Suraj Deshmukh's repositories
awesome-deep-vision
A curated list of deep learning resources for computer vision
categorical-encoding
A library of sklearn compatible categorical variable encoders
connexion
Swagger/OpenAPI First framework for Python on top of Flask with automatic endpoint validation and OAuth2 support
deep-learning-keras-euroscipy2016
# Deep Learning with Keras
deep-learning-models
Keras code and weights files for popular deep learning models.
dist-keras
Distributed deep learning with Keras and Apache Spark.
dl-models-for-qa
Keras DL models to answer 8th grade science multiple choice questions (Kaggle AllenAI competition).
doc-classify
Document classification.
FaceRecog-Keras
Face Recognition using Neural Networks implemented using keras library
headlines
Automatically generate headlines to short articles
imbalanced-learn
Python module to perform under sampling and over sampling with various techniques.
jobdescription2jobtitle
classify a job description (or noisy job title) into a ONET job title
keras-cam
Keras implementation of class activation mapping
keras-js
Run trained Keras models in the browser, with GPU support
kerasR
R interface to the keras library
lectures
Oxford Deep NLP 2017 course
machine-learning
Web-interface, and programmatic API for classification and regression predictions
music-auto_tagging-keras
Music auto-tagging models and trained weights in keras/theano
Neural-Style-Transfer
Implementation of Neural Style Transfer from the paper [A Neural Algorithm of Artistic Style](http://arxiv.org/abs/1508.06576) in Keras 1.0.6
Object-detection-with-deep-learning-and-sliding-window
Introduces an approach for object detection in an image with sliding window. The repository contains three files, make_data.py reads the image in gray scale and converts the image into a numpy array. The labels are also appended based on the file name. In this case, if the file name starts with "trn", then 1 is appended else 0. Finally, all the images and labels are saved into .npy file. The test-model-1.py file loads the images and converts the labels into two categories as we are doing binary classification of images. The model is built using keras with theano as backend. In this case, the best training accuracy was 80% since the data was just 500 images and the testing accuracy was 67%
question-answering
Question Answering System.
quiver
Interactive convnet features visualization for Keras
scikit-multilearn
A scikit-learn based module for multi-label et. al. classification
sklearn2pmml
Python library for converting Scikit-Learn pipelines to PMML
spark-sklearn
Scikit-learn integration package for Spark
TensorFlow-Tutorials
Simple tutorials using Google's TensorFlow Framework
VQA_LSTM_CNN
Train a deeper LSTM and normalized CNN Visual Question Answering model. This current code can get 58.16 on OpenEnded and 63.09 on Multiple-Choice on test-standard.