Skip to main content
ConductScienceScienceDex
Powered by ShareScore

Find research datasets worth reusing

Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.

29

datasets available to search

ShareScore release 0.9.0

Reset

Dataset results

29 results for “Few-shot”

Learn how ShareScore rates datasets ↗
zenodo48/100

MEWL: Few-shot multimodal word learning with referential uncertainty

<p><strong>Dataset Release for <a href="https://arxiv.org/abs/2306.00503">MEWL: Few-shot multimodal word learning with referential uncertainty&nbsp;(ICML 2023)&nbsp;</a></strong></p> <p><strong>GitHub:</strong> <a href="https://github.com/jianggy/MEWL">https://github.com/jianggy/MEWL</a></p> <p><strong>Abstract: </strong>Without explicit feedback, humans can rapidly learn the meaning of words. Children can acquire a new word after just a few passive exposures, a process known as fast mapping. This word learning capability is believed to be the most fundamental building block of multimodal understanding and reasoning. Despite recent advancements in multimodal learning, a systematic and rigorous evaluation is still missing for human-like word learning in machines. To fill in this gap, we introduce the MachinE Word Learning (MEWL) benchmark to assess how machines learn word meaning in grounded visual scenes. MEWL covers human&#39;s core cognitive toolkits in word learning: cross-situational reasoning, bootstrapping, and pragmatic learning. Specifically, MEWL is a few-shot benchmark suite consisting of nine tasks for probing various word learning capabilities. These tasks are carefully designed to be aligned with the children&#39;s core abilities in word learning and echo the theories in the developmental literature. By evaluating multimodal and unimodal agents&#39; performance with a comparative analysis of human performance, we notice a sharp divergence in human and machine word learning. We further discuss these differences between humans and machines and call for human-like few-shot word learning in machines.</p>

opencc-by-4.0May 2023View details →
zenodo48/100

learn2learn: Few-Shot Learning Datasets

<p>Few-shot learning datasets, including:</p> <ul> <li>mini-ImageNet</li> <li>tiered-ImageNet</li> <li>CIFAR-FS</li> <li>FC100</li> </ul> <p>New in 1.0.1:</p> <ul> <li>FGVC Fungi</li> <li>FGVC Aircrafts</li> <li>Describable Textures</li> <li>VGG Flowers</li> <li>CUB200</li> </ul> <p>Please cite the respective datasets if you use them, not this archive.</p>

opencc-by-4.0May 2023View details →
zenodo44/100

An Open-set Recognition and Few-Shot Learning Dataset for Audio Event Classification in Domestic Environments

<p>The problem of training a deep neural network with a small set of positive samples is known as few-shot learning (FSL). It is widely known that traditional deep learning (DL) algorithms usually show very good performance when trained with large datasets. However, in many applications, it is not possible to obtain such a high number of samples. In the image domain, typical FSL applications are those related to face recognition. In the audio domain, music fraud or speaker recognition can be clearly benefited from FSL methods. This paper deals with the application of FSL to the detection of specific and intentional acoustic events given by different types of sound alarms, such as door bells or fire alarms, using a limited number of samples. These sounds typically occur in domestic environments where many events corresponding to a wide variety of sound classes take place. Therefore, the detection of such alarms in a practical scenario can be considered an open-set recognition (OSR) problem. To address the lack of a dedicated public dataset for audio FSL, researchers usually make modifications on other available datasets. This paper is aimed at providing the audio recognition community with a carefully annotated dataset for FSL and OSR comprised of 1360 clips from 34 classes divided into pattern sounds&nbsp;and unwanted sounds. To facilitate and promote research in this area, results with two baseline systems (one trained from scratch and another based on transfer learning), are presented.</p> <p>&nbsp;</p>

