首页/数据集/Prompts used for measuring performance of GPT in diagnosis

Prompts used for measuring performance of GPT in diagnosis

Name: Prompts used for measuring performance of GPT in diagnosis
Creator: Lawrence Berkeley National Laboratory
Published: 2024-08-15T00:00:00+08:00

Zenodo2024-08-15 更新2026-04-07 收录

AI诊断

性能评估

数据链接：

https://zenodo.org/doi/10.5281/zenodo.13327451 数据链接链接失效反馈

官方服务：

资源简介：

Prompts used for measuring performance of GPT in diagnosis

应用场景：

提供机构：

Lawrence Berkeley National Laboratory

创建时间：

2024-08-15

传染病检测

性能评估

qRT-PCR, antigen ELISA, and serology results, Lagos, Nigeria.

NIAID Data Ecosystem50

Performance metrics with a background class weight of 0.2, 0.5, 1.0, and 1.5 for loss.

类别权重调优

性能评估

With the background weight of 1.5, all the classes have an equal weight. The Std Dev Diff represents the standard deviation over validation sections of the differences of the metrics to the default ba

Figshare2024-03-29 更新10

Diagnostic accuracy summary table.

性能评估

Diagnostic accuracy summary table.

NIAID Data Ecosystem10

AICs and MSEs based on Number of Infected Cases in Mauritius, from 12th April 2020 to 23rd April 2020.

传染病统计建模

性能评估

AICs and MSEs based on Number of Infected Cases in Mauritius, from 12th April 2020 to 23rd April 2020.

Figshare2020-07-10 更新10

rebase_vgs_gpt-oss-20b_rg_cognition_ns32_md4_bt0_1_seed42_rg_cognition_vgs_2gpu

智能问答

性能评估

该数据集包含多个字段，主要用于问题生成和回答任务。数据集的主要特征包括问题文本（question）、生成ID（generation_id）、生成的回答（generation）、标记数量（num_tokens）、奖励分数（reward）、问题索引（question_index）、目标文本（target）、任务类型（task）、预测值（vf_prediction）和级别（level）。数据集仅包含测试

Hugging Face2026-05-07 更新10

Prompts used for measuring performance of GPT in diagnosis

资源简介：

相关数据集