Live Notebook

You can run this notebook in a live session Binder or view it on Github.

Dask Bags

Dask Bag implements operations like map, filter, groupby and aggregations on collections of Python objects. It does this in parallel and in small memory using Python iterators. It is similar to a parallel version of itertools or a Pythonic version of the PySpark RDD.

Dask Bags are often used to do simple preprocessing on log files, JSON records, or other user defined Python objects.

Full API documentation is available here:

Start Dask Client for Dashboard

Starting the Dask Client is optional. It will provide a dashboard which is useful to gain insight on the computation.

The link to the dashboard will become visible when you create the client below. We recommend having it open on one side of your screen while using your notebook on the other side. This can take some effort to arrange your windows, but seeing them both at the same is very useful when learning.

from dask.distributed import Client, progress
client = Client(n_workers=4, threads_per_worker=1)



  • Workers: 4
  • Cores: 4
  • Memory: 8.36 GB

Create Random Data

We create a random set of record data and store it to disk as many JSON files. This will serve as our data for this notebook.

import dask
import json
import os

os.makedirs('data', exist_ok=True)              # Create data/ directory

b = dask.datasets.make_people()                 # Make records of people'data/*.json')   # Encode as JSON, write to disk

Read JSON data

Now that we have some JSON data in a file lets take a look at it with Dask Bag and Python JSON module.

!head -n 2 data/0.json
{"age": 39, "name": ["Wilfredo", "Silva"], "occupation": "Farmer", "telephone": "(489) 718-1463", "address": {"address": "760 Russia Drive", "city": "Allen Park"}, "credit-card": {"number": "4890 7910 1700 3335", "expiration-date": "05/23"}}
{"age": 25, "name": ["Erline", "Sloan"], "occupation": "Chandler", "telephone": "+1-(404)-661-5342", "address": {"address": "234 Shaw Brae", "city": "Wyoming"}, "credit-card": {"number": "3452 021586 06729", "expiration-date": "07/25"}}
import dask.bag as db
import json

b = db.read_text('data/*.json').map(json.loads)
dask.bag<loads, npartitions=10>
({'age': 39,
  'name': ['Wilfredo', 'Silva'],
  'occupation': 'Farmer',
  'telephone': '(489) 718-1463',
  'address': {'address': '760 Russia Drive', 'city': 'Allen Park'},
  'credit-card': {'number': '4890 7910 1700 3335',
   'expiration-date': '05/23'}},
 {'age': 25,
  'name': ['Erline', 'Sloan'],
  'occupation': 'Chandler',
  'telephone': '+1-(404)-661-5342',
  'address': {'address': '234 Shaw Brae', 'city': 'Wyoming'},
  'credit-card': {'number': '3452 021586 06729', 'expiration-date': '07/25'}})

Map, Filter, Aggregate

We can process this data by filtering out only certain records of interest, mapping functions over it to process our data, and aggregating those results to a total value.

b.filter(lambda record: record['age'] > 30).take(2)  # Select only people over 30
({'age': 39,
  'name': ['Wilfredo', 'Silva'],
  'occupation': 'Farmer',
  'telephone': '(489) 718-1463',
  'address': {'address': '760 Russia Drive', 'city': 'Allen Park'},
  'credit-card': {'number': '4890 7910 1700 3335',
   'expiration-date': '05/23'}},
 {'age': 50,
  'name': ['Spencer', 'Wood'],
  'occupation': 'Research Director',
  'telephone': '(431) 111-2477',
  'address': {'address': '1375 Eldridge Park', 'city': 'Coachella'},
  'credit-card': {'number': '4142 6029 2679 1462',
   'expiration-date': '07/17'}})
[7]: record: record['occupation']).take(2)  # Select the occupation field
('Farmer', 'Chandler')
b.count().compute()  # Count total number of records

Chain computations

It is common to do many of these steps in one pipeline, only calling compute or take at the end.

result = (b.filter(lambda record: record['age'] > 30)
           .map(lambda record: record['occupation'])
           .topk(10, key=1))
dask.bag<topk-aggregate, npartitions=1>

As with all lazy Dask collections, we need to call compute to actually evaluate our result. The take method used in earlier examples is also like compute and will also trigger computation.

