by asappresearch

asappresearch / flambe

An ML framework to accelerate research and its path to production.

224 Stars 29 Forks Last release: 6 months ago (v0.4.18) MIT License 661 Commits 19 Releases

Available items

No Items, yet!

The developer of this repository has not created any items for sale yet. Need a bug fixed? Help with integration? A different license? Create a request here:

.. raw:: html



.. image:: https://github.com/asappresearch/flambe/workflows/Run%20fast%20tests/badge.svg :target: https://github.com/asappresearch/flambe/actions :alt: Fast tests

.. image:: https://github.com/asappresearch/flambe/workflows/Run%20slow%20tests/badge.svg :target: https://github.com/asappresearch/flambe/actions :alt: Slow tests

.. image:: https://readthedocs.org/projects/flambe/badge/?version=latest :target: https://flambe.ai/en/latest/?badge=latest :alt: Documentation Status

.. image:: https://badge.fury.io/py/flambe.svg :target: https://badge.fury.io/py/flambe :alt: PyPI version


Welcome to Flambé, a

_-based library that allows users to:
  • Run complex experiments with multiple training and processing stages
  • Search over hyperparameters, and select the best trials
  • Run experiments remotely over many workers, including full AWS integration
  • Easily share experiment configurations, results, and model weights with others




.. code-block:: bash

pip install flambe

From source:

.. code-block:: bash

git clone [email protected]:asappresearch/flambe.git
cd flambe
pip install .

Getting started

Define an


.. code-block:: yaml


name: sst-text-classification


stage 0 - Load the Stanford Sentiment Treebank dataset and run preprocessing

dataset: !SSTDataset # this is a simple Python object, and the arguments to build it transform: # these arguments are passed to the init method text: !TextField label: !LabelField

Stage 1 - Define a model

model: !TextClassifier embedder: !Embedder embedding: !torch.Embedding # automatically use pytorch classes num_embeddings: [email protected] dataset.text.vocab_size # link to other components, and attributes embedding_dim: 300 embedding_dropout: 0.3 encoder: !PooledRNNEncoder input_size: 300 n_layers: !g [2, 3, 4] # grid search over any parameters hidden_size: 128 rnn_type: sru dropout: 0.3 output_layer: !SoftmaxLayer input_size: [email protected] model[embedder][encoder].rnn.hidden_size # also use inner-links output_size: [email protected] dataset.label.vocab_size

Stage 2 - Train the model on the dataset

train: !Trainer dataset: [email protected] dataset model: [email protected] model train_sampler: !BaseSampler val_sampler: !BaseSampler loss_fn: !torch.NLLLoss metric_fn: !Accuracy optimizer: !torch.Adam params: [email protected] train[model].trainable_params max_steps: 100 iter_per_step: 100

Stage 3 - Eval on the test set

eval: !Evaluator dataset: [email protected] dataset model: [email protected] train.model metric_fn: !Accuracy eval_sampler: !BaseSampler

Define how to schedule variants

schedulers: train: !ray.HyperBandScheduler

All objects in the

are subclasses of
, which are automatically registered to be used with YAML. Custom
implementations must implement
to add custom behavior when being executed.

Now just execute:

.. code-block:: bash

flambe example.yaml

Note that defining objects like model and dataset ahead of time is optional; it's useful if you want to reference the same model architecture multiple times later in the pipeline.

Progress can be monitored via the Report Site (with full integration with Tensorboard):

.. raw:: html


  • Native support for hyperparameter search: using search tags (see
    in the example) users can define multi variant pipelines. More advanced search algorithms will be available in a coming release!
  • Remote and distributed experiments: users can submit
    which will execute in a distributed way. Full
    integration is supported.
  • Visualize all your metrics and meaningful data using Tensorboard: log scalars, histograms, images, hparams and much more.
  • Add custom code and objects to your pipelines: extend flambé functionality using our easy-to-use extensions mechanism.
  • Modularity with hierarchical serialization: save different components from pipelines and load them safely anywhere.

Next Steps

Full documentation, tutorials and much more in https://flambe.ai


You can reach us at [email protected]

We use cookies. If you continue to browse the site, you agree to the use of cookies. For more information on our use of cookies please see our Privacy Policy.