Neptune-fastai Integration

The integration enables you to log fast.ai metrics to Neptune.

Requirements

Integration with the fastai framework is enabled as a part of neptune-contrib, an open source project curated by the Neptune team.

Install fast.ai before you continue. For more information, check the docs.

pip install neptune-contrib
pip install neptune-contrib[monitoring]

Create your databunch

from fastai.vision import *
path = untar_data(URLs.MNIST_TINY)

data = ImageDataBunch.from_folder(path, ds_tfms=(rand_pad(2, 28), []), bs=64)
data.normalize(imagenet_stats)

Create the learner, find your optimal learning rate, and plot it

learn = cnn_learner(data, models.resnet18, metrics=accuracy)
learn.lr_find()
learn.recorder.plot()

Create an experiment and add the neptune_monitor callback

import neptune
from neptunecontrib.monitoring.fastai import NeptuneMonitor

neptune.init(project_qualified_name='USER_NAME/PROJECT_NAME')

with neptune.create_experiment(params={'lr': 1e-2}):
    learn = cnn_learner(data, models.resnet18, metrics=accuracy,
                        callback_fns=[NeptuneMonitor])
    learn.fit_one_cycle(20, 1e-2)

Monitor your fastai training in Neptune

Now you can watch your fastai model training in Neptune!

Full fastai monitor script

Simply copy and paste it to fastai_example.py and run.

from fastai.vision import *
import neptune
from neptunecontrib.monitoring.fastai import NeptuneMonitor

neptune.init(project_qualified_name='USER_NAME/PROJECT_NAME')

path = untar_data(URLs.MNIST_TINY)

data = ImageDataBunch.from_folder(path, ds_tfms=(rand_pad(2, 28), []), bs=64)
data.normalize(imagenet_stats)

learn = cnn_learner(data, models.resnet18, metrics=accuracy)
learn.lr_find()
learn.recorder.plot()

with neptune.create_experiment(params={'lr': 1e-2}):
    learn = cnn_learner(data, models.resnet18, metrics=accuracy,
                        callback_fns=[NeptuneMonitor])
    learn.fit_one_cycle(20, 1e-2)