Skip to content

Commit

Permalink
Merge pull request #24 from eXascaleInfolab/naterq_taxidermy
Browse files Browse the repository at this point in the history 
1. documentation readme dataset and taxidermy of the families
  • Loading branch information
@qnater
qnater authored Oct 23, 2024
2 parents e928dc7 + 94c2619 commit 2192dfe
Show file tree
Hide file tree
Showing 2 changed files with 59 additions and 28 deletions.
6 changes: 3 additions & 3 deletions .idea/workspace.xml
View file

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

81 changes: 56 additions & 25 deletions imputegap/dataset/README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -17,11 +17,17 @@ The dataset's dimensions are 64 series, each containing 256 values.
This dataset is primarily used for the analysis of medical and brain-related data, with a focus on detecting predictable patterns in brain wave activity.

### Plots
The plots present a series of plots derived from the EEG-TEST dataset, illustrating various aspects of the data and preprocessing steps.
EEG-Test dataset - raw data 64x256 shows the full raw dataset, consisting of NxM time series.
EEG-Test dataset - raw data 20x400 provides a subset of the data, limited to 20 time series over 400 time steps, while EEG-Test dataset - raw data 01x400 focuses on a single time series extracted from the dataset.
Finally, EEG-Test - normalized 20x400 demonstrates the impact of "MIN-MAX" normalization on the raw data, applied to the same 20x400 subset.

![EEG-Test dataset - raw data 64x256](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/eeg-test/01_eeg-test-rawdata-NxM_graph.jpg)
![EEG-Test dataset - raw data 20x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/eeg-test/02_eeg-test-rawdata20x400_graph.jpg)
![EEG-Test dataset - raw data 01x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/eeg-test/03_eeg-test-rawdata01x400_graph.jpg)
![EEG-Test - normalized 20x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/eeg-test/04_eeg-test-normmin_maxdata01x400_graph.jpg)


### Features

| Category | Feature | Value |
Expand Down Expand Up @@ -84,11 +90,16 @@ The extracted dataset contains 1201 values across 33 series.
The goal of this dataset is to provide an alternative perspective on neuroscience-related datasets and to enable comparisons of results from different research studies utilizing the same technology.

### Plots
The plots present a series of plots derived from the EEG dataset, illustrating various aspects of the data and preprocessing steps.
EEG dataset - raw data 64x256 shows the full raw dataset, consisting of NxM time series.
EEG dataset - raw data 20x400 provides a subset of the data, limited to 20 time series over 400 time steps, while EEG dataset - raw data 01x400 focuses on a single time series extracted from the dataset.
Finally, EEG - normalized 20x400 demonstrates the impact of "MIN-MAX" normalization on the raw data, applied to the same 20x400 subset.

![EEG dataset - raw data 64x256](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/eeg/01_eeg-rawdata-NxM_graph.jpg)
![EEG dataset - raw data 20x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/eeg/02_eeg-rawdata20x400_graph.jpg)
![EEG dataset - raw data 01x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/eeg/03_eeg-rawdata01x400_graph.jpg)
![EEG - normalized 20x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/eeg/04_eeg-normmin_maxdata01x400_graph.jpg)

![EEG dataset - raw data 33x1201](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/eeg/01_eeg-rawdata-NxM_graph.jpg)
![EEG dataset - raw data 20x400](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/eeg/02_eeg-rawdata20x400_graph.jpg)
![EEG dataset - raw data 01x400](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/eeg/03_eeg-rawdata01x400_graph.jpg)
![EEG - normalized 20x400](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/eeg/04_eeg-normmin_maxdata01x400_graph.jpg)

### Features

Expand Down Expand Up @@ -144,11 +155,15 @@ Data was collected over a 36-month period, from January 2008 to February 2011, a
The dataset has a time granularity of 6 hours and consists of 100 time series, each containing 1,000 data points. This dataset is particularly valuable for testing algorithms designed to handle drift and outliers.

### Plots
The plots present a series of plots derived from the Drift dataset, illustrating various aspects of the data and preprocessing steps.
Drift dataset - raw data 64x256 shows the full raw dataset, consisting of NxM time series.
Drift dataset - raw data 20x400 provides a subset of the data, limited to 20 time series over 400 time steps, while Drift dataset - raw data 01x400 focuses on a single time series extracted from the dataset.
Finally, Drift - normalized 20x400 demonstrates the impact of "MIN-MAX" normalization on the raw data, applied to the same 20x400 subset.

