Reem Al-Halimi's repositories
Semantic-Text-Segmentation-with-Embeddings
Uses GloVe embeddings and greedy sequence segmentation to semantically segment a text document into any number of k segments.
Tokenization-and-Word-Embedding-Compatibility
The Quora Insincere Question Classification competition allows us to use the four embeddings: glove.840B.300d (GloVe), paragram_300_sl999 (paragram), wiki-news-300d-1M (wiki) and GoogleNews-vectors-negative300 (GoogleNews). In a kernel titled: "How to: Preprocessing when Using Embeddings", the author raises the issue of tokenization and its effect on how much of the training vocabulary is covered by words in an embedding. The author uses Google news embeddings to illustrate this point. In this kernel I expand on this point by exploring the effect of tokenization assumptions on the other three embeddings: GloVe, Paragram, and Wiki News.
stem_detection
Using deep learning to detect full stem instances in heavily occluded greenhouse images. The code also detects tomatoes and leaves in those images.
Browser-Based-Annotator
A browser based annotator where users can label segments with multiple labels, overlap labelled segments, label occluded objects where only disjoint parts appear, mark labelled objects as separate instances.
Quora-Insincere-Questions-Classification
Using a GRU-Attention-Capsule network to automatically detect insincere questions
Mask_RCNN_Nucli
An example using Mask_RCNN to detect nuclei from Kaggle's Data Science Bowl 2018 (see samples/nucli).
letter_images
Correspondence letters are sometimes saved as images. This repo uses standard business letter layout to extract information from the image of a letter such as: sender, recipient, date, signator, enclosures, and letter body. It also archives a given dataset of letter images for further search on the contents of those letters.
covid_data_analysis_example
A data analysis example
Python_annotation_tool
A python image annotation tool that allows multiple instance annotation of any given label
char-rnn
Multi-layer Recurrent Neural Networks (LSTM, GRU, RNN) for character-level language models in Torch
crfrnn_layer
A Lasagne layer (and supporting Theano Ops) for the CRF-as-RNN layer.
FastMaskRCNN
Mask RCNN in TensorFlow
fcn.berkeleyvision.org
Fully Convolutional Networks for Semantic Segmentation by Jonathan Long*, Evan Shelhamer*, and Trevor Darrell. CVPR 2015 and PAMI 2016.
pytorch-fpn
Feature Pyramid Networks in PyTorch
pytorch-semseg
Semantic Segmentation Architectures Implemented in PyTorch
PyTorch_examples
A set of examples around pytorch in Vision, Text, Reinforcement Learning, etc.
tensorflow-fcn-ciona17
Deep FCN Models for Semantic Segmentation of Aquatic Invasive Species in TensorFlow 1.x
Torch-vision
Datasets, Transforms and Models specific to Computer Vision