opencc-by-4.0Feb 2020View details →
zenodo44/100

DCASE 2024 Task 5: Few-shot Bioacoustic Event Detection Development Set

<p><strong>General Description:</strong></p> <p>The development set for task 5 of DCASE 2024 "Few-shot Bioacoustic Event Detection" consists of 217 audio files acquired from different bioacoustic sources. The dataset is split into training and validation sets.&nbsp;</p> <p>Multi-class annotations are provided for the training set with positive (POS), negative (NEG) and unkwown (UNK) values for each class. UNK indicates uncertainty about a class.&nbsp;</p> <p>Single-class (class of interest) annotations are provided for the validation set, with events marked as positive (POS) or unkwown (UNK) provided for the class of interest.&nbsp;</p> <p><strong>Folder Structure:</strong></p> <p><em>Development_set.zip</em></p> <p>|_Development_Set/</p> <p>&nbsp; &nbsp; |__Training_Set/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___JD/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___HT/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___BV/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___MT/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___WMW/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp;</p> <p>&nbsp; &nbsp; |__Validation_Set/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___HB/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___PB/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___ME/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; &nbsp; &nbsp;|___PB24/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___RD/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___PW/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp;</p> <p><em>Development_set_annotations.zip</em> has the same structure but contains only the *.csv files</p> <p>&nbsp;</p> <p><strong>Dataset statistics</strong></p> <p>Some statistics on this dataset are as follows, split between training and validation set and their sub-folders:</p> <p>-----------------------------------------------------<br>TRAINING SET<br>-----------------------------------------------------<br>Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;174<br>Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;21 hours<br>Total classes&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;47<br>Total events&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;14229<br>-----------------------------------------------------<br>TRAINING SET/BV<br>-----------------------------------------------------<br>Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;5<br>Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;10 hours<br>Total classes &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;11<br>Total events &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;9026<br>Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;24000 Hz<br>-----------------------------------------------------<br>TRAINING SET/HT<br>-----------------------------------------------------<br>Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;5<br>Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;5 hours<br>Total classes &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;5<br>Total events &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;611<br>Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;6000 Hz<br>-----------------------------------------------------<br>TRAINING SET/JD<br>-----------------------------------------------------<br>Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;1<br>Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;10 mins<br>Total classes&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;1<br>Total events&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;357<br>Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;22050 Hz<br>-----------------------------------------------------<br>TRAINING SET/MT<br>-----------------------------------------------------<br>Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2<br>Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;1 hour and 10 mins<br>Total classes&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;4<br>Total events&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;1294<br>Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;8000 Hz<br>-----------------------------------------------------<br>TRAINING SET/WMW<br>-----------------------------------------------------<br>Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;161<br>Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;4 hours and 40 mins<br>Total classes&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;26<br>Total events&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2941<br>Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;various sampling rates<br>-----------------------------------------------------</p> <p>-----------------------------------------------------<br>VALIDATION SET<br>-----------------------------------------------------<br>Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;| &nbsp; &nbsp;43<br>Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;| &nbsp; &nbsp;49 hours and 57 minutes<br>Total classes&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;| &nbsp; &nbsp;7<br>Total events &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;| &nbsp; &nbsp;3504<br>-----------------------------------------------------<br>VALIDATION SET/HB<br>-----------------------------------------------------<br>Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;10<br>Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2 hours and 38 minutes<br>Total classes &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;1<br>Total events &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;712<br>Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;44100 Hz<br>-----------------------------------------------------<br>VALIDATION SET/PB<br>-----------------------------------------------------<br>Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;6<br>Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;3 hours<br>Total classes&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2<br>Total events &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;292<br>Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;44100 Hz<br>-----------------------------------------------------<br>VALIDATION SET/ME<br>-----------------------------------------------------<br>Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2<br>Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;20 minutes<br>Total classes&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2<br>Total events &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;73<br>Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;44100 Hz<br>-----------------------------------------------------<br>VALIDATION SET/PB24<br>-----------------------------------------------------<br>Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;| &nbsp; &nbsp;4<br>Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2 hours<br>Total classes&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2<br>Total events &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;| &nbsp; &nbsp;350<br>Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;44100 Hz<br>-----------------------------------------------------<br>VALIDATION SET/RD<br>-----------------------------------------------------<br>Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;| &nbsp; &nbsp;6<br>Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; 18 hours<br>Total classes&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;| &nbsp; &nbsp;1<br>Total events &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;| &nbsp; &nbsp;1372<br>Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;48000 Hz<br>-----------------------------------------------------<br>VALIDATION SET/PW<br>-----------------------------------------------------<br>Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;| &nbsp; &nbsp;15<br>Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;24 hours<br>Total classes&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;| &nbsp; &nbsp;1<br>Total events &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;| &nbsp; &nbsp;705<br>Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;| &nbsp; &nbsp;96000 Hz<br>-----------------------------------------------------</p> <p><strong>Annotation structure</strong></p> <p>Each line of the annotation csv represents an event in the audio file. The column descriptions are as follows:</p> <p>TRAINING SET<br>---------------------<br>Audiofilename, Starttime, Endtime, CLASS_1, CLASS_2, ...CLASS_N</p> <p>VALIDATION SET<br>---------------------<br>Audiofilename, Starttime, Endtime, Q</p> <p>&nbsp;</p> <p><strong>Classes</strong></p> <p>DCASE2024_task5_training_set_classes.csv and DCASE2024_task5_validation_set_classes.csv provide a table with class code correspondence to class name for all classes in the Development set. Additionally, DCASE2024_task5_validation_set_classes.csv also provides a recording names column.</p> <p>DCASE2024_task5_training_set_classes.csv<br>---------------------<br>dataset, class_code, class_name</p> <p>DCASE2024_task5_validation_set_classes.csv<br>---------------------<br>dataset, recording, class_code, class_name</p> <p>&nbsp;</p> <p><strong>Evaluation Set</strong></p> <p>The Evaluation set for this task will be released on the 1 June 2024</p> <p><strong>Open Access:</strong></p> <p>This dataset is available under a Creative Commons Attribution 4.0 International (CC BY 4.0) license.<br>&nbsp;</p> <p><strong>Contact info:</strong></p> <p>Please send any feedback or questions to:</p> <p>Burooj Ghani - &nbsp;burooj.ghani@naturalis.nl | Ines Nolasco - i.dealmeidanolasco@qmul.ac.uk</p> <p>Alternately, join us on Slack: <a href="https://join.slack.com/t/dcase/shared_invite/zt-12zfa5kw0-dD41gVaPU3EZTCAw1mHTCA">task-fewshot-bio-sed</a></p> <p>&nbsp;</p>

opencc-by-4.0Mar 2024View details →
zenodo44/100

DCASE 2021 Task 5: Few-shot Bioacoustic Event Detection Evaluation Set

<p><strong>General Description</strong></p> <p>The evaluation set for task 5 of DCASE 2021 &quot;Few-shot Bioacoustic Event Detection&quot; consists of 31 audio files acquired from different bioacoustic sources.&nbsp;</p> <p>In Evaluation_Set_Annotations: the first 5 annotations are provided for each file, with events marked as positive (POS) for the class of interest. This is the same setup used during the DCASE 2021 challenge.</p> <p>In Evaluation_Set_Full_Annotations: the full annotations are provided for each file, with events marked as positive (POS) or unknown (UNK) for the class of interest.</p> <p>&nbsp;</p> <p><strong>Folder Structure</strong></p> <p><em>Evaluation_Set.zip contains audio files and annotation files with 5 first POS events (as used during DCASE 2021 challenge)</em></p> <p>|__Evaluation_Set/</p> <p>&nbsp; &nbsp; |___DC/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; |___ME/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; |___ML/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p><em>Evaluation_Set_Audio.zip</em>&nbsp;has the same structure but contains only the *.wav files.</p> <p><em>Evaluation_Set_Annotations.zip</em>&nbsp;has the same structure but contains only the *.csv files with first 5 POS annotations.</p> <p><em>Evaluation_Set_Full_Annotations.zip</em>&nbsp;has the same structure but contains only the *.csv files with all POS annotations.</p> <p>The subfolders denote different recording sources and there may or may not be overlap between classes of interest from different wav files.</p> <p>&nbsp;</p> <p><strong>Annotation structure</strong></p> <p>Each line of the annotation csv represents an event in the audio file. The column descriptions are as follows:<br> [ Audiofilename, Starttime, Endtime, Q ]</p> <p>&nbsp;</p> <p><strong>Classes</strong></p> <p>DCASE2021_task5_evaluation_set.csv provides a table with class code correspondance to class name for all the recordings of the Evaluation set.</p> <p>DCASE2021_task5_evaluation_set.csv<br> -------------------<br> dataset, recording, class_code, class_name</p> <p>&nbsp;</p> <p><strong>Development Set</strong></p> <p>The development set for the same task can be found at:&nbsp;<a href="https://doi.org/10.5281/zenodo.5412896">https://doi.org/10.5281/zenodo.5412896</a></p> <p>&nbsp;</p> <p><strong>Open Access</strong></p> <p>This dataset is available under a Creative Commons Attribution 4.0 International (CC BY 4.0) license.<br> &nbsp;</p> <p><strong>Contact info</strong></p> <p>Please send any feedback or questions to:<br> Veronica Morfi: g.v.morfi@qmul.ac.uk</p>

