Transfer learning on VGG16 using Keras with Caltech256 and Urban Tribes dataset. Dark knowledge in transfer learning.
-
Updated
Nov 21, 2017 - Jupyter Notebook
Transfer learning on VGG16 using Keras with Caltech256 and Urban Tribes dataset. Dark knowledge in transfer learning.
Semantic Description of Images- A deep learning model for describing images using natural language.
Identify question pair with the same intent using Convolutional Neural Network
Algorithm that identifies the dog's breed given an image of a dog, or the resembling dog breed of a human face. Uses a Haar feature-based cascade classifier to detect humans, a pre-trained ResNet-50 model to detect dogs, and a fine-tuned Xception model for dog breed classification. Achieves a test accuracy of around 85% on a test set of 800 images.
This repository will include the codes and pre-trained model for ICIP 2018 paper: "Skeleton to Color Map: A Novel Representation for 3D Action Recognition with Inception Residual Networks"
udacity Behavioral-Cloning-P3
Classify the trustworthiness of yelp reviews
CNN and ANN models trained with MNIST dataset.
Deploying CNN based model to production and generate Reminder
Deep Learning paper 2017/2018 Jheronimus Academy of Data Science (JADS)
Facial Expression Classifier For CNN model.Training CNN classifier to recognize people expressions (emotions) on Fer2013 dataset.
A simple CNN-classifier designed on keras to recognize Sign Language Digits.
Implementation of CNN-based models for entity target type identification
Using Convolutional Neural Networks to build a digital recognizer.
this is a jupyter notebook that displays how a CNN would be used to classify clothes. I want to use the saved model in a RCNN to use for realtime classification of articles of clothing, so watch out for another repo soon. Also this is apart of Siraj Raval 100 Days of ML
Multi-class Image classification of people driving cars
Digit Recognition using CNN
Some ML Problems And Templates
Developed a Deep Neural Network model which classifies the traffic signs.By using Digital Image Processing techniques likes Gray Scale Conversion,Histogram Equalization,Image normalization ,we preprocessed the images.By using Convoultional Neural Network model, from keras framework developed a working model. This model gives 96% accurate results.
Add a description, image, and links to the cnn-keras topic page so that developers can more easily learn about it.
To associate your repository with the cnn-keras topic, visit your repo's landing page and select "manage topics."