[('Hairdresser', 18),
 ('Kennel Hand', 15),
 ('Psychoanalyst', 15),
 ('Plant Manager', 14),
 ('Marketing Coordinator', 14),
 ('Undertaker', 14),
 ('Bank Messenger', 14),
 ('Technical Instructor', 14),
 ('Licensed Premises', 13),
 ('Beauty Therapist', 13)]

Transform and Store

Sometimes we want to compute aggregations as above, but sometimes we want to store results to disk for future analyses. For that we can use methods like to_textfiles and json.dumps, or we can convert to Dask Dataframes and use their storage systems, which we’ll see more of in the next section.

(b.filter(lambda record: record['age'] > 30)  # Select records of interest
  .map(json.dumps)                            # Convert Python objects to text
  .to_textfiles('data/processed.*.json'))     # Write to local disk

Convert to Dask Dataframes

Dask Bags are good for reading in initial data, doing a bit of pre-processing, and then handing off to some other more efficient form like Dask Dataframes. Dask Dataframes use Pandas internally, and so can be much faster on numeric data and also have more complex algorithms.

However, Dask Dataframes also expect data that is organized as flat columns. It does not support nested JSON data very well (Bag is better for this).

Here we make a function to flatten down our nested data structure, map that across our records, and then convert that to a Dask Dataframe.

({'age': 39,
  'name': ['Wilfredo', 'Silva'],
  'occupation': 'Farmer',
  'telephone': '(489) 718-1463',
  'address': {'address': '760 Russia Drive', 'city': 'Allen Park'},
  'credit-card': {'number': '4890 7910 1700 3335',
   'expiration-date': '05/23'}},)
def flatten(record):
    return {
        'age': record['age'],
        'occupation': record['occupation'],
        'telephone': record['telephone'],
        'credit-card-number': record['credit-card']['number'],
        'credit-card-expiration': record['credit-card']['expiration-date'],
        'name': ' '.join(record['name']),
        'street-address': record['address']['address'],
        'city': record['address']['city']
({'age': 39,
  'occupation': 'Farmer',
  'telephone': '(489) 718-1463',
  'credit-card-number': '4890 7910 1700 3335',
  'credit-card-expiration': '05/23',
  'name': 'Wilfredo Silva',
  'street-address': '760 Russia Drive',
  'city': 'Allen Park'},)
df =
age occupation telephone credit-card-number credit-card-expiration name street-address city
0 39 Farmer (489) 718-1463 4890 7910 1700 3335 05/23 Wilfredo Silva 760 Russia Drive Allen Park
1 25 Chandler +1-(404)-661-5342 3452 021586 06729 07/25 Erline Sloan 234 Shaw Brae Wyoming
2 27 Lumberjack (793) 041-7129 4001 7633 2769 3515 12/25 Marcel Morgan 92 Forest Hill Gate Santa Monica
3 22 Legal Assistant 431-149-6381 2413 7521 1311 8833 03/20 Verlene Marquez 386 Latona Terrace Tucson
4 50 Research Director (431) 111-2477 4142 6029 2679 1462 07/17 Spencer Wood 1375 Eldridge Park Coachella

We can now perform the same computation as before, but now using Pandas and Dask dataframe.

df[df.age > 30].occupation.value_counts().nlargest(10).compute()
Hairdresser              18
Psychoanalyst            15
Kennel Hand              15
Marketing Coordinator    14
Undertaker               14
Plant Manager            14
Technical Instructor     14
Bank Messenger           14
Travel Clerk             13
Training Co-ordinator    13
Name: occupation, dtype: int64

Learn More

You may be interested in the following links: