suppfig_end_session_histogram.py

"""
HISTOGRAM OF SESSION END STATUSES DURING TRAINING
Miles  Wells, UCL, 2019
"""
import os

import matplotlib.pyplot as plt
import numpy as np
import seaborn as sns

import datajoint as dj
from ibl_pipeline import acquisition
from paper_behavior_functions import \
    (figpath, query_sessions, query_subjects, group_colors, seaborn_style,
     FIGURE_HEIGHT, FIGURE_WIDTH, QUERY)

assert QUERY, 'This script requires a DataJoint instance, which was removed in Dec 2023.'

# Set default figure size.
save_path = figpath()  # Our figure save path
colors = group_colors()
seaborn_style()

endcriteria = dj.create_virtual_module('SessionEndCriteriaImplemented',
                                       'group_shared_end_criteria')
sessions = query_sessions().proj(session_start_date='date(session_start_time)')
subj_crit = query_subjects().aggr(
                     acquisition.Session(),
                     first_day='min(date(session_start_time))').proj('first_day')
session_num = (sessions * subj_crit).proj(n='DATEDIFF(session_start_date, first_day)')

df = (endcriteria.SessionEndCriteriaImplemented * session_num).fetch(format='frame')  # Fetch data

# Convert statuses to numerical
fig, ax = plt.subplots(1, 1, figsize=(FIGURE_WIDTH/4, FIGURE_HEIGHT))
ids = {k: v for v, k in enumerate(df['end_status'].unique())}
df['end_status_id'] = df['end_status'].map(ids)
bins = [0, 6, 13, 20, 27, 34]
ax = df.pivot(columns='end_status_id').n.plot(
    kind='hist', color=colors, bins=bins, stacked=True, density=True)  # weights=369,
ax.legend(ids.keys())
ax.set_xlabel('Session #')
ax.set_ylabel('Frequency')
plt.gcf().savefig(os.path.join(save_path, "suppfig_end_status_histogram.png"))

# Unity plot
max_n_days = 40
normalize = True
df = df.reset_index()
counts = np.array([[sum(df['end_status_id'].where(df['n'] == n_days) == criterion)
                    if n_days < max_n_days
                    else sum(df['end_status_id'].where(df['n'] >= n_days) == criterion)
                    for n_days in range(max_n_days+1)]
                   for criterion in np.sort(df['end_status_id'].unique())])

if normalize:
    counts = np.stack([n / sum(n) for n in counts.T]).T
    #  counts = np.stack([n / sum(n) for n in counts])

bar_l = range(1, counts.shape[1]+1)
#  bottom = np.zeros_like(bar_l).astype('float')
bottom = np.vstack((np.zeros((1, counts.shape[1])), np.cumsum(counts, axis=0)[:-1, :]))

fig, ax = plt.subplots(1, 1, figsize=(FIGURE_WIDTH / 2, FIGURE_HEIGHT))
for i in range(counts.shape[0]):
    ax.bar(bar_l, counts[i, :], bottom=bottom[i, :], width=1, label=list(ids.keys())[i],
           color=colors[i])

ax.set_xticks([1] + [i * 7 for i in range(1, round(max_n_days+7/7))])
ax.set_xticks([0, 10, 20, 30, 40])

ax.set_xlim([0, counts.shape[1]+.5])
ax.set_xlabel('Session #')
ax.set_ylabel('Proportion')
ax.legend(loc='upper right')
plt.tight_layout()
sns.despine(trim=False)
plt.gcf().savefig(os.path.join(save_path, "suppfig_end_status_histogram_normalized.png"), dpi=300)
plt.gcf().savefig(os.path.join(save_path, "suppfig_end_status_histogram_normalized.pdf"))