Skip to main content

Official Losswise library for Python

Project description

This is the official Losswise Python library. This library allows for server-side integration of Losswise.

Installation

The library can be installed using pip:

pip install losswise

Getting Started

First create an account on the Losswise website (https://losswise.com). This will automatically generate a unique API key.

Typical usage usually looks like this:

import random
import losswise

# replace with your own api key
losswise.set_api_key('your_api_key')

# replace with a string that identifies your model
session = losswise.Session(tag='my_dilated_convnet', max_iter=10, data={'num_params': 10000000})

# create empty graph for loss, keep track of minima here hence kind='min'
graph = session.Graph(title='loss', kind='min')

# track artificial loss over time
for x in xrange(10):
    train_loss = 1. / (0.1 + x + 0.1 * random.random())
    test_loss = 1.5 / (0.1 + x + 0.2 * random.random())
    graph.append(x, {'train_loss': train_loss, 'test_loss': test_loss})

# mark session as complete
session.done()

You can then view the visualization results on your dashboard.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

losswise-0.99.tar.gz (4.3 kB view hashes)

Uploaded Source

Built Distribution

losswise-0.99-py2.py3-none-any.whl (6.7 kB view hashes)

Uploaded Python 2 Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page