Stochastic gradient routines for Theano
Python
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
docs
downhill Update util.log with better info/formatting. Jun 3, 2018
examples Update examples to work with float32 or float64. Jan 8, 2017
test
.coveragerc Add automated build/coverage. Jun 7, 2015
.gitignore Ignore generated docs. Mar 4, 2015
.travis.yml Try skipping the conda --update for py3.4/travis. Jan 8, 2017
LICENSE Update util.log with better info/formatting. Jun 3, 2018
README.rst Move parameter-preparation to common location. Jan 10, 2017
setup.cfg Update setup.cfg to use newer heading format. Jan 8, 2017
setup.py

README.rst

DOWNHILL

The downhill package provides algorithms for minimizing scalar loss functions that are defined using Theano.

Several optimization algorithms are included:

All algorithms permit the use of regular or Nesterov-style momentum as well.

Quick Start: Matrix Factorization

Let's say you have 100 samples of 1000-dimensional data, and you want to represent your data as 100 coefficients in a 10-dimensional basis. This is pretty straightforward to model using Theano: you can use a matrix multiplication as the data model, a squared-error term for optimization, and a sparse regularizer to encourage small coefficient values.

Once you have constructed an expression for the loss, you can optimize it with a single call to downhill.minimize:

import downhill
import numpy as np
import theano
import theano.tensor as TT

FLOAT = 'df'[theano.config.floatX == 'float32']

def rand(a, b):
    return np.random.randn(a, b).astype(FLOAT)

A, B, K = 20, 5, 3

# Set up a matrix factorization problem to optimize.
u = theano.shared(rand(A, K), name='u')
v = theano.shared(rand(K, B), name='v')
z = TT.matrix()
err = TT.sqr(z - TT.dot(u, v))
loss = err.mean() + abs(u).mean() + (v * v).mean()

# Minimize the regularized loss with respect to a data matrix.
y = np.dot(rand(A, K), rand(K, B)) + rand(A, B)

# Monitor during optimization.
monitors = (('err', err.mean()),
            ('|u|<0.1', (abs(u) < 0.1).mean()),
            ('|v|<0.1', (abs(v) < 0.1).mean()))

downhill.minimize(
    loss=loss,
    train=[y],
    patience=0,
    batch_size=A,                 # Process y as a single batch.
    max_gradient_norm=1,          # Prevent gradient explosion!
    learning_rate=0.1,
    monitors=monitors,
    monitor_gradients=True)

# Print out the optimized coefficients u and basis v.
print('u =', u.get_value())
print('v =', v.get_value())

If you prefer to maintain more control over your model during optimization, downhill provides an iterative optimization interface:

opt = downhill.build(algo='rmsprop',
                     loss=loss,
                     monitors=monitors,
                     monitor_gradients=True)

for metrics, _ in opt.iterate(train=[[y]],
                              patience=0,
                              batch_size=A,
                              max_gradient_norm=1,
                              learning_rate=0.1):
    print(metrics)

If that's still not enough, you can just plain ask downhill for the updates to your model variables and do everything else yourself:

updates = downhill.build('rmsprop', loss).get_updates(
    batch_size=A, max_gradient_norm=1, learning_rate=0.1)
func = theano.function([z], loss, updates=list(updates))
for _ in range(100):
    print(func(y))  # Evaluate func and apply variable updates.

More Information

Source: http://github.com/lmjohns3/downhill

Documentation: http://downhill.readthedocs.org

Mailing list: https://groups.google.com/forum/#!forum/downhill-users