[BMVC 2022] Official repository for "How to Train Vision Transformer on Small-scale Datasets?"

This study presents the development and validation of AI models for both nodule detection and cancer classification tasks. This benchmarking across multiple datasets establishes the DLCSD as a reliable resource for lung cancer AI research.

Some large-scale MVC benchmark datases

Millions-Level Face/Human-Scene Image-Text Datasets

Collection of metadata of different Side Channel Attack datasets

Distributed computing workflow for generation and analysis of large scale molecular datasets obtained running multi-source multi-fidelity quantum chemical calculations of ground-state and excited-state properties

scava-datasets

The four social datasets

Unsupervised domain adaptation is critical in various computer vision tasks, such as object detection, instance segmentation, which aims to alleviate performance degradation caused by domain-shift and promotes the rapid application of models such as detection model in practical tasks. Previous works use adversarial training to align global features and local features to eventually minimize the domain discrepancy. However, there is no semantic information for each pix-feature in the local features, which causes the different categories of pixel-features to align and increase the difficulty of alignment. So a Semantic Consistency Alignment(SCA) model is proposed to align the pixel-feature with the same categories and mitigate negative transfer in this paper. Firstly, a Semantic Prediction Model(SPM) is designed in SCA to predict the categories of each pixel-feature, which is trained by the source domain according to the receptive field. Then the penultimate layer of SPM as a Semantic Vector Map(SVM) is jointed with the local feature to send to the local domain discriminator. Meanwhile, the output of SPM as an attention map is used to adjust the weight of the local domain discriminator. Extensive experiments for unsupervised domain adaptation on commonly used datasets demonstrate the effectiveness of our proposed approach for robust object detection in various domain shift scenarios

No description available

No description available

The four small social datasets and the IS algorithms code

This are Datasets is for cryptocurrency scam analysis and moedl development

Full-stack AI chatbot enabling natural language queries on 20M+ records (Postgres, MongoDB, LangChain, GPT-3.5)

No description available

This work was aimed at finding methods to identify the most distant proteins and most diverse subsets of proteins from large protein databases in a scalable and efficient way using a dataset of protein embeddings from SwissProt, data mining techniques and metaheuristics.

This includes the baselines, video datasets and the core code of our system

MSBA 6411 Project 1: This SCA repository consists of the analyses carried out on the official customer transactions dataset provided by the Sun Country Airlines.

No description available

SCA using neural networks and ASCAD dataset

repository with data

No description available

datasets_scatterplot

datasets for notiscam (linghacks)

No description available

This dataset presents the open datasets for scan and bim models available on the internet

datasets of raw text from reddit

Medical scans data sets for training image classifiers

The code behind the datasets for the USS data scavenger hunt.

No description available