The data was collected and represents the historical data of 6 currency pairs measured in 4-hour intervals: 4 major pairs such as GBP/USD, EUR/USD, USD/CHF, USD/JPY and 2 minor pairs such as EUR/GBP and GBP/JPY.The data set covers the period from 01-2000 to 07-2020   Three files exists: the sequence dataset, the label dataset, and the legenda.The legenda reports the follwoing information:BBSignal, BBBreakoutSignal, LabelStrategy:
BarType and HAType:


A dataset of beach volleyball serves movements annotated in 4 types (i.e. long float, long top spin, short float, short top spin) and free games. The dataset comprises the data of 10 users in 4 sessions, recorded using 4 inertial sensors per user.


The dataset comprises data from 10 users in 4 sessions of 4 players each (2 players were present in two sessions). The data were recorded using 4 BlueSense [1] inertial platforms per user. The sensors were placed on the torso, and on each segment of the dominant arm (i.e. upper arm, lower arm and hand).

The players were asked to perform 12 repetition of 4 type of serves, i.e. long float, long top spin, short float and short top spin. After the completion of the 48 serves (12 serves x 4 types), the players were asked to play freely in 2-vs-2 games, following the standard beach volleyball rules.

The dataset include one file per user per session. In the file, the data are formatted in single .txt files for every user. - Each file contains 54 columns:

  • 1 timestamp in milliseconds
  • 13 x 4 body position (torso, upper_arm, lower_arm, hand):
    • 9 for sensor data
      • 3 x accelerometer
      • 3 x gyroscope
      • 3 x magnetometer
    • 4 for quaternion format of orientation data (w, x, y, z)
  • 1 for the activity label ID:
    • 1001 Long float serve
    • 1002 Long top spin serve
    • 1003 Short float serve
    • 1004 Short top spin serve

[1] Daniel Roggen, Arash Pouryazdan, and Mathias Ciliberto. 2017. BlueSense: designing an extensible platform for wearable motion sensing, sensor research and IoT applications. In Proceedings of the 2018 International Conference on Embedded Wireless Systems and Networks. ACM.

Citation Request:

Use of this dataset in publications must be acknowledged by referencing the following publication:

Mathias Ciliberto, Luis Ponce Cuspinera, and Daniel Roggen. 2021. Collecting a dataset of gestures for skill assessment in the field: a beach volleyball serves case study. In Proceedings of the ACM International Joint Conference and International Symposium on Pervasive and Ubiquitous Computing and Wearable Computers, 2021.


Passwords that were leaked or stolen from sites. The Rockyou Dataset is about 14 million passwords.


Note that, you must register on, and join our team "new_media", finally download the dataset. Pretecting the user pravicy.  


We provide a public available database for arcing event detection. We design a platform for arcing fault simulation. The arc simulation is carried out in our local lab under room temperature. A general procedure to collect the arcing and normal current and voltage wave, is designed, which consists of turning on the load, generating arc, stoping arc, turning off the load. The data is collected by a 16bit, 10KHz high resolution recorder and a 12bit, 64000Hz low resolution sensor.