JoaoCarabetta

import textwrap
import cufflinks as cf
cf.go_offline()
cf.set_config_file(offline=False, world_readable=True)

def iplottitle(title, width=40):
    return '<br>'.join(textwrap.wrap(title, width))

cf.themes.THEMES['clean'] = {'layout': {'height': 500,
 'legend': {'bgcolor': 'white',

JoaoCarabetta

How to use

This class is supposed to be used inside a Jupyter notebook with Keplergl activated.

It just makes it easier to organize maps and save configuration files.

The advantage is that once the config is save, the map with the same identifier_string will load with the saved config.

Initialize Map

JoaoCarabetta

def line_polygon_intersection(line_df, poly_df):
    
    column_geom_poly = poly_df._geometry_column_name
    column_geom_line = line_df._geometry_column_name
    
    spatial_index = line_df.sindex
    bbox = poly_df.geometry.apply(lambda x: x.bounds)
    sidx = bbox.apply(lambda x: list(spatial_index.intersection(x)))
    
    nei = []

JoaoCarabetta

import plotly
import plotly.graph_objs as go
import cufflinks as cf
plotly.offline.init_notebook_mode(connected=True)

cf.go_offline()
cf.set_config_file(offline=False, world_readable=True)

# Add figures
figs = [df.query(f"group == '{group}'").figure()

JoaoCarabetta

from multiprocessing.pool import Pool
from functools import partial
import time

n_processes = 3

def func(extra, t):
    time.sleep(t + extra)
    print('t', t, 'extra', extra)
    

JoaoCarabetta

/* Stratified splits dataset into training and test 

It guarantees that each group has the minimum size to be split.
*/

with ssize as (
	select
		group
	from  to_split_table
	group by group

JoaoCarabetta

/* It calculates the mode of a records-like `maintable`
*/
with counter as (
  select
    service,
    array[
      cast(row('Bob',  count_if(name = 'Bob')) AS row(name varchar, age interger)),
      cast(row('Alice', count_if(name = 'Alice')) AS row(name varchar, age interger)),
      cast(row('Jane',  count_if(name = 'Jane')) AS row(name varchar, age interger))
    ] as users

JoaoCarabetta

import logging
from contextlib import contextmanager

log = logging.getLogger(__name__)

@contextmanager
def timed_log(name, time_chunk='seconds'):
    """Context manager to get execution time of parts of codes.
    To use, simply declares the context manager:
        ```with timed_log(name='useful', time_chunck='minutes'):

JoaoCarabetta

def persist_local(data, args, folder, 
                    id_keys=['experiment_id'], 
                    as_type='.parquet.gz',
                    save_path='persistance'):

    save_path = build_path(args, folder, id_keys, as_type, save_path)

    # Check if path exists, add path if not
    if not save_path.parent.exists():
        save_path.parent.mkdir(mode=777, parents=True, exist_ok=True)

JoaoCarabetta

def open_yaml(path):
    """
    Load yaml file.
    Parameters
    ----------
    path: pathlib.PosixPath
        Path to yaml file
    Return
    ------
    Dictionary