GLOM TensorFlow 
This Python package attempts to implement GLOM in TensorFlow, which allows advances made by several different groups transformers, neural fields, contrastive representation learning, distillation and capsules to be combined. This was suggested by Geoffrey Hinton in his paper "How to represent part-whole hierarchies in a neural network".
Further, Yannic Kilcher's video and Phil Wang's repo was very helpful for me to implement this project.
Installation
Run the following to install:
pip install glom-tf
Developing glom-tf
To install glom-tf, along with tools you need to develop and test, run the following in your virtualenv:
git clone https://github.com/Rishit-dagli/GLOM-TensorFlow.git
# or clone your own fork
cd GLOM-TensorFlow
pip install -e .[dev]
A bit about GLOM
The GLOM architecture is composed of a large number of columns which all use exactly the same weights. Each column is a stack of spatially local autoencoders that learn multiple levels of representation for what is happening in a small image patch. Each autoencoder transforms the embedding at one level into the embedding at an adjacent level using a multilayer bottom-up encoder and a multilayer top-down decoder. These levels correspond to the levels in a part-whole hierarchy.
Interactions among the 3 levels in one column
An example shared by the author was as an example when show a face image, a single column might converge on embedding vectors representing a nostril, a nose, a face, and a person.
At each discrete time and in each column separately, the embedding at a level is updated to be the weighted average of:
- bottom-up neural net acting on the embedding at the level below at the previous time
- top-down neural net acting on the embedding at the level above at the previous time
- embedding vector at the previous time step
- attention-weighted average of the embeddings at the same level in nearby columns at the previous time
For a static image, the embeddings at a level should settle down over time to produce similar vectors.
A picture of the embeddings at a particular time
Usage
from glomtf import Glom
model = Glom(dim = 512,
levels = 5,
image_size = 224,
patch_size = 14)
img = tf.random.normal([1, 3, 224, 224])
levels = model(img, iters = 12) # (1, 256, 5, 12)
# 1 - batch
# 256 - patches
# 5 - levels
# 12 - dimensions
Use the return_all = True argument to get all the column and level states per iteration. This also gives you access to all the level data across iterations for clustering, from which you can inspect the islands too.
from glomtf import Glom
model = Glom(dim = 512,
levels = 5,
image_size = 224,
patch_size = 14)
img = tf.random.normal([1, 3, 224, 224])
all_levels = model(img, iters = 12, return_all = True) # (13, 1, 256, 5, 12)
# 13 - time
# top level outputs after iteration 6
top_level_output = all_levels[7, :, :, -1] # (1, 256, 512)
# 1 - batch
# 256 - patches
# 512 - dimensions
Want to Contribute
๐โโ๏ธ
?
Awesome! If you want to contribute to this project, you're always welcome! See Contributing Guidelines. You can also take a look at open issues for getting more information about current or upcoming tasks.
Want to discuss?
๐ฌ
Have any questions, doubts or want to present your opinions, views? You're always welcome. You can start discussions.
Citations
@misc{hinton2021represent,
title = {How to represent part-whole hierarchies in a neural network},
author = {Geoffrey Hinton},
year = {2021},
eprint = {2102.12627},
archivePrefix = {arXiv},
primaryClass = {cs.CV}
}
License
Copyright 2020 Rishit Dagli
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.
12 Aug 6, 2022
8 Dec 11, 2022
46 Dec 11, 2022
62 Jan 3, 2023
46 Dec 14, 2022
26 Nov 23, 2022
22 Nov 30, 2022
1.6k Jan 6, 2023
0 Jul 20, 2022
![[ImgBot] Optimize images](https://avatars.githubusercontent.com/in/4706?v=4)
16 Dec 26, 2022
827 Dec 29, 2022
9 May 10, 2022
7 Nov 07, 2022
11 Dec 17, 2021
14 Dec 17, 2021
422 Dec 28, 2022
268 Jan 09, 2023
6 Dec 31, 2022
11 Sep 25, 2022
253 Jan 01, 2023
6 Dec 29, 2022
287 Nov 25, 2022
339 Oct 27, 2022
52 Nov 10, 2022
49 Sep 21, 2022
7 Mar 28, 2022
66 Nov 30, 2022
3 Jan 26, 2022
127 Dec 31, 2022