![Drift dataset - raw data 33x1201](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/drift/01_drift-rawdata-NxM_graph.jpg)
![Drift dataset - raw data 20x400](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/drift/02_drift-rawdata20x400_graph.jpg)
![Drift dataset - raw data 01x400](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/drift/03_drift-rawdata01x400_graph.jpg)
![Drift - normalized 20x400](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/drift/04_drift-normmin_maxdata01x400_graph.jpg)
![Drift dataset - raw data 64x256](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/drift/01_drift-rawdata-NxM_graph.jpg)
![Drift dataset - raw data 20x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/drift/02_drift-rawdata20x400_graph.jpg)
![Drift dataset - raw data 01x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/drift/03_drift-rawdata01x400_graph.jpg)
![Drift - normalized 20x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/drift/04_drift-normmin_maxdata01x400_graph.jpg)

### Features

Expand Down Expand Up @@ -203,11 +218,15 @@ The dataset exhibits a cyclic pattern with recurring peaks, suggesting the prese
This makes it particularly well-suited for time series imputation methods that are designed to detect and leverage seasonality in data.

### Plots
The plots present a series of plots derived from the Chlorine dataset, illustrating various aspects of the data and preprocessing steps.
Chlorine dataset - raw data 64x256 shows the full raw dataset, consisting of NxM time series.
Chlorine dataset - raw data 20x400 provides a subset of the data, limited to 20 time series over 400 time steps, while Chlorine dataset - raw data 01x400 focuses on a single time series extracted from the dataset.
Finally, Chlorine - normalized 20x400 demonstrates the impact of "MIN-MAX" normalization on the raw data, applied to the same 20x400 subset.

![Chlorine dataset - raw data 33x1201](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/chlorine/01_chlorine-rawdata-NxM_graph.jpg)
![Chlorine dataset - raw data 20x400](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/chlorine/02_chlorine-rawdata20x400_graph.jpg)
![Chlorine dataset - raw data 01x400](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/chlorine/03_chlorine-rawdata01x400_graph.jpg)
![Chlorine - normalized 20x400](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/chlorine/04_chlorine-normmin_maxdata01x400_graph.jpg)
![Chlorine dataset - raw data 64x256](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/chlorine/01_chlorine-rawdata-NxM_graph.jpg)
![Chlorine dataset - raw data 20x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/chlorine/02_chlorine-rawdata20x400_graph.jpg)
![Chlorine dataset - raw data 01x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/chlorine/03_chlorine-rawdata01x400_graph.jpg)
![Chlorine - normalized 20x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/chlorine/04_chlorine-normmin_maxdata01x400_graph.jpg)

### Features

Expand Down Expand Up @@ -260,11 +279,15 @@ This structure is particularly valuable for spatio-temporal modeling [Spatial-te
The dataset exhibits high variability, along with periodic or trend-like behavior and noise, making it suitable for advanced analytical techniques.

### Plots
The plots present a series of plots derived from the Climate dataset, illustrating various aspects of the data and preprocessing steps.
Climate dataset - raw data 64x256 shows the full raw dataset, consisting of NxM time series.
Climate dataset - raw data 20x400 provides a subset of the data, limited to 20 time series over 400 time steps, while Climate dataset - raw data 01x400 focuses on a single time series extracted from the dataset.
Finally, Climate - normalized 20x400 demonstrates the impact of "MIN-MAX" normalization on the raw data, applied to the same 20x400 subset.