opencc-by-4.0May 2021View details →
zenodo44/100

DCASE 2021 Task 5: Few-shot Bioacoustic Event Detection Development Set

<p><strong>General Description</strong></p> <p>The development set for task 5 of DCASE 2021 &quot;Few-shot Bioacoustic Event Detection&quot; consists of 19 audio files acquired from different bioacoustic sources. The dataset is split into training and validation Sets.&nbsp;</p> <p>Multi-class annotations are provided for the training set with positive (POS), negative (NEG) and unkwown (UNK) values for each class. UNK indicates uncertainty about a class.&nbsp;</p> <p>Single-class (class of interest) annotations are provided for the validation set, with events marked as positive (POS) or unkwown (UNK) provided for the class of interest.&nbsp;</p> <p>&nbsp;</p> <p><strong>Folder Structure</strong></p> <p><em>Development_Set.zip</em></p> <p>|_Development_Set/</p> <p>&nbsp; &nbsp; |__Training_Set/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___BV/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___HT/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___JD/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___MT/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; |__Validation_Set/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___HV/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___PB/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp;</p> <p><em>Development_Set_Audio.zip</em> has the same structure but contains only the *.wav files.</p> <p><em>Development_Set_Annotations.zip</em> has the same structure but contains only the *.csv files</p> <p>&nbsp;</p> <p><strong>Dataset statistics</strong></p> <p>Some statistics on this dataset are as follows, split between training and validation set and their sub-folders:</p> <p>-----------------------------------------------------<br> TRAINING SET<br> -----------------------------------------------------<br> Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;11<br> Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;14 hours and 20 mins<br> Total classes (excl. UNK)&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;19<br> Total events (excl. UNK)&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;4,686<br> -----------------------------------------------------<br> TRAINING SET/BV<br> -----------------------------------------------------<br> Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;5<br> Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;10 hours<br> Total classes (excl. UNK)&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;11<br> Total events (excl. UNK)&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2,662<br> Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;24,000 Hz<br> -----------------------------------------------------<br> TRAINING SET/HT<br> -----------------------------------------------------<br> Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;3<br> Total duration&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |&nbsp;&nbsp; &nbsp;3 hours<br> Total classes (excl. UNK)&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;3<br> Total events (excl. UNK)&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;435<br> Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;6,000 Hz<br> -----------------------------------------------------<br> TRAINING SET/JD<br> -----------------------------------------------------<br> Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;1<br> Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;10 mins<br> Total classes (excl. UNK)&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;1<br> Total events (excl. UNK)&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;355<br> Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;22,050 Hz<br> -----------------------------------------------------<br> TRAINING SET/MT<br> -----------------------------------------------------<br> Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2<br> Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;1 hour and 10 mins<br> Total classes (excl. UNK)&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;4<br> Total events (excl. UNK)&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;1,234<br> Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;8,000 Hz<br> -----------------------------------------------------</p> <p><br> -----------------------------------------------------<br> VALIDATION SET<br> -----------------------------------------------------<br> Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;8<br> Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;5 hours<br> Total classes (excl. UNK)&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;4<br> Total events (excl. UNK)&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;310<br> -----------------------------------------------------<br> VALIDATION SET/HV<br> -----------------------------------------------------<br> Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2<br> Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2 hours<br> Total classes (excl. UNK)&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2<br> Total events (excl. UNK)&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;50<br> Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;6,000 Hz<br> -----------------------------------------------------<br> VALIDATION SET/PB<br> -----------------------------------------------------<br> Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;6<br> Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;3 hours<br> Total classes (excl. UNK)&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2<br> Total events (excl. UNK)&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;260<br> Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;44,100 Hz<br> -----------------------------------------------------</p> <p>&nbsp;</p> <p><strong>Annotation structure</strong></p> <p>Each line of the annotation csv represents an event in the audio file. The column descriptions are as follows:</p> <p>TRAINING SET<br> ---------------------<br> Audiofilename, Starttime, Endtime, CLASS_1, CLASS_2, ...CLASS_N</p> <p>VALIDATION SET<br> ---------------------<br> Audiofilename, Starttime, Endtime, Q</p> <p>&nbsp;</p> <p><strong>Classes</strong></p> <p>DCASE2021_task5_training_set_classes.csv and&nbsp;DCASE2021_task5_validation_set_classes.csv provide a table with class code&nbsp;correspondace to class name for all classes in the Development set.</p> <p>DCASE2021_task5_training_set_classes.csv<br> ---------------------<br> dataset, class_code, class_name</p> <p>DCASE2021_task5_validation_set_classes.csv<br> ---------------------<br> dataset, recording, class_code, class_name</p> <p>&nbsp;</p> <p><strong>Evaluation Set</strong></p> <p>The Evaluation set for the same task can be found at:&nbsp;<a href="https://doi.org/10.5281/zenodo.5413149">https://doi.org/10.5281/zenodo.5413149</a></p> <p>&nbsp;</p> <p><strong>Open Access</strong></p> <p>This dataset is available under a Creative Commons Attribution 4.0 International (CC BY 4.0) license.</p> <p><br> <strong>Contact info</strong></p> <p>Please send any feedback or questions to:<br> Veronica Morfi: g.v.morfi@qmul.ac.uk<br> &nbsp;</p>

opencc-by-4.0Feb 2021View details →
zenodo44/100

Supporting information for "Few-Shot Learning Enables Population-Scale Analysis of Leaf Traits in Populus trichocarpa"

