Benchmarks
| Name | Description of Benchmark |
|---|---|
| SynMS | The PLANET SynMS (synthetic manuscript) family is a collection of databases of images of synthetically generated cursive writings.... |
| Name | Description of Benchmark |
|---|---|
| Counting | In this task we want to test the ability of counting. The goal is to predict all the next possible symbols of a string given the previous symbols.... |
| Name | Description of Benchmark |
|---|---|
| Place cell learning | Dataset The dataset consists of 17 distance sensors recorded during navigation of a mobile robot in a simulated environment called SINAR. SINAR... |
| NARMA30 classification | In this task data sampled from a NARMA system with a varying set of parameters must be instantaneous labeled correctly based on the input and output... |
| Name | Description of Benchmark |
|---|---|
| AURORA-2 | The AURORA-2 corpus (version 1.0) contains clean and noise corrupted versions of the isolated and connected digit utterances of the TIDIGITS corpus (... |
| TIMIT | The TIMIT Acoustic-Phonetic Continuous Speech Corpus, distributed by LDC (reference LDC93S1) is a relatively small corpus (1 CD) of read speech, and... |
| WSJ | The Wall Street Journal (WSJ) corpus, distributed by LDC, consists of read speech with texts drawn from a machine-readable corpus of Wall Street... |
| Name | Description of Benchmark |
|---|---|
| NARMA30 | This task concerns the modeling of the following 30th order system also known as NARMA which can be written as: For this task the output y(k) has to... |
| Name | Description of Benchmark |
|---|---|
| Triple generator | Three simple generators: Tent map, Sine, Random constant value are interleaving with equal probabilities. The task is to predict the time series as... |
| Mackey-Glass attractor | This task tackles the continuation of a Mackey-Glass attractor with delay tau 30. Therefore 10 sets of 3000 samples are sampled from the following... |