![Climate dataset - raw data 33x1201](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/climate/01_climate-rawdata-NxM_graph.jpg)
![Climate dataset - raw data 20x400](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/climate/02_climate-rawdata20x400_graph.jpg)
![Climate dataset - raw data 01x400](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/climate/03_climate-rawdata01x400_graph.jpg)
![Climate - normalized 20x400](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/climate/04_climate-normmin_maxdata01x400_graph.jpg)
![Climate dataset - raw data 64x256](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/climate/01_climate-rawdata-NxM_graph.jpg)
![Climate dataset - raw data 20x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/climate/02_climate-rawdata20x400_graph.jpg)
![Climate dataset - raw data 01x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/climate/03_climate-rawdata01x400_graph.jpg)
![Climate - normalized 20x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/climate/04_climate-normmin_maxdata01x400_graph.jpg)

### Features

Expand Down Expand Up @@ -312,11 +335,15 @@ The dataset exhibits high variability, along with periodic or trend-like behavio
## Meteo

### Plots
The plots present a series of plots derived from the Meteo dataset, illustrating various aspects of the data and preprocessing steps.
Meteo dataset - raw data 64x256 shows the full raw dataset, consisting of NxM time series.
Meteo dataset - raw data 20x400 provides a subset of the data, limited to 20 time series over 400 time steps, while Meteo dataset - raw data 01x400 focuses on a single time series extracted from the dataset.
Finally, Meteo - normalized 20x400 demonstrates the impact of "MIN-MAX" normalization on the raw data, applied to the same 20x400 subset.

![Meteo dataset - raw data 33x1201](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/meteo/01_meteo-rawdata-NxM_graph.jpg)
![Meteo dataset - raw data 20x400](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/meteo/02_meteo-rawdata20x400_graph.jpg)
![Meteo dataset - raw data 01x400](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/meteo/03_meteo-rawdata01x400_graph.jpg)
![Meteo - normalized 20x400](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/meteo/04_meteo-normmin_maxdata01x400_graph.jpg)
![Meteo dataset - raw data 64x256](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/meteo/01_meteo-rawdata-NxM_graph.jpg)
![Meteo dataset - raw data 20x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/meteo/02_meteo-rawdata20x400_graph.jpg)
![Meteo dataset - raw data 01x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/meteo/03_meteo-rawdata01x400_graph.jpg)
![Meteo - normalized 20x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/meteo/04_meteo-normmin_maxdata01x400_graph.jpg)

### Features

Expand Down Expand Up @@ -379,11 +406,15 @@ Example: 13 = observation period 12:41 to 13:40
## BAFU

### Plots
The plots present a series of plots derived from the BAFU dataset, illustrating various aspects of the data and preprocessing steps.
BAFU dataset - raw data 64x256 shows the full raw dataset, consisting of NxM time series.
BAFU dataset - raw data 20x400 provides a subset of the data, limited to 20 time series over 400 time steps, while BAFU dataset - raw data 01x400 focuses on a single time series extracted from the dataset.
Finally, BAFU - normalized 20x400 demonstrates the impact of "MIN-MAX" normalization on the raw data, applied to the same 20x400 subset.

![Bafu dataset - raw data 33x1201](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/bafu/01_bafu-rawdata-NxM_graph.jpg)
![Bafu dataset - raw data 20x400](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/bafu/02_bafu-rawdata20x400_graph.jpg)
![Bafu dataset - raw data 01x400](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/bafu/03_bafu-rawdata01x400_graph.jpg)
![Bafu - normalized 20x400](https://github.com/eXascaleInfolab/ImputeGAP/tree/main/imputegap/dataset/docs/bafu/04_bafu-normmin_maxdata01x400_graph.jpg)
![BAFU dataset - raw data 64x256](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/bafu/01_bafu-rawdata-NxM_graph.jpg)
![BAFU dataset - raw data 20x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/bafu/02_bafu-rawdata20x400_graph.jpg)
![BAFU dataset - raw data 01x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/bafu/03_bafu-rawdata01x400_graph.jpg)
![BAFU - normalized 20x400](https://github.com/eXascaleInfolab/ImputeGAP/raw/main/imputegap/dataset/docs/bafu/04_bafu-normmin_maxdata01x400_graph.jpg)

### Features

Expand Down Expand Up @@ -417,7 +448,7 @@ Example: 13 = observation period 12:41 to 13:40

| Data info | |
|--------------------|---------------------------------------------|
| Dataset codename | BAFU/bafu |
| Dataset codename | BAFU<br/>bafu |
| Dataset name | Hydrological data across multiple stations |
| Url/source | https://www.bafu.admin.ch/bafu/en/home.html |
| Granularity | 30 minutes |
Expand Down

0 comments on commit 2192dfe

Please sign in to comment.