• Stars
    star
    189
  • Rank 204,649 (Top 5 %)
  • Language
    Lua
  • License
    MIT License
  • Created over 9 years ago
  • Updated over 7 years ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

Deep metric learning using Triplet network

Deep Metric Learning Using Triplet Network

This code replicates the results from the paper โ€œDeep metric learning using Triplet networkโ€ (http://arxiv.org/abs/1412.6622).

It can train a TripletNet on any of the {Cifar10/100, STL10, SVHN, MNIST} datasets.

Data

You can get the needed data using the following repos:

Dependencies

Models

Available models are at the โ€œModelsโ€ directory. The basic Model.lua was used in the paper, while NiN based models achieve slightly better results.

Training

You can start training using:

th Main.lua -dataset Cifar10 -LR 0.1 -save new_exp_dir

Additional flags

Flag Default Value Description
modelsFolder ./Models/ Models Folder
network Model.lua Model file - must return valid network.
LR 0.1 learning rate
LRDecay 0 learning rate decay (in # samples
weightDecay 1e-4 L2 penalty on the weights
momentum 0.9 momentum
batchSize 128 batch size
optimization sgd optimization method
epoch -1 number of epochs to train (-1 for unbounded)
threads 8 number of threads
type cuda float or cuda
devid 1 device ID (if using CUDA)
load none load existing net weights
save time-identifier save directory
dataset Cifar10 Dataset - Cifar10, Cifar100, STL10, SVHN, MNIST
normalize 1 1 - normalize using only 1 mean and std values
whiten false whiten data
augment false Augment training data
preProcDir ./PreProcData/ Data for pre-processing (means,Pinv,P)

More Repositories

1

seq2seq.pytorch

Sequence-to-Sequence learning using PyTorch
Python
520
star
2

convNet.pytorch

ConvNet training using pytorch
Python
346
star
3

quantized.pytorch

Python
212
star
4

bigBatch

Code used to generate the results appearing in "Train longer, generalize better: closing the generalization gap in large batch training of neural networks"
Python
148
star
5

captionGen

Generate captions for an image using PyTorch
Jupyter Notebook
128
star
6

ImageNet-Training

ImageNet training using torch
Lua
102
star
7

utils.pytorch

Utilities for Pytorch
Python
90
star
8

DeepDream.torch

Torch version for https://github.com/google/deepdream
Lua
53
star
9

fix_your_classifier

Python
34
star
10

recurrent.torch

Recurrent modules for Torch
Lua
27
star
11

lmdb.torch

LMDB for Torch
Lua
26
star
12

norm_matters

Python
23
star
13

SemiSupContrast

Semi-supervised deep learning by metric embedding
Lua
19
star
14

DeepLearningCourse

Deep learning mini-course given at Technion
Jupyter Notebook
18
star
15

convNet.torch

Convolutional network training using Torch
Lua
18
star
16

captionGeneration.torch

Generate captions for an image using convolutional and recurrent networks
Jupyter Notebook
12
star
17

eladtools

Lua
11
star
18

ConvNet-torch

Training Deep Convolutional Networks on visual classification tasks
Lua
11
star
19

GoogLeNet.torch

Trained network models for Torch
9
star
20

convNet.tf

Convolutional network training using TensorFlow
Python
8
star
21

stl10.torch

STL10 Dataset on Torch
Lua
3
star
22

DataProvider.torch

Data providers for Torch
Lua
3
star
23

eladhoffer.github.io

CSS
3
star
24

colab-notebooks

Jupyter Notebook
2
star
25

seq2seq.torch

Lua
1
star
26

DescriptorLearning

Lua
1
star