Overview

ALP helps you experiment with a lot of machine learning models quickly. It provides you with a simple way of scheduling and recording experiments.

This library has been developped to work well with Keras and Scikit-learn but can suit a lot of other frameworks.

Documentation

http://python-alp.readthedocs.io/

Quickstart

Clone the repo and install the library:

git clone https://github.com/tboquet/python-alp.git
cd python-alp
python setup.py install

Install the Command Line Interface dependencies:

cd req
pip install -r requirements_cli.txt

Generate a base configuration using an absolute path:

alp --verbose genconfig --outdir=/path/to/a/directory --cpu

Launch the services:

alp --verbose service start /path/to/a/directory/.alp/containers.json

Check the status of your containers:

alp --verbose status /path/to/a/directory/.alp/containers.json

Log in to the Jupyter notebook you just launched in your browser @ localhost:440 using the password default.

Launch some experiments!

# we import numpy and fix the seed
import numpy as np
np.random.seed(1337)  # for reproducibility

# we import alp and Keras tools that we will use
import alp
from keras.datasets import mnist
from keras.models import Sequential
from keras.layers import Dense, Dropout, Activation, Flatten
from keras.utils import np_utils
import keras.backend as K
from keras.optimizers import Adam
from alp.appcom.ensembles import HParamsSearch

# if you use tensorflow you must use this configuration
# so that it doesn't use all of the GPU's memory (default config)
import tensorflow as tf

config = tf.ConfigProto(allow_soft_placement=True)
config.gpu_options.allow_growth = True
session = tf.Session(config=config)
K.set_session(session)

batch_size = 128
nb_classes = 10
nb_epoch = 12

# input image dimensions
img_rows, img_cols = 28, 28
# number of features to use
nb_features = 32

# the data, shuffled and split between train and test sets
(X_train, y_train), (X_test, y_test) = mnist.load_data()

X_train = X_train.astype('float32')
X_test = X_test.astype('float32')
X_train /= 255
X_test /= 255
print('X_train shape:', X_train.shape)
print(X_train.shape[0], 'train samples')
print(X_test.shape[0], 'test samples')

if K.image_dim_ordering() == 'th':
    X_train = X_train.reshape(X_train.shape[0], 1, img_rows, img_cols)
    X_test = X_test.reshape(X_test.shape[0], 1, img_rows, img_cols)
    input_shape = (1, img_rows, img_cols)
else:
    X_train = X_train.reshape(X_train.shape[0], img_rows, img_cols, 1)
    X_test = X_test.reshape(X_test.shape[0], img_rows, img_cols, 1)
    input_shape = (img_rows, img_cols, 1)

# convert class vectors to binary class matrices
Y_train = np_utils.to_categorical(y_train, nb_classes)
Y_test = np_utils.to_categorical(y_test, nb_classes)

# put the data in the form ALP expects
data, data_val = dict(), dict()
data["X"] = X_train[:500]
data["y"] = Y_train[:500]
data_val["X"] = X_test[:500]
data_val["y"] = Y_test[:500]

# Define and compile the model

model = Sequential()

model.add(Flatten(input_shape=input_shape))
model.add(Dense(nb_features))
model.add(Activation('relu'))
model.add(Dropout(0.25))

model.add(Dense(128))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Dense(nb_classes))
model.add(Activation('softmax'))

model.compile(loss='categorical_crossentropy',
              optimizer='adadelta',
              metrics=['accuracy'])

# Define you experiment

from alp.appcom.core import Experiment

expe = Experiment(model)

# Fit the model linked to your experiment
results = expe.fit([data], [data_val], nb_epoch=2, batch_size=batch_size)

# Predict using your model
expe.predict(data['X'])

Get started with the tutorial series!

• Free software: Apache license