Real-time collaboration for Jupyter Notebooks, Linux Terminals, LaTeX, VS Code, R IDE, and more,
all in one place. Commercial Alternative to JupyterHub.

Views: ⁵⁹²

Kernel: Python 3 (Ubuntu Linux)

Dask in Python 3 (Ubuntu Linux)

https://docs.dask.org/en/latest/

In [1]:

import dask
dask.__version__

Out[1]:

'1.0.0'

In [2]:

import dask
import dask.distributed
import os
dask.config.set({
    'temporary_directory': os.path.expanduser('~/tmp'),
    'scheduler.work-stealing': True
})

Out[2]:

<dask.config.set at 0x7f063001f748>

In [3]:

dask.config.config

Out[3]:

{'logging': {'distributed': 'info',
  'distributed.client': 'warning',
  'bokeh': 'critical',
  'tornado': 'critical',
  'tornado.application': 'error'},
 'require-encryption': False,
 'client-heartbeat-interval': 5000,
 'distributed': {'version': 2,
  'scheduler': {'allowed-failures': 3,
   'bandwidth': 100000000,
   'default-data-size': 1000,
   'transition-log-length': 100000,
   'work-stealing': True,
   'worker-ttl': None,
   'preload': [],
   'preload-argv': []},
  'worker': {'multiprocessing-method': 'forkserver',
   'use-file-locking': True,
   'connections': {'outgoing': 50, 'incoming': 10},
   'preload': [],
   'preload-argv': [],
   'profile': {'interval': 10, 'cycle': 1000},
   'memory': {'target': 0.6, 'spill': 0.7, 'pause': 0.8, 'terminate': 0.95}},
  'client': {'heartbeat': '5s'},
  'comm': {'compression': 'auto',
   'default-scheme': 'tcp',
   'socket-backlog': 2048,
   'recent-messages-log-length': 0,
   'timeouts': {'connect': 3, 'tcp': 30}},
  'dashboard': {'link': 'http://{host}:{port}/status', 'export-tool': False},
  'admin': {'tick': {'interval': 20, 'limit': 1000},
   'log-length': 10000,
   'log-format': '%(name)s - %(levelname)s - %(message)s',
   'pdb-on-err': False}},
 'temporary-directory': '/home/user/tmp',
 'scheduler': {'work-stealing': True}}

In [4]:

from dask.distributed import Client
client = Client('127.0.0.1:8786')
client

Out[4]:

The dashboard is actually at https://cocalc.com/{{ THE PROJECT UUID }}/server/8787/status

Websocket forwarding doesn't work, though ... hmm...

alternatively, start an X11 desktop in cocalc and run google-chrome at http://127.0.0.1:8787/status

data array similar to numpy arrays

In [5]:

import dask.array as da

In [6]:

x = da.random.random((3000, 3000), chunks=(300, 300))
x

Out[6]:

dask.array<random_sample, shape=(3000, 3000), dtype=float64, chunksize=(300, 300)>

In [7]:

y = x + x.T
z = y[::50, 1000:].mean(axis=1)
z

Out[7]:

dask.array<mean_agg-aggregate, shape=(60,), dtype=float64, chunksize=(6,)>

In [8]:

z.shape

Out[8]:

(60,)

In [9]:

out = z.compute()
out

Out[9]:

array([1.00894334, 0.98686389, 1.00412201, 1.00562101, 0.9984469 ,
       1.00773417, 1.0017398 , 0.99675565, 0.99054068, 0.98704365,
       0.99406761, 1.01857855, 1.00252324, 1.00853879, 0.98123995,
       0.99701221, 0.99059657, 0.99144697, 1.00628833, 0.9947498 ,
       0.99845961, 1.01338604, 1.00216002, 0.99366811, 0.99538552,
       1.03424335, 1.00016978, 0.99584787, 0.99663896, 1.0028844 ,
       0.99098811, 0.9955956 , 0.99308403, 1.01371317, 1.0229804 ,
       0.99350163, 1.00235196, 1.00826049, 0.9949777 , 0.98840276,
       1.00248748, 1.01109067, 1.01085646, 1.01437783, 1.00604058,
       0.99871721, 1.00370476, 0.99210787, 1.00258028, 0.99557382,
       1.0012086 , 0.99908451, 0.98371352, 1.00652402, 0.9849942 ,
       1.0135658 , 0.98824092, 0.99474543, 1.00656597, 0.9921391 ])

