from emod_api import schema_to_class as s2c
import pandas as pd
from emodpy_malaria.interventions.outbreak import add_campaign_event
[docs]def add_scale_larval_habitats(campaign, df=None,
start_day: int = 0, repetitions: int = 1,
timesteps_between_repetitions: int = 365):
"""
Reduce available larval habitat in a node-specific way.
Args:
campaign: campaign object to which the intervention will be added, and schema_path container
df: The dataframe containing habitat scale factors. **Examples**::
Scale TEMPORARY_RAINFALL by 3-fold for all nodes, all species:
df = pd.DataFrame({ 'TEMPORARY_RAINFALL': [3]})
Scale TEMPORARY_RAINFALL by 3-fold for all nodes, arabiensis only:
df = pd.DataFrame({ 'TEMPORARY_RAINFALL.arabiensis': [3]})
Scale differently by node ID:
df = pd.DataFrame({ 'NodeID' : [0, 1, 2, 3, 4],
'CONSTANT': [1, 0, 1, 1, 1],
'TEMPORARY_RAINFALL': [1, 1, 0, 1, 0]})
Scale differently by both node ID and species:
df = pd.DataFrame({ 'NodeID' : [0, 1, 2, 3, 4],
'CONSTANT.arabiensis': [1, 0, 1, 1, 1],
'TEMPORARY_RAINFALL.arabiensis': [1, 1, 0, 1, 0],
'CONSTANT.funestus': [1, 0, 1, 1, 1]})
Scale some habitats by species and others same for all species:
df = pd.DataFrame({ 'NodeID' : [0, 1, 2, 3, 4],
'CONSTANT.arabiensis': [1, 0, 1, 1, 1],
'TEMPORARY_RAINFALL.arabiensis': [1, 1, 0, 1, 0],
'CONSTANT.funestus': [1, 0, 1, 1, 1],
'LINEAR_SPLINE': [1, 1, 0, 1, 0]})
Scale nodes at different dates:
df = pd.DataFrame({ 'NodeID' : [0, 1, 2, 3, 4],
'CONSTANT': [1, 0, 1, 1, 1],
'TEMPORARY_RAINFALL': [1, 1, 0, 1, 0],
'Start_Day': [0, 30, 60, 65, 65]
})
start_day: The date that habitats are scaled for all scaling
actions specified in **df**. Used only if there is no
Start_Day column in **df**.
repetitions: The number of times to repeat the intervention.
timesteps_between_repetitions: The number of time steps between
repetitions.
Returns:
None
"""
if 'Start_Day' not in df.columns.values:
df['Start_Day'] = start_day
standard_columns = ['NodeID', 'Start_Day']
habitat_columns = [x for x in df.columns.values if x not in standard_columns]
habitat_names = list(set([x.split('.')[0] for x in habitat_columns]))
by_species = any(['.' in x for x in df.columns.values if x not in standard_columns])
by_node = 'NodeID' in df.columns.values
for start_day, df_by_date in df.groupby('Start_Day'):
for gn, gdf in df_by_date.groupby(habitat_columns):
habitat_scales = []
if not by_species:
if len(habitat_names) == 1:
habitat_scales.append({"Habitat": habitat_columns[0],
"Species": "ALL_SPECIES",
"Factor": float(gn[0])})
else:
for x, y in zip(habitat_columns, gn):
habitat_scales.append({"Habitat": x,
"Species": "ALL_SPECIES",
"Factor": float(y)})
else:
if len(habitat_names) == 1:
if len(habitat_columns) == 1:
habitat, sp = habitat_columns[0].split('.')
habitat_scales.append({"Habitat": habitat,
"Species": sp,
"Factor": float(gn[0])})
else:
habitat = habitat_columns[0].split('.')[0]
species = [x.split('.')[1] for x in habitat_columns]
for sp, x in zip(species, gn):
habitat_scales.append({"Habitat": habitat,
"Species": sp,
"Factor": float(x)})
else:
for ih, habitat in enumerate(habitat_names):
if habitat in habitat_columns:
habitat_scales.append({"Habitat": habitat,
"Species": "ALL_SPECIES",
"Factor": float(gdf.iloc[0][habitat])})
else:
h = [x for x in habitat_columns if habitat in x]
vals = [gn[x] for x in range(len(habitat_columns)) if habitat in habitat_columns[x]]
for x, y in zip(h, vals):
habitat_scales.append({"Habitat": habitat,
"Species": x.split('.')[1],
"Factor": float(y)})
if by_node:
node_ids = [int(x) for x in gdf['NodeID']]
else:
node_ids = None
add_habitat_reduction_event(campaign, start_day=start_day, node_ids=node_ids, habitat_scales=habitat_scales,
repetitions=repetitions,
timesteps_between_repetitions=timesteps_between_repetitions)
[docs]def add_habitat_reduction_event(campaign, start_day: int, node_ids: list, habitat_scales: list, repetitions: int,
timesteps_between_repetitions: int):
"""
Add a campaign event to reduce vector's larval habitat(s).
Args:
campaign: campaign object to which the intervention will be added, and schema_path container
start_day: The day the intervention is given out.
node_ids: List of nodes to which to distribute the intervention. [] or None, indicates all nodes
will get the intervention
habitat_scales: List of dictionaries for scaling larval habitats.
Examples::
[{"Habitat": "ALL_HABITATS", "Species": "ALL_SPECIES", "Factor": 0.5},
{"Habitat": "CONSTANT", "Species": "arabiensis", "Factor": 2}]
repetitions: The number of times an intervention is given, used with timesteps_between_repetitions. -1 means
the intervention repeats forever. Sets **Number_Repetitions**
timesteps_between_repetitions: The interval, in timesteps, between repetitions. Ignored if repetitions = 1.
Sets **Timesteps_Between_Repetitions**
Returns:
Nothing
"""
# configuring the intervention itself
scale_larval_habitat_intervention = s2c.get_class_with_defaults("ScaleLarvalHabitat", campaign.schema_path)
# scale_larval_habitat_intervention.Larval_Habitat_Multiplier = larval_habitat_multiplier_list
scale_larval_habitat_intervention.Larval_Habitat_Multiplier = habitat_scales
add_campaign_event(campaign=campaign, start_day=start_day, node_ids=node_ids, repetitions=repetitions,
timesteps_between_repetitions=timesteps_between_repetitions,
intervention=scale_larval_habitat_intervention)