<p><strong>Description</strong></p> <p>In this work, we use few-shot learning to segment the body and vein architecture of&nbsp;<em>P. trichocarpa</em>&nbsp;leaves from high-resolution scans obtained in the UC Davis common garden. Leaf and vein segmentation are formulated as separate tasks, in which convolutional neural networks (CNNs) are used to iteratively expand partial segmentations until reaching stopping criteria. Our leaf and vein segmentation approaches use just 50 and 8 manually traced images for training, respectively, and are applied to a set of 2,634 top and bottom leaf scans. We show that both methods achieve high segmentation accuracy and retain biologically realistic features. The leaf and vein segmentations are compared against a U-Net baseline model, and subsequently used to extract 68 morphological traits using traditional open-source image processing tools, which are validated using real-world physical measurements. For a biological perspective, we perform a genome-wide association study using the &quot;vein density&quot; trait to discover novel genetic architectures associated with multiple physiological processes relating to leaf development and function. In addition to sharing all of the few-shot learning code, we are releasing all images, manual segmentations, model predictions, 68 extracted leaf phenotypes, and a new set of SNPs called against the v4&nbsp;<em>P. trichocarpa</em>&nbsp;genome for 1,419 genotypes.</p> <p><strong>Directories:</strong></p> <pre><code>Few-shot learning for p. trichocarpa leaf traits ├── data │ ├── genomes │ │ ├── Ptri_V4_Nisq1.[...].bed │ │ ├── Ptri_V4_Nisq1.[...].bim │ │ └── Ptri_V4_Nisq1.[...].fam │ ├── images │ │ └── *.jpeg │ ├── leaf_masks │ │ └── *.png │ ├── leaf_preds │ │ └── *.png │ ├── leaf_unet_preds │ │ └── *.png │ ├── results │ │ ├── digital_traits.tsv │ │ ├── gwas_results.csv │ │ ├── manual_traits.tsv │ │ ├── vein_density_blups.tsv │ │ └── vein_density_tps_adj.tsv │ ├── vein_bce_preds │ │ └── *.png │ ├── vein_bce_probs │ │ └── *.png │ ├── vein_fl_preds │ │ └── *.png │ ├── vein_fl_probs │ │ └── *.png │ ├── vein_masks │ │ └── *.png │ ├── vein_unet_bce_preds │ │ └── *.png │ ├── vein_unet_bce_probs │ │ └── *.png │ ├── vein_unet_fl_preds │ │ └── *.png │ ├── vein_unet_fl_probs │ │ └── *.png ├── figures │ └── *.png ├── logs │ ├── leaf_tracer_256.txt │ ├── leaf_unet_256.txt │ ├── vein_grower_bce_128.txt │ ├── vein_grower_fl_128.txt │ ├── vein_unet_bce_128.txt │ └── vein_unet_fl_128.txt ├── models │ ├── BuildCNN.py │ ├── BuildUNet.py │ ├── LeafTracer.py │ └── VeinGrower.py ├── notebooks │ ├── Figures.ipynb │ ├── GrowerInference.ipynb │ ├── GrowerTraining.ipynb │ ├── TracerInference.ipynb │ ├── TracerTraining.ipynb │ ├── UNetLeafSegmentation.ipynb │ └── UNetVeinSegmentation.ipynb ├── utils │ ├── GetLowestGPU.py │ ├── ImageLoader.py │ ├── LeafGenerator.py │ ├── ModelWrapperGenerator.py │ ├── TimeRemaining.py │ ├── TraceInitializer.py │ ├── UNetTileGenerator.py │ └── VeinGenerator.py └── weights ├── leaf_tracer_256_best_val_model.save ├── leaf_unet_256_best_val_model.save ├── vein_grower_bce_128_best_val_model.save ├── vein_grower_fl_128_best_val_model.save ├── vein_unet_bce_128_best_val_model.save └── vein_unet_fl_128_best_val_model.save </code></pre> <p><strong>Data:</strong></p> <p>The&nbsp;<code>data</code>&nbsp;folder includes all images, ground truth segmentations, predicted segmentations, and extracted leaf traits. All images encode the sample ID in the file name by indicating the treatment, block, row, position, and leaf side, respectively. For example, the file,&nbsp;<code>C_1_1_2_bot.jpeg</code>, indicates the control treatment, block 1, row 1, position 2, and the bottom side of the leaf. Tabulated results include position IDs as well as the corresponding genotype IDs.</p> <ul> <li>The&nbsp;<code>images</code>&nbsp;folder includes the 2,906 high-resolution leaf scans taken in the field.</li> <li>The&nbsp;<code>leaf_masks</code>&nbsp;folder includes 50 ground truth segmentations used for training the leaf tracing algorithm.</li> <li>The&nbsp;<code>leaf_preds</code>&nbsp;folder includes the 2,906 predicted segmentations from the leaf tracing algorithm.</li> <li>The&nbsp;<code>leaf_unet_preds</code>&nbsp;folder includes the 2,906 predicted segmentations from the U-Net model for leaf segmentation.</li> <li>The&nbsp;<code>vein_masks</code>&nbsp;folder includes 8 ground truth segmentations used for training the vein growing algorithm.</li> <li>The&nbsp;<code>vein_*_preds</code>&nbsp;folder includes the 1,453 predicted segmentations from the vein growing algorithm, where * specifies the loss function (bce: binary cross-entropy, fl: focal loss).</li> <li>The&nbsp;<code>vein_*_probs</code>&nbsp;folder includes the 1,453 predicted probability maps from the vein growing algorithm before thresholding, where * specifies the loss function (bce: binary cross-entropy, fl: focal loss).</li> <li>The&nbsp;<code>vein_unet_*_preds</code>&nbsp;folder includes the 1,453 predicted segmentations from the U-Net model for vein segmentation, where * specifies the loss function (bce: binary cross-entropy, fl: focal loss).</li> <li>The&nbsp;<code>vein_unet_*_probs</code>&nbsp;folder includes the 1,453 predicted probability maps from the U-Net model for vein segmentation before thresholding, where * specifies the loss function (bce: binary cross-entropy, fl: focal loss).