All Projects → mmehdig → Seq2seq Attention Model

mmehdig / Seq2seq Attention Model

Licence: mit
An implementation for attention model in Keras for sequence to sequence model.

Projects that are alternatives of or similar to Seq2seq Attention Model

Nearby exoplanet map
Creating a stylised pseudo-3D exoplanet map using matplotlib
Stars: ✭ 15 (-25%)
Mutual labels:  jupyter-notebook
Core Stories
All the notebooks for the analysis of Emotional Arcs within the Project Gutenberg corpus, see "The emotional arcs of stories are dominated by six basic shapes"
Stars: ✭ 15 (-25%)
Mutual labels:  jupyter-notebook
Grab Aiforsea
Entry for Grab's AI for S.E.A. challenge
Stars: ✭ 20 (+0%)
Mutual labels:  jupyter-notebook
Udacity Deep Learning Nanodegree
This is just a collection of projects that made during my DEEPLEARNING NANODEGREE by UDACITY
Stars: ✭ 15 (-25%)
Mutual labels:  jupyter-notebook
Azure Webapp W Cntk
Deployment template for Azure WebApp, CNTK, Python 3 (x64) and sample model
Stars: ✭ 15 (-25%)
Mutual labels:  jupyter-notebook
Sage
Home of the Semi-Analytic Galaxy Evolution (SAGE) galaxy formation model
Stars: ✭ 15 (-25%)
Mutual labels:  jupyter-notebook
Seq 2 Seq Ocr
Handwritten text recognition with Keras
Stars: ✭ 15 (-25%)
Mutual labels:  jupyter-notebook
Brain Tumor Segmentation Keras
Keras implementation of the multi-channel cascaded architecture introduced in the paper "Brain Tumor Segmentation with Deep Neural Networks"
Stars: ✭ 20 (+0%)
Mutual labels:  jupyter-notebook
Syde 522
Stars: ✭ 15 (-25%)
Mutual labels:  jupyter-notebook
Anda
Code for our ICAR 2019 paper "ANDA: A Novel Data Augmentation Technique Applied to Salient Object Detection"
Stars: ✭ 20 (+0%)
Mutual labels:  jupyter-notebook
Mj583
J583 Advanced Interactive Media
Stars: ✭ 15 (-25%)
Mutual labels:  jupyter-notebook
Facealignmentcompare
Empirical Study of Recent Face Alignment Methods
Stars: ✭ 15 (-25%)
Mutual labels:  jupyter-notebook
Deep Learning Experiments
Notes and experiments to understand deep learning concepts
Stars: ✭ 883 (+4315%)
Mutual labels:  jupyter-notebook
Attend Infer Repeat Pytorch
S.M.Ali Eslam et.al. Attend, Infer, Repeat: Fast Scene Understanding with Generative Models ICML16
Stars: ✭ 15 (-25%)
Mutual labels:  jupyter-notebook
Intrusion Detection System
I have tried some of the machine learning and deep learning algorithm for IDS 2017 dataset. The link for the dataset is here: http://www.unb.ca/cic/datasets/ids-2017.html. By keeping Monday as the training set and rest of the csv files as testing set, I tried one class SVM and deep CNN model to check how it works. Here the Monday dataset contains only normal data and rest of the days contains both normal and attacked data. Also, from the same university (UNB) for the Tor and Non Tor dataset, I tried K-means clustering and Stacked LSTM models in order to check the classification of multiple labels.
Stars: ✭ 20 (+0%)
Mutual labels:  jupyter-notebook
Pytorch Bicubic Interpolation
Bicubic interpolation for PyTorch
Stars: ✭ 15 (-25%)
Mutual labels:  jupyter-notebook
Ud810 Intro Computer Vision
My solutions for Udacity's "Introduction to Computer Vision" MOOC
Stars: ✭ 15 (-25%)
Mutual labels:  jupyter-notebook
P5deeplearn
deeplearn.js meets p5
Stars: ✭ 20 (+0%)
Mutual labels:  jupyter-notebook
Altair Catplot
Utility to generate plots with categorical variables using Altair.
Stars: ✭ 20 (+0%)
Mutual labels:  jupyter-notebook
Lstm Sentiment Analysis
Sentiment Analysis with LSTMs in Tensorflow
Stars: ✭ 886 (+4330%)
Mutual labels:  jupyter-notebook

seq2seq-attention-model

An implementation for attention model in Keras for sequence to sequence model.

Known bugs

  • This implementation is wrong! The attention model is using the future information in order to predict the next word but since we only tested it on known answers to see how the attention can be used as implicit alignment, the result truns to be fine. The bug is fixable. I have other versions of this code but I have no plan in near future to update this repository.

  • The dataset I used is not available now because it wasn't my contribution to this experiment. Maybe in future I can add it here. But first I have to ask if I am alowed. But the idea was to have a synthetic data which two languages represent two different syntax for sentence generation. You can generate such dataset on your own and if you want to test on natural language you need a parallel corpora but of course the vocabulary size might be an issue.

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].