def calculate_bee_lifespans_from_hatchdates():
"""
Uses the Last_day_alive.csv cache and the get_hatchdate functions from bb_utils
to calculate the lifespan (length-of-life) in days for each bee. The lifespan will be NaN if hatchdate is NaT.
"""
meta = bb_utils.meta.BeeMetaInfo()
last_alive_path = detections_cache_path + 'Last_day_alive.csv'
last_day_alive_df = pd.read_csv(last_alive_path,
parse_dates=['max'],
usecols=['max', 'bee_id'])
last_day_alive_df.index = last_day_alive_df['bee_id']
lifespan = pd.DataFrame()
for ix in last_day_alive_df['bee_id']:
birthday = meta.get_hatchdate(
bb_utils.ids.BeesbookID.from_dec_9(ix)).date()
temp_lifespan = pd.DataFrame(
[[ix, (last_day_alive_df.loc[ix]['max'].date() - birthday).days]],
columns=['id', 'lifespan'])
lifespan = pd.concat([lifespan, temp_lifespan])
lifespan.index = last_day_alive_df['bee_id']
lifespan = lifespan['lifespan']
lifespan = filter_out_fake_deaths(lifespan)
return lifespan
def calculate_bee_lifespans_combined():
last_alive_path = detections_cache_path + 'Last_day_alive.csv'
first_alive_path = detections_cache_path + 'First_day_alive.csv'
last_day_alive_df = pd.read_csv(last_alive_path,
parse_dates=['max'],
usecols=['max', 'bee_id'])
first_day_alive_df = pd.read_csv(first_alive_path,
parse_dates=['min'],
usecols=['min', 'bee_id'])
last_day_alive_df.index = last_day_alive_df['bee_id']
first_day_alive_df.index = first_day_alive_df['bee_id']
meta = bb_utils.meta.BeeMetaInfo()
lifespan_from_detections = calculate_bee_lifespans_from_detections()
lifespan = pd.DataFrame()
for ix in last_day_alive_df['bee_id']:
birthday = meta.get_hatchdate(
bb_utils.ids.BeesbookID.from_dec_9(ix)).date()
if pd.isnull(birthday):
birthday = first_day_alive_df.loc[ix]['min'].date()
temp_lifespan = pd.DataFrame(
[[ix, (last_day_alive_df.loc[ix]['max'].date() - birthday).days]],
columns=['id', 'lifespan'])
lifespan = pd.concat([lifespan, temp_lifespan])
lifespan.index = last_day_alive_df['bee_id']
lifespan = lifespan['lifespan']
lifespan = filter_out_fake_deaths(lifespan)
return lifespan
def get_all_bee_ids():
meta = bb_utils.meta.BeeMetaInfo()
bees = []
for id in range(0,4096):
bee_id = bb_utils.ids.BeesbookID.from_dec_9(id)
if meta.get_hatchdate(bee_id) != np.datetime64('NaT'):
bees.append(bee_id)
bee_ids_as_beesbookid_format = list(bees)
bee_ids_as_ferwar_format = map(lambda i: i.as_ferwar(), bee_ids_as_beesbookid_format) #as ferwar
return (list(bee_ids_as_ferwar_format), bee_ids_as_beesbookid_format)
def get_random_bee_ids(amount):
meta = bb_utils.meta.BeeMetaInfo()
bees = set()
while len(bees) < amount:
bee_id_dec = random.randint(0, 4097)
bee_id = bb_utils.ids.BeesbookID.from_dec_9(bee_id_dec)
if meta.get_hatchdate(bee_id) != np.datetime64('NaT'):
bees.add(bee_id)
bee_ids_as_beesbookid_format = list(bees)
bee_ids_as_ferwar_format = map(lambda i: i.as_ferwar(), bee_ids_as_beesbookid_format) #as ferwar
return (list(bee_ids_as_ferwar_format), bee_ids_as_beesbookid_format)
# ### Get a group of bees to work on and calculate their ages
# In[3]:
#TODO: Filter out dead bees
#TODO: Move this out to a function
#Get all bees/n random bees/forager group 20
(bee_ids_as_ferwar_format, bee_ids_as_beesbookid_format) = get_all_bee_ids()
#Calculate the ages for each bee
bee_days_since_birth = []
#TODO: calculate ages from the dataframe later (after filtering)
for id in bee_ids_as_beesbookid_format:
bee_days_since_birth.append((datetime_start - meta.get_hatchdate(id)).days)
# ### Load a PRESENCE.csv cache (saved from the DB_TO_DETECTIONS notebook)
# In[86]:
(csv_name, csv_path) = create_presence_cache_filename(num_hours,
datetime_start,
num_intervals_per_hour)
print('Starting with ' + csv_path)
presence_df = pd.read_csv(csv_path).iloc[:, 1:]
print(presence_df.shape)
presence_df.head()
# In[87]: