shoaibmerajsami Goto Github PK

attention-module

Official PyTorch code for "BAM: Bottleneck Attention Module (BMVC2018)" and "CBAM: Convolutional Block Attention Module (ECCV2018)"

classification-of-digits-mnist-

contrastive-unpaired-translation

Contrastive unpaired image-to-image translation, faster and lighter training than cyclegan (ECCV 2020, in PyTorch)

coupledverifier

fingerprint-antispoofing-domain-adaptation-and-adversarial-learning

Fingerprint Antispoofing/PAD (forgery detection) with deep learning.

kagglecompetitions

source code of my different kaggle competitions

lenet-5

PyTorch implementation of LeNet-5 with live visualization

mtcnn

pytorch-radial-basis-function-layer

An implementation of an RBF layer/module using PyTorch.

shoaibmerajsami

Config files for my GitHub profile.

siamese-triplet

Siamese and triplet networks with online pair/triplet mining in PyTorch

som

In this project I write SOM(Self-Organazing-Map) Algorithm for Character recognition(There was 7 character and for each one 3 different example in train data). this Algorithm is competitive Algorithm, for example in this problem we define 7 random point (same dimension with characters) and in each iteration we find the nearest character to this random point and update the value of this random point with this approach: it was added by the nearest neuron (which is multiplied with Learning Rate). Diamond and linear Model difference is in the way of updating in this two model we update the nearest random point too (nearest in linear shape or Diamond shape)