CV--Swin TransFromer, you should change the Backbone again

previously on Hello everyone, VIsion Transformer has reached its sixth day. With the passage of time, the difficulty is getting greater and greater. I believe you have learned from ResNet, ViT Transformer and Diet. Today, you have learned the best ICCV paper Swin Transformer in 2021. Swin Transformer directly killed the list on major pound li ...

Posted by mlavwilson on Fri, 03 Dec 2021 18:22:11 -0800

[MAE] implementation of Masked Autoencoders and visualization of pre training

Masked Autoencoders Are Scalable Vision Learners MAE proposes a self supervised training method, which can effectively and train the model and improve the performance of the model. This project realizes the self-monitoring training part and visualizes the training process. network structure The structure of MAE is relatively simple. It is c ...

Posted by derzok on Fri, 03 Dec 2021 17:10:04 -0800

Parakeet: teach you how to build a Chinese speech synthesis model (training and prediction)

Chinese speech synthesis scheme based on PaddleSpeech Chinese speech synthesis scheme: https://aistudio.baidu.com/aistudio/projectdetail/2791125?contributionType=1Paddlespeech source code: https://github.com/PaddlePaddle/PaddleSpeech Chinese standard female voice library (10000 sentences) brief introduction Speech synthesis is a technology ...

Posted by nblackwood on Fri, 03 Dec 2021 14:10:57 -0800

AdvSemiSeg realizes semi supervised semantic segmentation by generating confrontation

Project description Reproduction of Paper "Adversarial Learning for Semi-Supervised Semantic Segmentation" with PaddlePaddle. This project reproduces the classic paper "Adversarial Learning for Semi-Supervised Semantic Segmentation" in semi supervised semantic segmentation field based on PaddlePaddle, and achieves the index of thesi ...

Posted by atomm on Fri, 03 Dec 2021 14:00:05 -0800

PPASR streaming and non streaming speech recognition

PPASR streaming and non streaming speech recognition The project will be divided into three stages: beginner ,Progressive class and Final stage Branch, which is currently the final level and continuously maintains the version. PPASR Chinese name PaddlePaddle Chinese speech recognition (PaddlePaddle Automatic Speech Recognition), is a PaddlePa ...

Posted by vitch2002 on Tue, 30 Nov 2021 08:54:09 -0800

Meter reading based on PaddleX

Industrial pointer meter reading 1 project description In this project, we mainly introduce how to use target detection and semantic segmentation to realize the reading of pointer meter. In the power and energy plant, the meter readings need to be monitored regularly to ensure the normal operation of equipment and the safety of the plant. Ho ...

Posted by iceman2g on Mon, 29 Nov 2021 16:46:31 -0800

It's possible to create endless images with words. Dale's pre training model starts

It is possible to create endless images with words. The Russian version of Dall-E pre training model starts (including automatic translation, Chinese can be entered) Look, look, this dalle is really wonderful. Write what you read and draw what you think. Don't miss your brain hole. "Avocado chair" "Beau ...

Posted by thatsme on Sat, 27 Nov 2021 19:20:51 -0800

Combined with PaddleSeg and PyQt, a new crown lung CT ground glass lesion detection gadget was made

PaddleSeg PaddleSeg is an end to end image segmentation development kit developed by PaddlePaddle, covering a large number of high quality segmentation models in different directions, such as high precision and lightweight. Through modular design, two application modes of configuration driver and API call are provided to help developers more c ...

Posted by txhoyt on Sat, 20 Nov 2021 01:32:55 -0800

Let AI do homework: typo single test based on paddelnlp taskflow

introduction In previous project ERNIE for CSC: can't you tell the difference? Here comes the Savior! In, we introduced how to train a text error correction model from 0. At the end of the article, the Taskflow of PaddleNLP is mentioned. This paper launches an interesting project on Taskflow based on text error correction: how to make AI do t ...

Posted by izy on Thu, 18 Nov 2021 20:30:16 -0800

[PaddleNLP] thousand words data set: emotion analysis - SKEP

[PaddleNLP] thousand words data set: emotion analysis - SKEP This project uses the pre training model SKEP complete Thousand words data set: emotion analysis match It contains three sub tasks: sentence level emotion classification, evaluation object level emotion classification and viewpoint extraction Welcome to praise, fork and pay attention ...

Posted by davidklonski on Sat, 06 Nov 2021 03:35:18 -0700