patnicolas / floorplan

Evaluate various techniques to extract and organize information from a floor plan

Home Page:https://patricknicolas.blogspot.com

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Patrick Nicolas - Last update 08.04.2023

Topology Graph Neural Network for floor plans

The objective is this repository is to evaluate various techniques to extract and organize information from a floor plan

References

Data flow

Here are the steps to upload floor plan and generate procurement log:
1 User enters name and email address.
2 User uploads floor plan as PDF file.
3a System fires email to Selection.AI with user name, email and attached floor plan.
3b System notifies sender/user the floor plan has been received with an estimated completion date.
...
4 Company generates procurement log.
5 Systems fires email to user with attached procurement log.

Environment

Package Version
python 3.9.16
torch 2.0.1
openai 0.27.1
matplotlib 3.7.1
scikit-learn 1.2.2
numpy 1.24.3
pandas 2.0.2
langchain 0.0.2
polars 0.17.0
fastapi 0.97.0
uvicorn 0.22.0
requests 2.31.0
pydantic 1.10.9
jinja2 3.1.2.

Versions

Date Version
06.27.2023 0.1
08.04.2023 0.2

Deployment

Python modules

pip install -r requirements

aiohttp==3.8.4
aiosignal==1.3.1
annotated-types==0.5.0
anyio==3.7.1
async-timeout==4.0.2
attrs==23.1.0
certifi==2023.5.7
charset-normalizer==3.1.0
click==8.1.6
distlib==0.3.7
docopt==0.6.2
exceptiongroup==1.1.2
fastapi==0.100.1
filelock==3.12.2
frozenlist==1.3.3
h11==0.14.0
idna==3.4
Jinja2==3.1.2
joblib==1.2.0
MarkupSafe==2.1.3
multidict==6.0.4
pbr==5.11.1
pipreqs==0.4.13
platformdirs==3.10.0
pydantic==2.1.1
pydantic_core==2.4.0
python-multipart==0.0.6
requests==2.31.0
sniffio==1.3.0
starlette==0.27.0
stevedore==5.1.0
threadpoolctl==3.1.0
tqdm==4.65.0
typing_extensions==4.7.1
uvicorn==0.23.2
virtualenv==20.24.2
virtualenv-clone==0.5.7
virtualenvwrapper==4.8.4
yarg==0.1.9
yarl==1.9.2
boto3~=1.28.18
cryptography~=41.0.3
argparse~=1.4.0
jmespath~=1.0.1
beautifulsoup4~=4.12.2
soupsieve~=2.4.1
cffi~=1.15.1
future~=0.18.2
Pillow~=9.5.0
pip~=23.1.2
configparser~=6.0.0
wheel~=0.38.4
docutils~=0.20.1
auth~=0.5.3
tornado~=6.3.2
botocore~=1.31.18
httplib2~=0.22.0
six~=1.15.0
wrapt~=1.15.0
PyYAML~=6.0
pytz~=2023.3
s3transfer~=0.6.1
Cython~=3.0.0
setuptools~=65.5.1
sympy~=1.12
mpmath~=1.3.0
selenium~=2.53.6
Unidecode~=1.3.6
SQLAlchemy~=2.0.15
python-dateutil~=2.8.2
numexpr~=2.8.4
fsspec~=2023.6.0
gunicorn~=21.2.0
greenlet~=2.0.2
networkx~=3.1
tenacity~=8.2.2
contourpy~=1.0.7
fonttools~=4.39.4
packaging~=23.1
tiktoken~=0.4.0
regex~=2023.6.3
pycparser~=2.21
pyparsing~=3.0.9
cycler~=0.11.0
kiwisolver~=1.4.4
zipp~=3.16.2
dl~=0.1.0
marshmallow~=3.19.0
toml~=0.10.2
Brotli~=1.0.9
keyring~=24.2.0
urllib3~=1.25.4
dnspython~=2.4.1
rsa~=4.9
pyasn1~=0.5.0
falcon~=3.1.1
Werkzeug~=2.3.6
cachetools~=5.3.1
google-api-python-client~=2.95.0
google-api-core~=2.11.1
google-auth~=2.22.0
google-auth-httplib2~=0.1.0
google-auth-oauthlib~=1.0.0
googleapis-common-protos~=1.60.0
blinker~=1.6.2
eventlet~=0.33.3
oauthlib~=3.2.2
uritemplate~=4.1.1

Deployment/hosting

The minimum requirements are support for FastAPI Web interface and PostgreSQL Heroku/Production-standard

radiant-thicket-07669
Auto Cert Mgmt: true
Dynos: web: 1
Git URL: https://git.heroku.com/radiant-thicket-07669.git
Owner: pnicolas57@yahoo.com
Region: us
Repo Size: 4 MB
Slug Size: 159 MB
Stack: heroku-22
Web URL: https://radiant-thicket-07669-41bc01837f36.herokuapp.com/

Step 1: Review/validate requirement.txt.

Step 2: Specify Python version

echo python-3.9.6 > runtime.txt

Step 3: Set configuration variable

heroku config
heroku config:set TOKEN= ...
heroku config
heroku config:unset TOKEN // If needed.

Step 4: update GIT repository.

git add // if necessary.
git commit -m ‘…’ .
git push

Step 5: Build, launch and monitor application.

git push heroku main
heroku open
heroku logs --tail

Step 6: HTTPS support.

heroku certs:auto:enable –a app-name.
heroku certs:auto –a app-name.

AI models

There are several options to generate a bill of material from floor plan:

Graph convolutional neural networks

Direct image processing using a sequence of convolutional neural network and graph convolutional neural networks:
Modeling sequence

Generative AI

Convert grayscale and predefined shape of pixels into vocabulary for fine tuning of a large language model such as GPT or LLMA

Spreadsheet itemization

The concept is to match patterns in the floor plan to a spreadsheet entry (row). The unique spreadsheet entries, partially defined become the labels for training the label.
The objective is to find graphic pattern that match each entry in the spreadsheet.

Spreadsheet itemization

Neural Optical Understanding

From a Meta paper: https://arxiv.org/abs/2308.13418v1

Copyright

Copyright notice

Todo list as of 08.07.2023

  • Redesign/refresh GUI.
  • Test with various browsers.
  • Switch to HTTPS - Set up but failed "CDN not returning HTTP challenge".
  • Fix issue with drag-drop file with UploadFile [Optional].

About

Evaluate various techniques to extract and organize information from a floor plan

https://patricknicolas.blogspot.com

License:Apache License 2.0


Languages

Language:HTML 59.4%Language:Python 28.5%Language:PowerShell 5.4%Language:JavaScript 3.2%Language:Shell 1.8%Language:CSS 1.8%Language:Procfile 0.0%