vowpal_porpoise
Lightweight python wrapper for vowpal_wabbit.
Why: Scalable, blazingly fast machine learning.
Install
- Install vowpal_wabbit. Clone and run
make - Install cython.
pip install cython - Clone vowpal_porpoise
- Run:
python setup.py installto install.
Now can you do: import vowpal_porpoise from python.
Examples
Standard Interface
Linear regression with l1 penalty:
from vowpal_porpoise import VW
# Initialize the model
vw = VW(moniker='test', # a name for the model
passes=10, # vw arg: passes
loss='quadratic', # vw arg: loss
learning_rate=10, # vw arg: learning_rate
l1=0.01) # vw arg: l1
# Inside the with training() block a vw process will be
# open to communication
with vw.training():
for instance in ['1 |big red square',\
'0 |small blue circle']:
vw.push_instance(instance)
# here stdin will close
# here the vw process will have finished
# Inside the with predicting() block we can stream instances and
# acquire their labels
with vw.predicting():
for instance in ['1 |large burnt sienna rhombus',\
'0 |little teal oval']:
vw.push_instance(instance)
# Read the predictions like this:
predictions = list(vw.read_predictions_())
L-BFGS with a rank-5 approximation:
from vowpal_porpoise import VW
# Initialize the model
vw = VW(moniker='test_lbfgs', # a name for the model
passes=10, # vw arg: passes
lbfgs=True, # turn on lbfgs
mem=5) # lbfgs rank
Latent Dirichlet Allocation with 100 topics:
from vowpal_porpoise import VW
# Initialize the model
vw = VW(moniker='test_lda', # a name for the model
passes=10, # vw arg: passes
lda=100, # turn on lda
minibatch=100) # set the minibatch size
Scikit-learn Interface
vowpal_porpoise also ships with an interface into scikit-learn, which allows awesome experiment-level stuff like cross-validation:
from sklearn.cross_validation import StratifiedKFold
from sklearn.grid_search import GridSearchCV
from vowpal_porpoise.sklearn import VW_Classifier
GridSearchCV(
VW_Classifier(loss='logistic', moniker='example_sklearn',
passes=10, silent=True, learning_rate=10),
param_grid=parameters,
score_func=f1_score,
cv=StratifiedKFold(y_train),
).fit(X_train, y_train)
Check out example_sklearn.py for more details
Library Interace (DISABLED as of 2013-08-12)
Via the VW interface:
with vw.predicting_library():
for instance in ['1 |large burnt sienna rhombus', \
'1 |little teal oval']:
prediction = vw.push_instance(instance)
Now the predictions are returned directly to the parent process, rather than having to read from disk. See examples/example1.py for more details.
Alternatively you can use the raw library interface:
import vw_c
vw = vw_c.VW("--loss=quadratic --l1=0.01 -f model")
vw.learn("1 |this is a positive example")
vw.learn("0 |this is a negative example")
vw.finish()
Currently does not support passes due to some limitations in the underlying vw C code.
Need more examples?
- example1.py: SimpleModel class wrapper around VP (both standard and library flavors)
- example_library.py: Demonstrates the low-level vw library wrapper, classifying lines of alice in wonderland vs through the looking glass.
Why
vowpal_wabbit is insanely fast and scalable. vowpal_porpoise is slower, but only during the initial training pass. Once the data has been properly cached it will idle while vowpal_wabbit does all the heavy lifting. Furthermore, vowpal_porpoise was designed to be lightweight and not to get in the way of vowpal_wabbit's scalability, e.g. it allows distributed learning via --nodes and does not require data to be batched in memory. In our research work we use vowpal_porpoise on an 80-node cluster running over multiple terabytes of data.
The main benefit of vowpal_porpoise is allowing rapid prototyping of new models and feature extractors. We found that we had been doing this in an ad-hoc way using python scripts to shuffle around massive gzipped text files, so we just closed the loop and made vowpal_wabbit a python library.
How it works
Wraps the vw binary in a subprocess and uses stdin to push data, temporary files to pull predictions. Why not use the prediction labels vw provides on stdout? It turns out that the python GIL basically makes streamining in and out of a process (even asynchronously) painfully difficult. If you know of a clever way to get around this, please email me. In other languages (e.g. in a forthcoming scala wrapper) this is not an issue.
Alternatively, you can use a pure api call (vw_c, wrapping libvw) for prediction.
Contact
Joseph Reisinger @josephreisinger
Contributors
- Austin Waters ([email protected])
- Joseph Reisinger ([email protected])
- Daniel Duckworth ([email protected])
License
Apache 2.0
0 Sep 06, 2022
32 Oct 10, 2022
111 Jan 08, 2023
554 Dec 30, 2022
315 Jan 01, 2023
3 Feb 15, 2022
16 Dec 23, 2022
326 Jan 02, 2023
329 Jan 03, 2023
74 Dec 12, 2022
75 Nov 11, 2022
94 Dec 18, 2022
3 Aug 20, 2022
4 Sep 15, 2022
46 Dec 29, 2022
1 Dec 11, 2021
3 Dec 15, 2021
1.5k Jan 02, 2023
38 Dec 28, 2022
1 Dec 09, 2021