</li> <li>The&nbsp;<code>genomes</code>&nbsp;folder includes the set of SNPs called against the v4&nbsp;<em>P. trichocarpa</em>&nbsp;genome for 1,419 genotypes with a README file detailing the steps taken.</li> <li>The&nbsp;<code>results</code>&nbsp;folder includes <ul> <li>Raw values of the 68 predicted leaf traits in&nbsp;<code>digital_traits.tsv</code></li> <li>Manually measured values of petiole length and width in&nbsp;<code>manual_traits.tsv</code></li> <li>Thin plate spline (TPS) adjusted values of the vein density trait in&nbsp;<code>vein_density_tps_adj.tsv</code></li> <li>Best linear unbiased prediction (BLUP) adjusted values of the vein density trait in&nbsp;<code>vein_density_blups.tsv</code></li> <li>GWAS results for the vein density trait, including chromosome positions and corresponding P values, in&nbsp;<code>gwas_results.csv</code></li> </ul> </li> </ul> <p><strong>Figures:</strong></p> <p>The&nbsp;<code>figures</code>&nbsp;folder includes all figures and videos used in the manuscript. See&nbsp;<code>notebooks/Figures.ipynb</code>&nbsp;for the methods used to generate these figures.</p> <p><strong>Logs:</strong></p> <p>The&nbsp;<code>logs</code>&nbsp;folder includes logs of CNN convergence for the training and validation sets during model training for the leaf tracing CNN vein growing CNN, and U-Net models. The file names include the model, loss function (bce: binary cross-entropy, fl: focal loss), and size of the input window for each method (e.g., 128 for the vein growing CNN).</p> <p><strong>Models:</strong></p> <p>The&nbsp;<code>models</code>&nbsp;folder includes the CNN implementations in PyTorch as well as the leaf tracing and vein growing algorithms at inference time.</p> <ul> <li><code>BuildCNN.py</code>&nbsp;defines the CNN architecture for leaf tracing or vein growing, with user-specified input shape, output shape, layers, and output activation functions.</li> <li><code>BuildUNet.py</code>&nbsp;defines the U-Net architecture for leaf and vein segmentation, with user-specified input/output shape, layers, and output activation functions.</li> <li><code>LeafTracer.py</code>&nbsp;defines the leaf tracing algorithm at inference time.</li> <li><code>VeinGrower.py</code>&nbsp;defines the vein growing algorithm at inference time.</li> </ul> <p><strong>Notebooks:</strong></p> <p>The&nbsp;<code>notebooks</code>&nbsp;folder includes Jupyter notebooks used for model training, model inference, and figure generation.</p> <ul> <li><code>Figures.ipynb</code>&nbsp;is used to generate all of the manuscript figures.</li> <li><code>GrowerTraining.ipynb</code>&nbsp;is used to train the vein growing CNN.</li> <li><code>GrowerInference.ipynb</code>&nbsp;is used to apply the vein growing algorithm to the 1,453 leaf bottom images.</li> <li><code>TracerTraining.ipynb</code>&nbsp;is used to train the leaf tracing CNN.</li> <li><code>TracerInference.ipynb</code>&nbsp;is used to apply the leaf tracing algorithm to the 2,906 leaf top and bottom images.</li> <li><code>UNetLeafSegmentation.ipynb</code>&nbsp;is used to train and apply U-Net for leaf segmentation.</li> <li><code>UNetVeinSegmentation.ipynb</code>&nbsp;is used to train and apply U-Net for vein segmentation.</li> </ul> <p><strong>Utils:</strong></p> <p>The&nbsp;<code>utils</code>&nbsp;folder includes utility scripts implemented in Python that assist in model training and inference.</p> <ul> <li><code>ImageLoader.py</code>&nbsp;loads image/mask pairs for sampling training/validation tiles.</li> <li><code>LeafGenerator.py</code>&nbsp;generates inputs/outputs for the leaf tracing CNN.</li> <li><code>VeinGenerator.py</code>&nbsp;generates inputs/outputs for the vein growing CNN.</li> <li><code>UNetTileGenerator.py</code>&nbsp;generates inputs/outputs for the U-Net model.</li> <li><code>GetLowestGPU.py</code>&nbsp;identifies available GPUs using the&nbsp;<code>nvidia-smi</code>&nbsp;command and selects the one with lowest memory usage, if none available the device is set to CPU.</li> <li><code>ModelWrapperGenerator.py</code>&nbsp;wraps the PyTorch CNN and data loaders with similar functionality to the Keras Model class in TensorFlow (e.g., model.fit(...)).</li> <li><code>TimeRemaining.py</code>&nbsp;is used by the model wrapper to estimate remaining time left per epoch.</li> <li><code>TraceInitializer.py</code>&nbsp;is used by the tracing algorithm at inference time to initialize the leaf trace using automatic thresholding.</li> </ul> <p><strong>Weights:</strong></p> <p>The&nbsp;<code>weights</code>&nbsp;folder includes the CNN parameters from the epoch resulting in the best validation error. The file names include the model, loss function (bce: binary cross-entropy, fl: focal loss), and size of the input window for each method (e.g., 128 for the vein growing CNN). The weights are loaded into the CNN models for inference.</p> <p><strong>Citation:</strong></p> <pre><code>@article{ doi:10.34133/plantphenomics.0072, author = {John Lagergren and Mirko Pavicic and Hari B. Chhetri and Larry M. York and Doug Hyatt and David Kainer and Erica M. Rutter and Kevin Flores and Jack Bailey-Bale and Marie Klein and Gail Taylor and Daniel Jacobson and Jared Streich }, title = {Few-Shot Learning Enables Population-Scale Analysis of Leaf Traits in Populus trichocarpa}, journal = {Plant Phenomics}, volume = {0}, number = {ja}, pages = {}, year = {}, doi = {10.34133/plantphenomics.0072}, URL = {https://spj.science.org/doi/abs/10.34133/plantphenomics.0072}, eprint = {https://spj.science.org/doi/pdf/10.34133/plantphenomics.0072}, }</code></pre>

