Jiatao Gu's Projects
Official implementation of AnimateDiff.
[WIP] Attention Is All You Need (Vaswani et al. 2017) by Chainer.
RL Experiments from our paper "Backpropagation Through the Void": https://arxiv.org/abs/1711.00123. Lovingly forked from OpenAI's RL Baseline repo.
Bidirectional Attention Flow
Official PyTorch implementation of the paper Progressive Deblurring of Diffusion Models for Coarse-to-Fine Image Synthesis.
Transform an image into a pencil-sketch
Computation Graph Toolkit
Variational AutoEncoder based on http://nbviewer.ipython.org/gist/duschendestroyer/a41fcab5f7f9ffa45387
Code to train and use models from "Charagram: Embedding Words and Sentences via Character n-grams".
A simple memory manager for CUDA designed to help Deep Learning frameworks manage memory
Microsoft Cognitive Toolkit (CNTK), an open source deep-learning toolkit
incorporating copying mechanism in sequence-to-sequence learning
A language-agnostic interface to TensorBoard
PyTorch CTC Decoder bindings
Current CV of Jiatao Gu
Deep Convolutional Generative Adversarial Networks
Jiatao Gu's homepage.
A dependency based RNN composition methods for phrases and sentences.
Efficient implementation of Depthwise Conv2d. The main code is from https://github.com/MegEngine
[NeurIPS'19] Deep Equilibrium Models
dgk_lost_conv 中文对白语料 chinese conversation corpus
This repository contains the code for the CVPR 2020 paper "Differentiable Volumetric Rendering: Learning Implicit 3D Representations without 3D Supervision"
Official PyTorch Implementation of "Scalable Diffusion Models with Transformers"
Implement Decoupled Neural Interfaces using Synthetic Gradients in Pytorch