fnbm-current-sweep-chart-loss-vs-absorption-gap
收藏资源简介:
本数据集名为fnbm-current-sweep-chart-loss-vs-absorption-gap,是用于fnbm-current模型的规范比较图表,具体展示损失与吸收间隙(loss_vs_absorption_gap)的关系。数据集规模极小,仅包含1行数据,有1个列字段,列名为plotly_json,存储为字符串类型,但未提供具体内容描述。数据集由脚本sync_sweep_results.py生成,属于评估结果(eval_result)产物,采用Plotly进行可视化,无超参数或外部输入数据集。适用于模型性能分析或可视化展示任务,用户可通过HuggingFace datasets库直接加载使用。
This dataset is named fnbm-current-sweep-chart-loss-vs-absorption-gap and is used for a specification comparison chart of the fnbm-current model, specifically showing the relationship between loss and absorption gap (loss_vs_absorption_gap). The dataset is extremely small, containing only 1 row of data with 1 column field named plotly_json, stored as a string type, but no specific content description is provided. The dataset is generated by the script sync_sweep_results.py and is a product of evaluation results (eval_result), using Plotly for visualization, with no hyperparameters or external input datasets. It is suitable for model performance analysis or visualization tasks, and users can load it directly via the HuggingFace datasets library.
数据集概述:fnbm-current-sweep-chart-loss-vs-absorption-gap
该数据集包含一个用于 fnbm-current 模型的规范对比图表,主题为 损失 vs 吸收间隙 (loss_vs_absorption_gap)。
基本信息
- 许可证: MIT
- 标签:
fnbm-current,sweep-charts - 数据规模: 1 行,1 列
数据结构
- 列: 仅包含
plotly_json列,类型为字符串,用于存储 Plotly 图表数据。详细描述未提供。
生成参数
- 脚本名称:
sync_sweep_results.py - 模型:
fnbm-current - 实验名称/ID:
fnbm-current - 可视化类型: plotly
- 制品类型: eval_result(评估结果)
- 超参数: 无
- 输入数据集: 无
使用方式
可使用 datasets 库加载,示例代码:
python
from datasets import load_dataset
dataset = load_dataset("arushram/fnbm-current-sweep-chart-loss-vs-absorption-gap", split="train") print(f"Loaded {len(dataset)} rows")