opengpl-2.0Jan 2023View details →
zenodo40/100

DCASE 2022 Task 5: Few-shot Bioacoustic Event Detection Development Set

<p><strong>General Description:</strong></p> <p>The development set for task 5 of DCASE 2022&nbsp;&quot;Few-shot Bioacoustic Event Detection&quot; consists of 192 audio files acquired from different bioacoustic sources. The dataset is split into training and validation sets.&nbsp;</p> <p>Multi-class annotations are provided for the training set with positive (POS), negative (NEG) and unkwown (UNK) values for each class. UNK indicates uncertainty about a class.&nbsp;</p> <p>Single-class (class of interest) annotations are provided for the validation set, with events marked as positive (POS) or unkwown (UNK) provided for the class of interest.&nbsp;</p> <p><strong>this version (3):</strong><br> * fixes issues with annotations from HB set</p> <p>&nbsp;</p> <p><strong>Folder Structure:</strong></p> <p><em>Development_Set.zip</em></p> <p>|_Development_Set/</p> <p>&nbsp; &nbsp; |__Training_Set/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___JD/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___HT/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___BV/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___MT/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___WMW/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp;</p> <p>&nbsp; &nbsp; |__Validation_Set/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___HB/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___PB/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |___ME/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp;</p> <p><em>Development_Set_Annotations.zip</em>&nbsp;has the same structure but contains only the *.csv files</p> <p>&nbsp;</p> <p><strong>## Dataset statistics</strong></p> <p>Some statistics on this dataset are as follows, split between training and validation set and their sub-folders:</p> <p>-----------------------------------------------------<br> TRAINING SET<br> -----------------------------------------------------<br> Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;174<br> Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;21 hours<br> Total classes&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;47<br> Total events&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;14229<br> -----------------------------------------------------<br> TRAINING SET/BV<br> -----------------------------------------------------<br> Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;5<br> Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;10 hours<br> Total classes &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;11<br> Total events &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;9026<br> Ratio event/duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;0.04<br> Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;24000 Hz<br> -----------------------------------------------------<br> TRAINING SET/HT<br> -----------------------------------------------------<br> Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;5<br> Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;5 hours<br> Total classes &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;5<br> Total events &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;611<br> Ratio event/duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;0.05<br> Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;6000 Hz<br> -----------------------------------------------------<br> TRAINING SET/JD<br> -----------------------------------------------------<br> Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;1<br> Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;10 mins<br> Total classes&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;1<br> Total events&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;357<br> Ratio event/duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;0.06<br> Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;22050 Hz<br> -----------------------------------------------------<br> TRAINING SET/MT<br> -----------------------------------------------------<br> Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2<br> Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;1 hour and 10 mins<br> Total classes&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;4<br> Total events&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;1294<br> Ratio event/duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;0.04<br> Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;8000 Hz<br> -----------------------------------------------------<br> TRAINING SET/WMW<br> -----------------------------------------------------<br> Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;161<br> Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;4 hours and 40 mins<br> Total classes&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;26<br> Total events&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2941<br> Ratio event/duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;0.24<br> Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;various sampling rates<br> -----------------------------------------------------</p> <p>-----------------------------------------------------<br> VALIDATION SET<br> -----------------------------------------------------<br> Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;18<br> Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;5 hours and 57 minutes<br> Total classes&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;5<br> Total events &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;1077<br> -----------------------------------------------------<br> VALIDATION SET/HB<br> -----------------------------------------------------<br> Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;10<br> Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2 hours and 38 minutes<br> Total classes &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;1<br> Total events &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;712<br> Ratio event/duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;0.7<br> Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;44100 Hz<br> -----------------------------------------------------<br> VALIDATION SET/PB<br> -----------------------------------------------------<br> Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;6<br> Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;3 hours<br> Total classes&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2<br> Total events &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;292<br> Ratio event/duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;0.003<br> Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;44100 Hz<br> -----------------------------------------------------<br> VALIDATION SET/ME<br> -----------------------------------------------------<br> Number of audio recordings&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2<br> Total duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;20 minutes<br> Total classes&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;2<br> Total events &nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;73<br> Ratio event/duration&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;0.01<br> Sampling rate&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;&nbsp;&nbsp; &nbsp;|&nbsp;&nbsp; &nbsp;44100 Hz<br> -----------------------------------------------------</p> <p>&nbsp;</p> <p><strong>Annotation structure</strong></p> <p>Each line of the annotation csv represents an event in the audio file. The column descriptions are as follows:</p> <p>TRAINING SET<br> ---------------------<br> Audiofilename, Starttime, Endtime, CLASS_1, CLASS_2, ...CLASS_N</p> <p>VALIDATION SET<br> ---------------------<br> Audiofilename, Starttime, Endtime, Q</p> <p>&nbsp;</p> <p><strong>Classes</strong></p> <p>DCASE2022_task5_training_set_classes.csv and&nbsp;DCASE2022_task5_validation_set_classes.csv provide a table with class code&nbsp;correspondence to class name for all classes in the Development set.</p> <p>DCASE2022_task5_training_set_classes.csv<br> ---------------------<br> dataset, class_code, class_name</p> <p>DCASE2022_task5_validation_set_classes.csv<br> ---------------------<br> dataset, recording, class_code, class_name</p> <p>&nbsp;</p> <p><strong>Evaluation Set</strong></p> <p>The Evaluation set for this task will be released on the 1st of June 2022</p> <p><strong>Open Access:</strong></p> <p>This dataset is available under a Creative Commons Attribution 4.0 International (CC BY 4.0) license.<br> &nbsp;</p> <p><strong>Contact info:</strong></p> <p>Please send any feedback or questions to:</p> <p>Ines Nolasco -&nbsp;&nbsp;i.dealmeidanolasco@qmul.ac.uk</p>

opencc-by-4.0Mar 2022View details →
zenodo40/100

Few-Shot Learning for Issue Report Classification

<p>This dataset is a sample of the NLBSE23 dataset for Issue Report Classification, manually re-labeled.</p>

opencc-by-4.0Feb 2023View details →
zenodo40/100

Delaunay data set learn2learn l2l for meta-learning and few-shot learning

<p>Delaunay data set learn2learn l2l for meta-learning and few-shot learning. We split it into 3 meta-train, meta-val and meta-test sets.&nbsp;</p> <p>&nbsp;</p> <p>For details of original data see:&nbsp;https://github.com/camillegontier/DELAUNAY_dataset</p>

opencc-by-4.0Mar 2023View details →
zenodo40/100

Supporting information for "Few-Shot Learning Enables Population-Scale Analysis of Leaf Traits in Populus trichocarpa"

<p>Raw VCF file for <em>Populus trichocarpa</em> V4 SNPs - Biallelic SNP variants were called for 1492 Populus trichocarpa genotypes using GATK Version 4.0 software.</p>

opencc-by-4.0Jan 2023View details →
zenodo36/100

Explainable few-shot learning workflow for detecting invasive and exotic tree species

<p>This is the supporting dataset of research work: <a href="Link"><strong>Explainable few-shot learning workflow for detecting</strong></a> <a href="Link"><strong>invasive and exotic tree species</strong></a>. (Link to be added after the publication) In this research, we presents a workflow that tackles both challenges by proposing an explainable few-shot learning workflow for detecting invasive and exotic tree species in the Atlantic Forest of Brazil using Unmanned Aerial Vehicle (UAV) images. By integrating a Siamese network with explainable AI (XAI), the workflow enables the classification of tree species with minimal labeled data while providing visual, case-based explanations for the predictions.</p> <p>The workflow is accessible in <a href="Link">this GitHub repository</a> (Link to be added after the publication). The required dataset of this workflow in provided in this Zenodo repository.</p> <p>This dataset repository has the following contents</p> <ul> <li> <p>uav_img.zip: the UAV orthomosaic image (.tif) of the study area used in this research, with related metadata</p> </li> <li>tree_labels.zip: the labels of trees created by expert, available in .shp and .gpkg</li> <li> <p>cutouts.zip: tree cutouts used in this study. They are two sub-directories:</p> <ul> <li>all_cutouts: all the candidated cutouts from three sources. See the README.md file insisde this folder for more information</li> <li>selected cutout: the manually selected cutouts from all cutouts used for training.</li> </ul> </li> <li> <p>training_pairs_20000.zarr.zip: training data created for base network traning. It is created by pairing the selected cutouts.</p> </li> <li>netflora.zip: Netflora workflow prediction results</li> <li>optimized_models.zip: Optimized base models (shallow and deep) and refined models with different shots/fold setup.</li> <li>n_fold_x_validation.zip: data pairs for refinement traing, with n fold and x valiation setup.</li> </ul>

opencc-by-4.0Sep 2024View details →
zenodo36/100

PRICER: Leveraging Few-Shot Learning with Fine-Tuned Large Language Models for Unstructured Economic Data

