Tags: lstm, pytorch | Score: 2 | Answered: No

Show, Attend, and Tell | a PyTorch Tutorial to Image Captioning (⭐ 2889 | Python)

Wanted to share a PyTorch implementation of an alternative attention mechanism based on wave physics. How it works: 1. QKV projection (standard) 2. Bilinear scatter — deposit values onto a co...

We implemented a Soft Decision Tree (SDT) and a Short-term Memory Soft Decision Tree (SM-SDT) using PyTorch. The methods were extensively tested on simulated and clinical datasets. The SDT was visuali...

A PyTorch Implementation of Single Shot MultiBox Detector (⭐ 5229)

A faster pytorch implementation of faster r-cnn (⭐ 7866)

My attempt to implement "World Models" by David Ha and Jurgen Schmidhuber (⭐ 3 | Python)

18 torch.Tensor.view() Simply put, torch.Tensor.view() which is inspired by numpy.ndarray.reshape() or numpy.reshape(), creates a new view of the tensor, as long as …

Jun 11, 2018 · In this view x.view(-1) is a weird flatten layer but missing the squeeze (i.e. adding a dimension of 1). Adding this squeeze or removing it is usually important for the code to …

18 torch.Tensor.view() Simply put, torch.Tensor.view() which is inspired by numpy.ndarray.reshape() or numpy.reshape(), creates a new view of the tensor, as long as the new shape is compatible with th...

Jun 11, 2018 · In this view x.view(-1) is a weird flatten layer but missing the squeeze (i.e. adding a dimension of 1). Adding this squeeze or removing it is usually important for the code to actual...

18 torch.Tensor.view() Simply put, torch.Tensor.view() which is inspired by numpy.ndarray.reshape() or numpy.reshape(), creates a new view of the tensor, as long as …

Jun 11, 2018 · In this view x.view(-1) is a weird flatten layer but missing the squeeze (i.e. adding a dimension of 1). Adding this squeeze or removing it is usually important for the code to …

Burgess et al. "MONet: Unsupervised Scene Decomposition and Representation" (⭐ 89)

18 torch.Tensor.view() Simply put, torch.Tensor.view() which is inspired by numpy.ndarray.reshape() or numpy.reshape(), creates a new view of the tensor, as long as the new shape is compatible with th...

Jun 11, 2018 · In this view x.view(-1) is a weird flatten layer but missing the squeeze (i.e. adding a dimension of 1). Adding this squeeze or removing it is usually important for the code to actual...

18 torch.Tensor.view() Simply put, torch.Tensor.view() which is inspired by numpy.ndarray.reshape() or numpy.reshape(), creates a new view of the tensor, as long as the new shape is compatible with th...

Jun 11, 2018 · In this view x.view(-1) is a weird flatten layer but missing the squeeze (i.e. adding a dimension of 1). Adding this squeeze or removing it is usually important for the code to actual...

Custom Implementation of the famous Transformer Architecture from scratch based on the Seminal Paper Attention is All You Need by Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jone...

I built torch-continuum, a library that auto-detects your GPU and applies the right hardware-specific optimizations for you. One line before your training loop: `import torch_continuum` `torch_...