In [10]:

(z[:100].sum() / z[:100].shape[0]).compute()

Out[10]:

1.0002978610915068

functions and native lists

In [11]:

fn = lambda x : x**3 - x**2 + 1

def neg(a):
    import time
    time.sleep(.1)
    return -a

A = client.map(fn, range(10))
B = client.map(neg, A)
C = client.map(lambda a, b: a+b, A, B)
total = client.submit(sum, B)
total.result()

Out[11]:

-1750

In [12]:

client.gather(C)

Out[12]:

[0, 0, 0, 0, 0, 0, 0, 0, 0, 0]

loops?

In [13]:

def fib(x):
    if x <= 1:
        return 1
    else:
        return fib(x - 2) + fib(x - 1)

In [14]:

[fib(_) for _ in range(10)]

Out[14]:

[1, 1, 2, 3, 5, 8, 13, 21, 34, 55]

Dask Bags

In [15]:

import dask.bag as db
b1 = db.from_sequence(range(-1000, 1000), npartitions=50)
b1

Out[15]:

dask.bag<from_se..., npartitions=50>

In [16]:

import operator

In [17]:

is_odd = lambda x : x % 2 == 0

#b1.groupby(is_odd).map(lambda k_v : (k_v[0], sum(k_v[1]))).compute()
b1.foldby(is_odd, operator.add, 0).compute()

Out[17]:

[(True, -1000), (False, 0)]

Dask Delayed

In [18]:

from dask import delayed

In [19]:

inc = lambda x : x+1
from operator import add

In [20]:

z = delayed(0)
for i in range(5):
    x = delayed(inc)(i)
    y = delayed(inc)(delayed(add)(i, x))
    z = delayed(add)(z, y)
z.compute()

Out[20]:

30

In [21]:

z.vizualize(filename='dask-delayed-1.svg')

Out[21]:

Delayed('vizualize-38f8c38e-2766-4979-aa1d-a76835e14c0f')

Dask ML

https://dask-ml.readthedocs.io/en/latest/

In [22]:

import dask_ml

In [23]:

from dask_ml.preprocessing import Categorizer, DummyEncoder
from dask_ml.linear_model import LogisticRegression

In [24]:

lr = LogisticRegression()
lr

Out[24]:

LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,
          intercept_scaling=1.0, max_iter=100, multiclass='ovr', n_jobs=1,
          penalty='l2', random_state=None, solver='admm',
          solver_kwargs=None, tol=0.0001, verbose=0, warm_start=False)

In [25]:

from sklearn.externals.joblib import parallel_backend
with parallel_backend('dask') as pb:
    print(pb[0])

Out[25]:

<sklearn.externals.joblib._dask.DaskDistributedBackend object at 0x7f0611245d68>

In [0]:

In [0]:

In [0]:

In [0]:

Real-time collaboration for Jupyter Notebooks, Linux Terminals, LaTeX, VS Code, R IDE, and more,
all in one place. Commercial Alternative to JupyterHub.

Dask in Python 3 (Ubuntu Linux)

data array similar to numpy arrays

functions and native lists

loops?

Dask Bags

Dask Delayed

Dask ML

Product

Resources

Company

Real-time collaboration for Jupyter Notebooks, Linux Terminals, LaTeX, VS Code, R IDE, and more, all in one place. Commercial Alternative to JupyterHub.

Dask in Python 3 (Ubuntu Linux)

data array similar to numpy arrays

functions and native lists

loops?

Dask Bags

Dask Delayed

Dask ML

Real-time collaboration for Jupyter Notebooks, Linux Terminals, LaTeX, VS Code, R IDE, and more,
all in one place. Commercial Alternative to JupyterHub.