<p>Describes the taxonomy used in the paper "PRICER: Leveraging Few-Shot Learning with Fine-Tuned Large Language Models for Unstructured Economic Data", presented at the Second Workshop on Semantic Technologies and Deep Learning Models for Scientific, Technical and Legal Data<em>&nbsp;</em>at the Extended Semantic Web Conference (ESWC)&nbsp;2024.</p>

opencc-by-4.0May 2024View details →
zenodo36/100

DCASE 2022 Task 5: Few-shot Bioacoustic Event Detection Evaluation Set

<p><strong>General Description</strong></p> <p>The evaluation set for task 5 of DCASE 2022&nbsp;&quot;Few-shot Bioacoustic Event Detection&quot; consists of 46 audio files acquired from different bioacoustic sources.&nbsp;</p> <p>The first 5 annotations are provided for each file, with events marked as positive (POS) for the class of interest.&nbsp;</p> <p>This dataset is to be used for evaluation purposes during the task</p> <p><strong>Folder Structure</strong></p> <p><em>Evaluation_Set.zip</em></p> <p>&nbsp; &nbsp; |___DC/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; |___CT/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; |___CHE/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; |___MGE/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; |___MS/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; |___QU/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp;</p> <p><em>Evaluation_Set_5shots.zip</em>&nbsp;has the same structure but contains only the *.wav files.</p> <p><em>Evaluation_Set_5shots_annotations_only.zip</em>&nbsp;has the same structure but contains only the *.csv files</p> <p>The subfolders denote different recording sources and there may or may not be overlap between classes of interest from different wav files.</p> <p><strong>Annotation structure</strong></p> <p>Each line of the annotation csv represents an event in the audio file. The column descriptions are as follows:<br> [ Audiofilename, Starttime, Endtime, Q ]</p> <p><strong>Development Set</strong></p> <p>The development set for the same task can be found at:&nbsp;<a href="http://doi.org/10.5281/zenodo.4543504">https://doi.org/10.5281/zenodo.6012309</a></p> <p><strong>Open Access</strong></p> <p>This dataset is available under a Creative Commons Attribution 4.0 International (CC BY 4.0) license.<br> &nbsp;</p> <p><strong>Contact info</strong></p> <p>Please send any feedback or questions to:<br> Ines Nolasco: i.dealmeidanolasco@qmul.ac.uk</p>

