Skip to content

Commit

Permalink
Full annotated code
Browse files Browse the repository at this point in the history
  • Loading branch information
@alexbelgium
alexbelgium authored Oct 30, 2024
1 parent 2e9bee8 commit e3591e0
Showing 1 changed file with 68 additions and 47 deletions.
115 changes: 68 additions & 47 deletions scripts/dynamic_plot.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -4,19 +4,36 @@
import numpy as np
from datetime import datetime
import os
from utils.helpers import get_settings # Import to access COLOR_SCHEME

# Define custom colorscales for light and dark modes
COLOR_SCALES = {
"light": [
[0.0, '#E0F2E9'], [0.2, '#A3D8A1'], [0.4, '#70BD70'],
[0.6, '#46A846'], [0.8, '#2E7D2E'], [1.0, '#004D00']
],
"dark": [
[0.0, '#F0F0F0'], [0.2, '#BDBDBD'], [0.4, '#969696'],
[0.6, '#737373'], [0.8, '#525252'], [1.0, '#252525']
]
}
from utils.helpers import get_settings # Import if needed for other settings

# Fetch color scheme setting and choose corresponding colorscale and text color
conf = get_settings()
color_scheme = conf.get('COLOR_SCHEME', 'light')

if color_scheme == 'dark':
PLOT_BGCOLOR='#F0F0F0'
PAPER_BGCOLOR='#7F7F7F'
CUSTOM_COLOR_SCALE = [
[0.0, PLOT_BGCOLOR],
[0.2, '#BDBDBD'],
[0.4, '#969696'],
[0.6, '#737373'],
[0.8, '#525252'],
[1.0, '#252525']
]
else:
PLOT_BGCOLOR='#FFFFFF'
PAPER_BGCOLOR='#7BC58A'
CUSTOM_COLOR_SCALE = [
[0.0, PLOT_BGCOLOR],
[0.1, '#E0F2E9'],
[0.2, '#A3D8A1'],
[0.4, '#70BD70'],
[0.6, '#46A846'],
[0.8, '#2E7D2E'],
[1.0, '#004D00']
]

ALL_HOURS = list(range(24))


Expand All @@ -28,9 +45,9 @@ def normalize_logarithmic(arr):
return np.log(arr / min_val) / np.log(np.max(arr) / min_val) if np.max(arr) > min_val else arr - min_val


def determine_text_color(z, threshold=0.5):
"""Determines text color (black or white) based on normalized value of z."""
return np.where(z > threshold, 'white', 'black')
def determine_text_color(z, threshold=0.8):
"""Determines text color (darkgrey or white) based on normalized value of z."""
return np.where(z == 0, PLOT_BGCOLOR, np.where(z > threshold, PLOT_BGCOLOR, '#1A1A1A'))


def add_annotations(fig, text_array, text_colors, col, row, species_list, all_hours, annotations):
Expand All @@ -41,7 +58,7 @@ def add_annotations(fig, text_array, text_colors, col, row, species_list, all_ho
if current_text:
annotations.append(dict(
x=0, y=species, text=current_text, showarrow=False,
font=dict(color=current_color, size=10),
font=dict(color=current_color, size=12),
xref=f'x{col}', yref=f'y{col}', xanchor='center', yanchor='middle'
))
elif col == 3: # Multi-column heatmap
Expand All @@ -51,7 +68,7 @@ def add_annotations(fig, text_array, text_colors, col, row, species_list, all_ho
if current_text:
annotations.append(dict(
x=hour, y=species, text=current_text, showarrow=False,
font=dict(color=current_color, size=10),
font=dict(color=current_color, size=12),
xref='x3', yref='y3', xanchor='center', yanchor='middle'
))

Expand All @@ -74,11 +91,6 @@ def create_plotly_heatmap(df_birds, now):
Conf=('Confidence', 'max')
).reset_index().fillna({'Conf': 0, 'Count': 0})

# Fetch color scheme setting and choose corresponding colorscale and text color
conf = get_settings()
color_scheme = conf.get('COLOR_SCHEME', 'light')
color_scale = COLOR_SCALES.get(color_scheme, COLOR_SCALES["light"])

# Summarize data for heatmap axes
df_birds_summary = plot_dataframe.groupby('Com_Name').agg(
Count=('Count', 'sum'),
Expand All @@ -89,58 +101,70 @@ def create_plotly_heatmap(df_birds, now):
species_list = df_birds_summary['Com_Name'].tolist()

# Normalize values and prepare text annotations
z_confidence = normalize_logarithmic(df_birds_summary['Conf'].values.reshape(-1, 1) * 100)
text_confidence = np.char.add(np.round(df_birds_summary['Conf'].values).astype(int).astype(str), ' %')
z_confidence = normalize_logarithmic(df_birds_summary['Conf'].values.reshape(-1, 1)) * 100
text_confidence = np.char.add((df_birds_summary['Conf'].values * 100).round().astype(int).astype(str), ' %')

z_detections = normalize_logarithmic(df_birds_summary['Count'].values.reshape(-1, 1))
text_detections = df_birds_summary['Count'].astype(str).values # Use actual counts for annotations
text_color_detections = determine_text_color(z_detections, threshold=0.5, color_scheme=color_scheme)
text_color_detections = determine_text_color(z_detections, threshold=0.5) # Removed color_scheme

df_hourly = plot_dataframe.pivot_table(index='Com_Name', columns='Hour', values='Count', aggfunc='sum').fillna(0)
df_hourly = df_hourly.reindex(species_list).fillna(0).reindex(columns=ALL_HOURS, fill_value=0)
z_hourly = normalize_logarithmic(df_hourly.values)
text_hourly = df_hourly.astype(int).astype(str).values # Use actual counts for hourly annotations
text_color_hourly = determine_text_color(z_hourly, threshold=0.5, color_scheme=color_scheme)
text_color_hourly = determine_text_color(z_hourly, threshold=0.5) # Removed color_scheme

# Create subplots
fig = make_subplots(rows=1, cols=3, shared_yaxes=True, column_widths=[0.1, 0.1, 0.7], horizontal_spacing=0.02)

# Heatmap Traces
# Prepare structured customdata arrays for each heatmap trace
custom_data_confidence = np.array([{'confidence': conf * 100} for conf in df_birds_summary['Conf'].values]).reshape(-1, 1)
custom_data_count = np.array([{'count': count} for count in df_birds_summary['Count'].values]).reshape(-1, 1)

# Add traces with updated customdata structure and hovertemplate
fig.add_trace(go.Heatmap(
z=z_confidence, customdata=df_birds_summary['Conf'].values, x=['Confidence'], y=species_list,
colorscale=color_scale, showscale=False, hovertemplate='Species: %{y}<br>Max Confidence: %{customdata:.0f}%<extra></extra>',
z=z_confidence, customdata=custom_data_confidence, x=['Confidence'], y=species_list,
colorscale=CUSTOM_COLOR_SCALE, showscale=False,
hovertemplate='Species: %{y}<br>Max Confidence: %{customdata.confidence:.0f}%<extra></extra>',
xgap=1, ygap=1, zmin=0, zmax=1
), row=1, col=1)

fig.add_trace(go.Heatmap(
z=z_detections, customdata=df_birds_summary['Count'].values, x=['Count'], y=species_list,
colorscale=color_scale, showscale=False, hovertemplate='Species: %{y}<br>Total Counts: %{customdata}<extra></extra>',
z=z_detections, customdata=custom_data_count, x=['Count'], y=species_list,
colorscale=CUSTOM_COLOR_SCALE, showscale=False,
hovertemplate='Species: %{y}<br>Total Counts: %{customdata.count}<extra></extra>',
xgap=1, ygap=1, zmin=0, zmax=1
), row=1, col=2)

fig.add_trace(go.Heatmap(
z=z_hourly, customdata=df_hourly.values, x=ALL_HOURS, y=species_list,
colorscale=color_scale, showscale=False, hovertemplate='Species: %{y}<br>Hour: %{x}<br>Detections: %{customdata}<extra></extra>',
colorscale=CUSTOM_COLOR_SCALE, showscale=False, hovertemplate='Species: %{y}<br>Hour: %{x}<br>Detections: %{customdata}<extra></extra>',
xgap=1, ygap=1, zmin=0, zmax=1
), row=1, col=3)

# Annotations
annotations = []
add_annotations(fig, text_confidence.reshape(-1, 1), determine_text_color(z_confidence, threshold=0.5, color_scheme=color_scheme), col=1, row=1, species_list=species_list, all_hours=ALL_HOURS, annotations=annotations)
add_annotations(fig, text_detections.reshape(-1, 1), text_color_detections, col=2, row=1, species_list=species_list, all_hours=ALL_HOURS, annotations=annotations)
add_annotations(fig, text_hourly, text_color_hourly, col=3, row=1, species_list=species_list, all_hours=ALL_HOURS, annotations=annotations)
add_annotations(fig, text_confidence.reshape(-1, 1), determine_text_color(z_confidence, threshold=0.5),
col=1, row=1, species_list=species_list, all_hours=ALL_HOURS, annotations=annotations)
add_annotations(fig, text_detections.reshape(-1, 1), text_color_detections,
col=2, row=1, species_list=species_list, all_hours=ALL_HOURS, annotations=annotations)
add_annotations(fig, text_hourly, text_color_hourly,
col=3, row=1, species_list=species_list, all_hours=ALL_HOURS, annotations=annotations)
fig.update_layout(annotations=annotations)

# Layout configuration
fig.update_layout(
title=dict(text=f"<b>{main_title}</b><br><span style='font-size:14px;'>{subtitle}</span>", x=0.5, y=0.97, xanchor='center', yanchor='top', font=dict(size=24)),
title=dict(text=f"<b>{main_title}</b><br><span style='font-size:14px;'>{subtitle}</span>",
x=0.5, y=0.97, xanchor='center', yanchor='top', font=dict(size=24)),
autosize=True, height=max(600, len(species_list) * 25 + 100),
yaxis=dict(autorange='reversed', tickfont=dict(size=10), showticklabels=True, ticklabelstandoff=15, fixedrange=True),
xaxis1=dict(title='Max confidence', showticklabels=False, title_font=dict(size=10), fixedrange=True),
xaxis2=dict(title='Total counts', showticklabels=False, title_font=dict(size=10), fixedrange=True),
xaxis3=dict(title='Hour', tickfont=dict(size=10), tickmode='linear', dtick=1, fixedrange=True),
margin=dict(l=20, r=20, t=80, b=80), clickmode='event+select',
plot_bgcolor='#CCCCCC', paper_bgcolor='#7F7F7F', font=dict(size=10), dragmode=False
yaxis=dict(autorange='reversed', tickfont=dict(size=12), showticklabels=True, ticklabelstandoff=15, fixedrange=True),
xaxis1=dict(title='Max Confidence', showticklabels=False, title_font=dict(size=12), fixedrange=True),
xaxis2=dict(title='Total Counts', showticklabels=False, title_font=dict(size=12), fixedrange=True),
xaxis3=dict(title='Hour', tickfont=dict(size=12), tickmode='linear', dtick=1, fixedrange=True),
margin=dict(l=20, r=20, t=80, b=80), clickmode='event+select',
plot_bgcolor=PAPER_BGCOLOR,
paper_bgcolor=PAPER_BGCOLOR,
font=dict(size=12, color='#000000'), dragmode=False
)
fig.update_xaxes(showgrid=False, zeroline=False)
fig.update_yaxes(showgrid=False, zeroline=False)
Expand Down Expand Up @@ -197,6 +221,3 @@ def create_plotly_heatmap(df_birds, now):
os.makedirs(output_dir, exist_ok=True)
with open(os.path.join(output_dir, 'interactive_daily_plot.html'), 'w') as f:
f.write(html_str)

# Clear figure to reset layout and prevent issues in future runs
fig.clear()

0 comments on commit e3591e0

Please sign in to comment.