Skip to main content

osprey is an easy-to-use tool for hyperparameter optimization for machine learning algorithms in python using scikit-learn (or using scikit-learn compatible APIs).

Project description

osprey is an easy-to-use tool for hyperparameter optimization for machine learning algorithms in python using scikit-learn (or using scikit-learn compatible APIs).

Each osprey experiment combines an dataset, an estimator, a search space (and engine), cross validation and asynchronous serialization for distributed parallel optimization of model hyperparameters.

Example (with mixtape models/datasets)

$ cat config.yaml
  eval_scope: mixtape
  eval: |
        ('featurizer', DihedralFeaturizer(types=['phi', 'psi'])),
        ('cluster', MiniBatchKMeans()),
        ('msm', MarkovStateModel(n_timescales=5, verbose=False)),

    min: 10
    max: 100
    type: int
      - ['phi', 'psi']
      - ['phi', 'psi', 'chi1']
   type: enum

cv: 5

  name: mdtraj
    trajectories: ~/local/msmbuilder/Tutorial/XTC/*/*.xtc
    topology: ~/local/msmbuilder/Tutorial/native.pdb
    stride: 1

    uri: sqlite:///osprey-trials.db

Then run osprey worker. You can run multiple parallel instances of osprey worker simultaniously on a cluster too.

$ osprey worker config.yaml
= osprey is a tool for machine learning hyperparameter optimization. =

osprey version:  0.2_10_g18392d9_dirty-py2.7.egg
time:            October 27, 2014 10:44 PM
hostname:        dn0a230538.sunet
cwd:             /private/var/folders/yb/vpt17lxs67vf02qpvgvjrc5m0000gn/T/tmpDgBwlU
pid:             99407

Loading config file:     config.yaml...
Loading trials database: sqlite:///osprey-trials.db (table = "trials")...

Loading dataset...
  100 elements without labels
Instantiated estimator:
  Pipeline(steps=[('featurizer', DihedralFeaturizer(sincos=True, types=['phi', 'psi'])), ('tica', tICA(gamma=0.05, lag_time=1, n_components=4, weighted_transform=False)), ('cluster', MiniBatchKMeans(batch_size=100, compute_labels=True, init='k-means++',
        init_size=None, max_iter=100, max_no_improvement=...toff=1, lag_time=1, n_timescales=5, prior_counts=0,
         reversible_type='mle', verbose=False))])
Hyperparameter search space:
  featurizer__types         (enum)    choices = (['phi', 'psi'], ['phi', 'psi', 'chi1'])
  cluster__n_clusters       (int)         10 <= x <= 100

Beginning iteration                                              1 / 1
History contains: 0 trials
Choosing next hyperparameters with random...
  {'cluster__n_clusters': 20, 'featurizer__types': ['phi', 'psi']}

Fitting 5 folds for each of 1 candidates, totalling 5 fits
[Parallel(n_jobs=1)]: Done   1 jobs       | elapsed:    0.3s
[Parallel(n_jobs=1)]: Done   5 out of   5 | elapsed:    1.8s finished
Success! Model score = 4.080646
(best score so far   = 4.080646)

1/1 models fit successfully.
time:         October 27, 2014 10:44 PM
elapsed:      4 seconds.
osprey worker exiting.

You can dump the database to JSON or CSV with osprey dump.


# grab the latest version from github
$ pip install git+git://
# or clone the repo yourself and run ``
$ git clone
$ cd osprey && python install


  • six
  • pyyaml
  • numpy
  • scikit-learn
  • sqlalchemy
  • hyperopt (recommended, required for engine=hyperopt_tpe)
  • scipy (optional, for testing)
  • nose (optional, for testing)

On python2.6, the argparse and importlib backports are also required

Project details

Download files

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

Files for osprey, version 0.3
Filename, size File type Python version Upload date Hashes
Filename, size osprey-0.3.tar.gz (34.3 kB) File type Source Python version None Upload date Hashes View hashes

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page