公开数据集
数据结构 ? 2.46M
Data Structure ?
* 以上分析是由系统提取分析形成的结果,具体实际数据为准。
README.md
Context
In machine learning there is a long path from understanding to intuition. I have created many data files of traditional electronics test pattern to see the response of different activation, loss, optimizers, and metrics in Keras. These files should give some ability to test drive your chosen type of machine learning with a very deliberate data set.
I wanted something that was infinitely predicable to see how all the different settings effected the algorithms to set a base line for me to gain intuition as to how they should behave once I make more complex models.
Content
These files most contain single line 10,000 example patterns in sine, cosine, triangle, and others. Frequency and amplitude in some change through out the set. One has 2,500 example with 4 features of a sine wave 90 degrees out of phase from each other. The values are all between zero and one so no scaling should be necessary.
CosineDecAmpFreqInc, CosineDecreasingAmp, CosineIncAmpFreqInc, CosineIncAmpFreqSlowing, ExponentialDecayTenWaves, ExponentialRiseTenWaves, FourSineWaves, LinearFall, LinearRise, Lorentz, Multitone, Pulse10Waves, Pulse10WavesInverted, RandomSamples, SinFiveWaves, SinFourtyWaves, SinTenWaves, SinTwentyWaves, 30,000 SquareFiveWaves, SquareTenWaves, SweepOneToFive, SweepOneToTwo, SweepOneToTwoPointFive, SyncPattern, TriangleFiveWaves, TriangleTenWaves.csv
Acknowledgements
Some were generated using Tektronics ArbExpress and modified in Excel for scale. Some I generated in c#.
Inspiration
How about a good Toy example of a LSTM in Keras with multivariate data and a single prediction of one of the columns. I did the 4 sine wave .csv to try this. So far the examples I have found just average all of them.
×
帕依提提提温馨提示
该数据集正在整理中,为您准备了其他渠道,请您使用
注:部分数据正在处理中,未能直接提供下载,还请大家理解和支持。
暂无相关内容。
暂无相关内容。
- 分享你的想法
去分享你的想法~~
全部内容
欢迎交流分享
开始分享您的观点和意见,和大家一起交流分享.
数据使用声明:
- 1、该数据来自于互联网数据采集或服务商的提供,本平台为用户提供数据集的展示与浏览。
- 2、本平台仅作为数据集的基本信息展示、包括但不限于图像、文本、视频、音频等文件类型。
- 3、数据集基本信息来自数据原地址或数据提供方提供的信息,如数据集描述中有描述差异,请以数据原地址或服务商原地址为准。
- 1、本站中的所有数据集的版权都归属于原数据发布者或数据提供方所有。
- 1、如您需要转载本站数据,请保留原数据地址及相关版权声明。
- 1、如本站中的部分数据涉及侵权展示,请及时联系本站,我们会安排进行数据下线。