Skip to main content

benderthon

Set of utilities to work easier with Bender.

Currently there’s support for TensorFlow and Caffe, but we are working on more stuff!

Works on Python 2.7.+ and 3.+, with Tensorflow 1.2+.

To install:

pip install benderthon

TensorFlow is required too. The simplest way to install it is:

pip install tensorflow

There are other ways, see Installing Tensorflow. Benderthon does not install it by default to let the usage of a custom installation.

tf-freeze

Utility to convert TensorFlow checkpoints into minimal frozen graphs.

Usage

To take the checkpoint in checkpoint_path.ckpt, whose output is yielded by the node named Tanh, and save it to graph_with_weights.pb:

benderthon tf-freeze checkpoint_path.ckpt graph_with_weights.pb Tanh

Sample

The file sample.py contains a network example for MNIST dataset with 2 convolutional layers and 2 dens layers. If you run it, it will generate checkpoints files with prefix checkpoints/mnist.ckpt:

./sample.py

Then you can get a minimal protobuf version with the weights frozen:

benderthon tf-freeze checkpoints/mnist.ckpt output/mnist.pb Prediction

The generated file occupies half the original checkpoints (26MB to 13MB).

You can also get only the graph, which occupies just 13kB:

benderthon tf-freeze --no-weights checkpoints/mnist.ckpt output/mnist_only_graph.pb Prediction

To save the weights in a separate path for later processing:

benderthon tf-freeze --only-weights checkpoints/mnist.ckpt weights/ Prediction

caffe-freeze

This module cannot be accessed from the command line utility, it should be used from Python code, importing benderthon.caffe_freeze.

You need caffeflow package installed first:

pip install -e git://github.com/xmartlabs/caffeflow.git@4618f89#egg=caffeflow

Development

This utility is under development and the API is not stable. So, do not heavily rely on it.

To install locally you should do ./setup.py install, but first have pandoc and pypandoc installed.

License

Copyright 2017 Xmartlabs SRL.

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

   http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.

Metadata

Release files for benderthon 0.3.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Built distribution (wheel)

Table of built distributions (wheels) for benderthon 0.3.0
File Interpreter ABI Platform Actions
benderthon-0.3.0-py2.py3-none-any.whl Python 3, Python 2 none any Details for benderthon-0.3.0-py2.py3-none-any.whl

Release files / benderthon-0.3.0-py2.py3-none-any.whl

Download URL benderthon-0.3.0-py2.py3-none-any.whl
Size 14.0 kB
Tags Python 2 Python 3
SHA-256 checksum
How to use checksums
bd02b42bccb5ba2e00e6518036449e2c17c784518a6dd358dd4b69f83a8f24f0
BLAKE2b-256 checksum
How to use checksums
d6410388ce95d9e8ec6c452c4af2d259f18a1fec9f4318250379fd28164b34f3
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No

Security for benderthon 0.3.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page