• Research software close
• GB close

an open-sourced highly automated machine learning Python framework for data-driven geochemistry discovery

The beta release adds in several minor changes, the most notable being the module name change.

Antibody facility

Replication package of "Mutation-based Consistency Testing for Evaluating the Code Understanding Capability of LLMs", Ziyu Li and Donghwan Shin, to appear in the Proceedings of the 3rd International Conference on AI Engineering - Software Engineering for AI (CAIN 2024). In this paper, we propose a novel method to systematically assess the code understanding performance of LLMs, particularly focusing on subtle differences between code and its descriptions, by introducing code mutations to existing code generation datasets. Code mutations are small changes that alter the semantics of the original code, creating a mismatch with the natural language description. We apply different types of code mutations, such as operator replacement and statement deletion, to generate inconsistent code-description pairs. We then use these pairs to test the ability of LLMs to correctly detect the inconsistencies. We propose a new LLM testing method, called Mutation-based Consistency Testing (MCT), and conduct a case study on the two popular LLMs, GPT-3.5 and GPT-4, using the state-of-the-art code generation benchmark, HumanEval-X, which consists of six programming languages (Python, C++, Java, Go, JavaScript, and Rust). We compare the performance of the LLMs across different types of code mutations and programming languages and analyze the results. We find that the LLMs show significant variation in their code understanding performance and that they have different strengths and weaknesses depending on the mutation type and language.

In this work, we explore the potential of multi-domain multi-branch convolutional neural networks (CNNs) for identifying comparatively rare giant radio galaxies from large volumes of survey data, such as those expected for new-generation radio telescopes like the SKA and its precursors. The approach presented here allows models to learn jointly from multiple survey inputs, in this case, NVSS and FIRST, as well as incorporating numerical redshift information. We find that the inclusion of multi-resolution survey data results in the correction of 39\% of the misclassifications seen from equivalent single domain networks for the classification problem considered in this work. We also show that the inclusion of redshift information can moderately improve the classification of giant radio galaxies. The data files we uploaded here are the exemplar model weights of the 7 architectures implemented in our work, with the model datasets coming out when Wong et al. in prep. (2022) get published.

Chromium 3' gene expression sequencing data of trypanosoma brucei brucei parasites take from murine infection. Samples were collect after 7 and 23 days infection. The cell line used contains a construct for RNAi depletion of the hyp2 gene to perturb stumpy form development via quorum sensing. Data includes raw and filtered gene transcript count matrices, as well as processed seurat objects. Code includes all code used to process, integrate and analyse data as described in publication. 

First release of the Tracks specification developed by the Cell Migration Standardisation Organisation.

individual is an R package which provides users a set of useful primitive elements for specifying individual based models (IBMs), with special attention to models for infectious disease epidemiology. Users build models by specifying variables for each characteristic describing individuals in the simulated population using data structures from the package. individual provides efficient methods for finding subsets of individuals based on these variables, or cohorts. Cohorts can then be targeted for variable updates or scheduled for events. Variable updates queued during a time step are executed at the end of a discrete time step, and the code places no restrictions on how individuals are allowed to interact. These data structures are designed to provide an intuitive way for users to turn their conceptual model of a system into executable code, which is fast and memory efficient.

The Energy Calculation Extension is a jupyter notebook that quantifies the energy expenditure for the outputs of simulation runs from the Urban Traffic Simulation (ABM) published here: Sedar Olmez, Obi Thompson Sargoni, Alison Heppenstall, Daniel Birks, Annabel Whipp, Ed Manley (2021, March 22). ���3D Urban Traffic Simulator (ABM) in Unity��� (Version 1.1.0). CoMSES Computational Model Library. Retrieved from: https://www.comses.net/codebases/32e7be8c-b05c-46b2-9b5f-73c4d273ca59/releases/1.1.0/ The notebook takes as input, outputs from the aforementioned model and produces statistical analysis and quantifies the energy used by electric motored vehicles. The complete datasets used in the journal article: An Agent-Based Simulation of Heterogeneous Driver Behaviour and its Impact on Electric Energy Consumption in Urban Space. Can be found at the following source: @article{Olmez2021, author = "Sedar Olmez and Alison Heppenstall", title = "{Drive Cycle Data from the 3D Urban Traffic Simulator (ABM) in Unity (version 1.1.0)}", year = "2021", month = "11", url = "https://figshare.com/articles/dataset/Drive_Cycle_Data_from_the_3D_Urban_Traffic_Simulator_ABM_in_Unity_version_1_1_0_/17099858", doi = "10.6084/m9.figshare.17099858.v1" } This repository contains a reproduced lite version of the code, to run all the experiments, you need to download all PHEV and ICEV csv datasets from the above link.

includes unit tests for open api spec and chart coordinates chart coordinates served from disk. Created once on deploy