CNN architecture for Image Classification
Project description
This is a project intended to create a new CNN Network for Image Classification. Fas14MNet can also be used for Multi-Label Image Classification and in this project we try to create a CNN that will work equally well with for Traditional AI architectures as well as a Federated Architecture and Fas14MNet helps us achieve that. Fas14MNet is inspired from the basic architecture of a ResNet50 model but is much lighter while having ~15M parameters compared to 23M+ parameters in a ResNet50 model.
Steps to run:
-
from fas14mnet.net import Fas14MNet
-
model = Fas14MNet(num_classes)
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
File details
Details for the file Fas14MNet-0.1.2.linux-x86_64.tar.gz.
File metadata
- Download URL: Fas14MNet-0.1.2.linux-x86_64.tar.gz
- Upload date:
- Size: 4.3 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/3.7.1 importlib_metadata/4.6.1 pkginfo/1.8.2 requests/2.26.0 requests-toolbelt/0.9.1 tqdm/4.56.0 CPython/3.8.10
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | 6150af298f030f04182c8c0ff8935ce5277114dbda12f1a1a7ad135b61a2b6bc |
|
| MD5 | 3f45e23e98d16a748312be6b14d2a761 |
|
| BLAKE2b-256 | 0a3af8975512d403cd70a211c7e5b24871b0584b9250e24092539a7864b52fcd |
File details
Details for the file Fas14MNet-0.1.2-py3-none-any.whl.
File metadata
- Download URL: Fas14MNet-0.1.2-py3-none-any.whl
- Upload date:
- Size: 3.7 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/3.7.1 importlib_metadata/4.6.1 pkginfo/1.8.2 requests/2.26.0 requests-toolbelt/0.9.1 tqdm/4.56.0 CPython/3.8.10
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | 666f833f5b937524ade6a5447761cbb822ca330c3dabe98b21cb575021cbaae8 |
|
| MD5 | 1a12dc338e445b94f3218478729e3dd5 |
|
| BLAKE2b-256 | 0137e85cd14a80e30928b2f224078981466c343d770863d793cf2cadbfc40e9c |
