blog
What have we done?
Adjusting for Bias with Procedural Data
By: Shesh Narayan Gupta, Nicholas Bear Brown 3D software are now capable of producing highly realistic images that look nearly indistinguishable from real images. This raises the question: can real datasets be enhanced with 3D-rendered data? We investigate this...
Gauguin
Gauguin is a project with five objectives: To automate EDA To create a python based AutoEDA tool similar to the many AutoML frameworks where a user uploads some data selects a target of interest and the tool creates thousands of visualizations giving insight into the...
Kindo
Kindo is a reinforcement learning high-level API enabling developers and analysts to use Stable Baselines 3 and TF-Agents algorithms. Stable Baselines 3 is powered by PyTorch. TF-Agents is powered by Tensorflow 2.X Kindo enables to train models using both Tensorflow...
Typography-MNIST (TMNIST)
This dataset is inspired by the MNIST database for handwritten digits. It consists of images representing digits from 0-9 produced using 2,990 google fonts files. The dataset consists of a single file: TMNIST_Data.csvThis file consists of 29,900 examples with labels...
PANOSE Typography
The PANOSE dataset annotates thousands of typefaces with 10 concatenated numeric values in which a typographer classifies a type by Serif Style, Weight, Proportion, Contrast, Stroke Variation, Arm Style, Letterform, Midline, and X-height.
BCH and Harvard Medical School Projects in machine learning for medical image analysis
Estimation of diffusion parameters from under-sampled measurements with deep learning Parameters inferred from the diffusion signal measured at every voxel in diffusion-weighted magnetic resonance imaging, such as mean diffusivity and fractional anisotropy, reflect...