Update README.md
Browse files
README.md
CHANGED
|
@@ -1,9 +1,26 @@
|
|
| 1 |
---
|
| 2 |
tags:
|
|
|
|
|
|
|
|
|
|
| 3 |
- HAR
|
|
|
|
|
|
|
|
|
|
| 4 |
---
|
| 5 |
Part of MONSTER: <https://arxiv.org/abs/2502.15122>.
|
| 6 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
***PAMAP2*** is a collection of data obtained from three Inertial Measurement Units (IMUs) placed on the wrist of the dominant arm, chest, and ankle, as well as 1 ECG heart rate [1]. The data was recorded at a frequency of 100Hz. The dataset includes annotated information about human activities performed by 9 subjects, each with their own unique physical characteristics. The majority of the subjects are male and have a dominant right hand. Notably, the dataset includes only one female subject (ID 102) and one left-handed subject (ID 108). In total, there are 18 different human activity classes represented in the dataset. To ensure an unbiased evaluation, we divide the dataset into cross-validation folds based on the subjects.
|
| 8 |
|
| 9 |
[1] Attila Reiss and Didier Stricker. (2012). Introducing a new benchmarked dataset for activity monitoring. In *16<sup>th</sup> International Symposium on Wearable Computers*, pages 108–109.
|
|
|
|
| 1 |
---
|
| 2 |
tags:
|
| 3 |
+
- time series
|
| 4 |
+
- time series classification
|
| 5 |
+
- monster
|
| 6 |
- HAR
|
| 7 |
+
pretty_name: PAMAP2
|
| 8 |
+
size_categories:
|
| 9 |
+
- 10K<n<100K
|
| 10 |
---
|
| 11 |
Part of MONSTER: <https://arxiv.org/abs/2502.15122>.
|
| 12 |
|
| 13 |
+
|PAMAP2||
|
| 14 |
+
|-|-:|
|
| 15 |
+
|Category|Audio|
|
| 16 |
+
|Num. Examples|38,856|
|
| 17 |
+
|Num. Channels|52|
|
| 18 |
+
|Length|100|
|
| 19 |
+
|Sampling Freq.|100 Hz|
|
| 20 |
+
|Num. Classes|12|
|
| 21 |
+
|License|Other|
|
| 22 |
+
|Citations|[1]|
|
| 23 |
+
|
| 24 |
***PAMAP2*** is a collection of data obtained from three Inertial Measurement Units (IMUs) placed on the wrist of the dominant arm, chest, and ankle, as well as 1 ECG heart rate [1]. The data was recorded at a frequency of 100Hz. The dataset includes annotated information about human activities performed by 9 subjects, each with their own unique physical characteristics. The majority of the subjects are male and have a dominant right hand. Notably, the dataset includes only one female subject (ID 102) and one left-handed subject (ID 108). In total, there are 18 different human activity classes represented in the dataset. To ensure an unbiased evaluation, we divide the dataset into cross-validation folds based on the subjects.
|
| 25 |
|
| 26 |
[1] Attila Reiss and Didier Stricker. (2012). Introducing a new benchmarked dataset for activity monitoring. In *16<sup>th</sup> International Symposium on Wearable Computers*, pages 108–109.
|