False)
thres = get_threshold(
data_file='experiments/ablations/reg_pe_no_oe/bach_grades.csv',
column='grade',
aggregate='75p',
)
plt.axhline(y=thres,
dashes=(2, 2),
label='Lowest Bach\ngrade threshold',
color=PLOT_COLORS['bach'])
xlim = range(num_epochs)
for model_label, median_grades in median_dict.items():
plt.plot(xlim,
median_grades[:num_epochs],
label=PLOT_LABELS[model_label],
color=PLOT_COLORS[model_label])
plt.title('Median Grade of Generations During Training')
ax.set_xticks([i + 1 for i in xlim])
ax.set_xticklabels([str(i) for i in xlim])
for label in ax.get_xaxis().get_ticklabels()[1::2]:
label.set_visible(False)
# plt.legend(loc='right')
handles, labels = ax.get_legend_handles_labels()
lgd = ax.legend(handles,
labels,
loc='upper center',
bbox_to_anchor=(-0.2, 0.5))
plt.ylabel('Grade')
plt.xlabel('Epoch')
plt.savefig('plots/median_grades_per_epoch.png', bbox_inches='tight')