GitPlanet

Cheap and reliable Node.js hosting starts at $3/month, and $1/month static HTML hosting

Created with love in Canada, visit hostnodejs.com today

Feel like to post an Ad? Learn Details
All Projects → betterenvi → Qa Rankit

betterenvi / Qa Rankit

Licence: mit
QA - Answer Selection (Rank candidate answers for a given question)

Labels

jupyter-notebookcnnlstm

Projects that are alternatives of or similar to Qa Rankit

Video Classification
Tutorial for video classification/ action recognition using 3D CNN/ CNN+RNN on UCF101
Stars: ✭ 543 (+1710%)
Mutual labels:  jupyter-notebook, cnn, lstm
Tensorflow Tutorial
Some interesting TensorFlow tutorials for beginners.
Stars: ✭ 893 (+2876.67%)
Mutual labels:  jupyter-notebook, cnn, lstm
Pytorch Pos Tagging
A tutorial on how to implement models for part-of-speech tagging using PyTorch and TorchText.
Stars: ✭ 96 (+220%)
Mutual labels:  jupyter-notebook, cnn, lstm
Neural Networks
All about Neural Networks!
Stars: ✭ 34 (+13.33%)
Mutual labels:  jupyter-notebook, cnn, lstm
Neural Image Captioning
Implementation of Neural Image Captioning model using Keras with Theano backend
Stars: ✭ 12 (-60%)
Mutual labels:  jupyter-notebook, cnn, lstm
End To End Sequence Labeling Via Bi Directional Lstm Cnns Crf Tutorial
Tutorial for End-to-end Sequence Labeling via Bi-directional LSTM-CNNs-CRF
Stars: ✭ 87 (+190%)
Mutual labels:  jupyter-notebook, cnn, lstm
Deeplearning tutorials
The deeplearning algorithms implemented by tensorflow
Stars: ✭ 1,580 (+5166.67%)
Mutual labels:  jupyter-notebook, cnn, lstm
Keras basic
keras를 이용한 딥러닝 기초 학습
Stars: ✭ 39 (+30%)
Mutual labels:  jupyter-notebook, cnn, lstm
Natural Language Processing With Tensorflow
Natural Language Processing with TensorFlow, published by Packt
Stars: ✭ 222 (+640%)
Mutual labels:  jupyter-notebook, cnn, lstm
Screenshot To Code
A neural network that transforms a design mock-up into a static website.
Stars: ✭ 13,561 (+45103.33%)
Mutual labels:  jupyter-notebook, cnn, lstm
Pytorch Learners Tutorial
PyTorch tutorial for learners
Stars: ✭ 97 (+223.33%)
Mutual labels:  jupyter-notebook, cnn, lstm
Image Captioning
Image Captioning using InceptionV3 and beam search
Stars: ✭ 290 (+866.67%)
Mutual labels:  jupyter-notebook, cnn, lstm
Lidc nodule detection
lidc nodule detection with CNN and LSTM network
Stars: ✭ 187 (+523.33%)
Mutual labels:  jupyter-notebook, cnn, lstm
Pytorch Sentiment Analysis
Tutorials on getting started with PyTorch and TorchText for sentiment analysis.
Stars: ✭ 3,209 (+10596.67%)
Mutual labels:  jupyter-notebook, cnn, lstm
Cryptocurrencyprediction
Predict Cryptocurrency Price with Deep Learning
Stars: ✭ 453 (+1410%)
Mutual labels:  jupyter-notebook, cnn, lstm
Telemanom
A framework for using LSTMs to detect anomalies in multivariate time series data. Includes spacecraft anomaly data and experiments from the Mars Science Laboratory and SMAP missions.
Stars: ✭ 589 (+1863.33%)
Mutual labels:  jupyter-notebook, lstm
Multi Class Text Classification Cnn Rnn
Classify Kaggle San Francisco Crime Description into 39 classes. Build the model with CNN, RNN (GRU and LSTM) and Word Embeddings on Tensorflow.
Stars: ✭ 570 (+1800%)
Mutual labels:  cnn, lstm
Cvnd exercises
Exercise notebooks for CVND.
Stars: ✭ 622 (+1973.33%)
Mutual labels:  jupyter-notebook, cnn
Stockpriceprediction
Stock Price Prediction using Machine Learning Techniques
Stars: ✭ 700 (+2233.33%)
Mutual labels:  jupyter-notebook, lstm
Text Classification
Implementation of papers for text classification task on DBpedia
Stars: ✭ 682 (+2173.33%)
Mutual labels:  cnn, lstm
View All Similar Projects ➔

QA-rankit

Rank candidate answers for a given question.

Models

Unsupervised

We first tried some unsupervised models. Although these models are straightforward and simple, they work effectively!

  • Word Overlap Count
  • IDF weighted
  • Q-A distance
  • ...
Supervised

We can use those metrics calculated in unsurpervised models as features of surpervised models. Besides, we can employ other models like CNN and LSTM to extract more features.

In this program, we tried following models:

  • Random Forest
  • Logistic Regression
  • Mixed CNN
  • Mixed LSTM

Among these models, the mixed LSTM achieved best performance. LSTM

Source code

main.ipynb

Main code, edited using Jupyter Notebook.

You'd better open this file using Jupyter Notebook.

If you dont's have Jupyter Notebook installed on your computer, please try main.py.

What main.ipynb does:

  • read data
  • preprocess data
  • extract features
  • fit models (models are implements in other source files)
  • evaluate models
  • predict on test dataset
main.py

.py version of main.ipynb.

MyLSTM.py

Implementation of an adapted LSTM model using Keras.

MyCNN.py

Implementation of an adapted CNN model using Keras.

MyGA.py, ParamGA.py

Implementaion of an adapted Genetic Algorithm using DEAP.

This can be used to find respectable parameters for sklearn models, like RandomForestClassifier.

PairWiseRanker.py

Implementation of pairwise ranking algorithm.

util.py, word2vec_util.py

Implementation of some tools.

Running environment settings

This program is developed under Python 2.7.

Packages that this program uses include:

  • Pandas
  • Numpy
  • DEAP
  • NLTK
  • Keras

This program also uses learning2rank. This original repository of learning2rank is https://github.com/shiba24/learning2rank. I forked it and made some modifications. The repository is https://github.com/betterenvi/learning2rank. Therefore the modified version will be better if you want to use learning2rank.

learning2rank also uses some packages, please install them if you want to use learning2rank.

It's possible that I miss some packages that this program actually uses. Therefore, I used the following command to generate requirements.txt file:

$ pip freeze > requirements.txt

then you can run the following command:

$ pip install -r requirements.txt

Actually, many packages listed in requirements.txt have been included in Anaconda.

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].