Skip to main content

A fastai based framework for similarity learning

Project description

similarity-learning

Docs

See https://irad-zehavi.github.io/similarity-learning/

Install

pip install similarity_learning

How to use

As an nbdev library, similarity_learning supports import * (without importing unwanted symbols):

from similarity_learning.all import *

Now we can train a pair-matcher. First let’s construct dataloaders of pairs:

from fastai.vision.all import *
pairs = Pairs(Imagenette(160), .1)
dls = pairs.dls(after_item=Resize(128),
                after_batch=Normalize.from_stats(*imagenet_stats))

To get quick results, we can use the body of a pretrained model as a backbone for our Siamese neural network:

classifier = resnet34(weights=ResNet34_Weights.DEFAULT)
siamese = ThresholdSiamese(create_body(model=classifier, cut=-1)).to(dls.device)
siamese.fit_threshold(dls.train)
(1.0099999904632568, 0.8962054252624512)

Let’s see how good it is:

learn = Learner(dls, siamese, metrics=accuracy)
learn.validate()
(#2) [0.5453092455863953,0.8877550959587097]
learn.show_results()

Not bad, but we can do better with finetuning:

learn.fit(5, 1e-4)
learn.validate()
(#2) [0.26150667667388916,0.954081654548645]
learn.show_results()

We can also consider the distribution of feature-space distances compared to the decision threshold:

siamese.plot_distance_histogram(dls.valid)

See the rest of the docs for more examples, including more visualizations, comparison of loss functions, and facial recognition.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

similarity-learning-0.0.4.tar.gz (22.5 kB view details)

Uploaded Source

Built Distribution

similarity_learning-0.0.4-py3-none-any.whl (22.6 kB view details)

Uploaded Python 3

File details

Details for the file similarity-learning-0.0.4.tar.gz.

File metadata

  • Download URL: similarity-learning-0.0.4.tar.gz
  • Upload date:
  • Size: 22.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.9.16

File hashes

Hashes for similarity-learning-0.0.4.tar.gz
Algorithm Hash digest
SHA256 951dd1c853449b9f1798aa73a6c8a8e56c3c6c9281c26ee009924f63158489f9
MD5 32d00f72e8da7ca23fd97f96dcca6bcc
BLAKE2b-256 9e22e31b169acd14a3be1552444275cdabc484480af5a2463e070844d52aa92b

See more details on using hashes here.

File details

Details for the file similarity_learning-0.0.4-py3-none-any.whl.

File metadata

File hashes

Hashes for similarity_learning-0.0.4-py3-none-any.whl
Algorithm Hash digest
SHA256 4000cb5b013803b8df9c9bf4dee6e5c712e725e194ceaa254ea6f1add1553c14
MD5 5da32a5d57814b35396f1bbcaf78c6fa
BLAKE2b-256 a37daa220fe06e5b24833c33866706bdde1e7cc6e6dc057dbe943933606dc66b

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page