opencc-by-4.0May 2022View details →
zenodo36/100

OpenEarthMap Land Cover Mapping Few-Shot Learning Challenge

<pre><strong>***The challenge is over, please use this verion for post-challenge research. This version contains all the data and files, <br>except the labels of the query-set of the testset that are withheld and researchers can submit the predictions at the <br>challenge leaderboard for evaluation***</strong></pre> <p><strong>Overview</strong></p> <p>This challenge is co-organized with the <a href="https://sites.google.com/view/l3divu2024/overview" target="_blank" rel="noopener">L3D-IVU 2024 CVPR</a> workshop. The challenge is an extension of the&nbsp;<a href="https://open-earth-map.org/" target="_blank" rel="noopener">OpenEarthMap </a>benchmark dataset for a generalized few-shot semantic segmentation (GFSS) task. The challenge aims to evaluate and benchmark learning methods for few-shot semantic segmentation on the OpenEarthMap dataset to promote research on geoinformatics for social good. The motivation is to enable researchers to develop few-shot learning algorithms for high-resolution RS image semantic segmentation.</p> <p><strong>Page</strong></p> <p><a href="https://cliffbb.github.io/OEM-Fewshot-Challenge/" target="_blank" rel="noopener">https://cliffbb.github.io/OEM-Fewshot-Challenge/</a></p> <p><strong>Baseline</strong></p> <p>The baseline model for the challenge is available&nbsp;<a href="https://github.com/cliffbb/OEM-Fewshot-Challenge" target="_blank" rel="noopener">here</a>.</p> <p><strong>Leaderboard</strong></p> <p>The <a href="https://codalab.lisn.upsaclay.fr/competitions/19210" target="_blank" rel="noopener">challenge leaderboard </a>is opened for post-challenge research to enable researchers to evaluate the predictions on the query-set of the testset.</p> <p><strong>Description</strong></p> <p>The dataset has been designed for remote sensing few-shot learning, particularly, for GFSS tasks in land cover mapping. The dataset consists of only 408 samples from the original <a href="https://open-earth-map.org/" target="_blank" rel="noopener">OpenEarthMap</a> dataset for RS image semantic segmentation. It extends the original 8 semantic classes of the OpenEarthmap benchmark to 15 classes, which is split into 7:4:4 for <em>train_base_class</em>,&nbsp;<em>val_novel_class</em>, and&nbsp;<em>test_novel_class</em> disjointed sets, respectively (i.e., <em>train_base_class</em>&nbsp;&cap;&nbsp;<em>val_novel_class</em>&nbsp;&cap;&nbsp;<em>test_novel_class</em> = &empty;). The 408 samples are also split into 258 as `trainset`, 50 as `valset`, and 100 as `testset`. The `trainset` is for pre-training a backbone network. It contains only the images and labels of the&nbsp;<em>train_base_class</em> split. Both the `valset` and the `testset` consist of a <em>support set</em> and a <em>query set</em> for a <strong>5-shot</strong>&nbsp; with <strong>4 novel classes</strong>&nbsp; and <strong>7 base classes</strong>&nbsp; GFSS task. The `valset` and the `testset` contain the images and labels of the <em>val_novel_class</em>&nbsp;and the&nbsp;<em>test_novel_class</em> splits, respectively.</p> <p>The challenge is in two phases: development phase and evaluation phase. The `valset` is for the development phase and the `testsets` is for the evaluation phase. Both `valset` and `testset`&nbsp; have 20 image-label pair examples, 5-set examples for each of the 4 novel classes in the <em>support set</em>. The `valset` and the `testset` contain an additional 30 images and 80 images, respectively, in the <em>query set</em>, which are to be predicted using the 20 labelled images in their&nbsp;<em>support set</em>.&nbsp;The labels for each image in the <em>support sets</em> do not contain any of the <em>train_base_class</em> split. Also, in each 5-set examples, the labels contain only one novel class (i.e., one novel class per 5-set examples). However, in the `valset`, the labels for the images in the&nbsp;<em>query set</em> contain both <em>train_base_class</em>&nbsp;and <em>val_novel_class</em>; and in the `testset, the labels for the images in the <em>query set</em> contain both&nbsp; <em>train_base_class</em> and <em>test_novel_class</em> `. Note that both the&nbsp;<em>support set</em>&nbsp;and&nbsp;<em>query set</em> in the `valset` are different from the ones&nbsp;in the `testset`.</p> <pre><br>File Structure and Content (All files are in `.tif` format): ----------------------------------------------------------- 1. **trainset.zip**: - Contains `images` and `labels` folders - `images` folder: 258 images of size 1024x1024 with a GSD (Ground Sampling Distance) of 0.6-1m.<br> - `labels` folder: 258 segmentation masks of the images in the `images` folder. 2. **valset.zip**:<br> - Contains `images` and `labels` folders<br> - `images` folder: 50 images of size 1024x1024 with a GSD (Ground Sampling Distance) of 0.6-1m.<br> - `labels` folder: 20 labels of the ``support set`` images in the `images` folder. The labels for<br> the 30 ``query set`` images in the `images` folder are withheld.<br>3. **testset.zip**:<br> - Contains `images` and `labels` folders<br> - `images` folder: 100 images of size 1024x1024 with a GSD (Ground Sampling Distance) of 0.6-1m.<br> - `labels` folder: 20 labels of the ``support set`` images in the `images` folder. The labels for<br> the 80 ``query set`` images in the `images` folder are withheld.<br> 4. **train.txt**:<br> - Contains a list of file names in the `trainset.zip`.<br> <br>3. **val.json** and **test.json**:<br> - Contains a list of file names the in the `valset.zip` and `testset.zip`, respectively. Below is<br> the structure of the `val.json` and `test.json` files.<br> - fnames = {<br> {"support_set": {8: ["filename_1.tif", "filename_2.tif", ...., "filename_5.tif"],<br> 9: ["filename_1.tif", "filename_2.tif", ...., "filename_5.tif"],<br> 10: ["filename_1.tif", "filename_2.tif", ...., "filename_5.tif"],<br> 11: ["filename_1.tif", "filename_2.tif", ...., "filename_5.tif"]},<br> {"query_set": ["filename_1.tif", "filename_2.tif", "filename_3.tif", ...<br> ...., <br> "filename_n.tif"]} <br> }<br> Land Cover Mapping Classes Strucure: ------------------------------------<br>1. **The `trainset`:<br> classId2className = {<br> &nbsp; &nbsp; # ***Base classes*** &nbsp; &nbsp; 1: 'tree', &nbsp; &nbsp; 2: 'rangeland', &nbsp; &nbsp; 3: 'bareland', &nbsp; &nbsp; 4: 'agric land type 1', &nbsp; &nbsp; 5: 'road type 1', &nbsp; &nbsp; 6: 'sea, lake, &amp; pond', &nbsp; &nbsp; 7: 'building type 1'<br> }<br><br>2. **The `valset` and `testset`:<br> classId2className = {<br> &nbsp; &nbsp; # ***Base classes*** &nbsp; &nbsp; 1: 'tree', &nbsp; &nbsp; 2: 'rangeland', &nbsp; &nbsp; 3: 'bareland', &nbsp; &nbsp; 4: 'agric land type 1', &nbsp; &nbsp; 5: 'road type 1', &nbsp; &nbsp; 6: 'sea, lake, &amp; pond', &nbsp; &nbsp; 7: 'building type 1'<br> # ***Novel classes***<br> &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 8: '',<br> &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 9: '',<br> &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 10: '',<br> &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 11: ''<br> }<br><br> - The class names for the ***Novel classes*** depends on the data set.<br><br> For the `valset`, the class names can be updated as:<br> {<br> 8: 'road type 2',<br> &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 9: 'river',<br> &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 10: 'boat &amp; ship',<br> &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 11: 'agric land type 2'<br> }<br><br> For the `testset`, the class names can be updated as:<br> {<br> 8: 'vehicle &amp; cargo-trailer',<br> &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 9: 'parking space',<br> &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 10: 'sports field',<br> &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 11: 'building type 2'<br> }<br><br><strong>License</strong></pre> <p>See <a href="../records/7223446" target="_blank" rel="noopener">OpenEarthMap</a></p>

opencc-by-4.0Jan 2024View details →
zenodo36/100

DCASE 2023 Task5: Few-shot Bioacoustic Event Detection: Evaluation set

<p><strong>General Description</strong></p> <p>The evaluation set for task 5 of DCASE 2023&nbsp;&quot;Few-shot Bioacoustic Event Detection&quot; consists of 8 subsets of data representing different acoustic sources, in total there are 66 audio files.&nbsp;</p> <p>The first 5 annotations are provided for each file, with events marked as positive (POS) for the class of interest.&nbsp;</p> <p>This dataset is to be used for evaluation purposes during the task.</p> <p>&nbsp;</p> <p><strong>Folder structure</strong></p> <p>Audiofiles and annotation files are organized across 3 different zip files:</p> <p><em>Eval_1.zip</em></p> <p>&nbsp; &nbsp; |___CHE23/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; |___CW/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p><em>Eval_2.zip</em></p> <p>&nbsp; &nbsp; |___MGE/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; |___MS/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; |___QU/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp;<em>Eval_3.zip</em></p> <p>&nbsp; &nbsp; |___DC/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; |___CT/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p>&nbsp; &nbsp; |___CHE/</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.wav</p> <p>&nbsp; &nbsp; &nbsp; &nbsp; |____*.csv</p> <p><em>Annotations_only.zip</em>&nbsp;contains the *.csv files organised in the same subfolders.</p> <p>*The subfolders denote different recording sources and there may or may not be overlap between classes of interest from different wav files.</p> <p>*Note that there can be different target classes within the same subfolder</p> <p>&nbsp;</p> <p><strong>Annotation structure</strong></p> <p>Each line of the annotation csv represents an event in the audio file. The column descriptions are as follows:<br> [ Audiofilename, Starttime, Endtime, Q ]</p> <p><strong>Development Set</strong></p> <p>The development set for the same task can be found at:&nbsp;<a href="http://doi.org/10.5281/zenodo.4543504">https://doi.org/10.5281/zenodo.6012309</a>.&nbsp;</p> <p><strong>Open Access</strong></p> <p>This dataset is available under a Creative Commons Attribution 4.0 International (CC BY 4.0) license.<br> &nbsp;</p> <p><strong>Contact info</strong></p> <p>Please send any feedback or questions to:<br> Ines Nolasco: i.dealmeidanolasco@qmul.ac.uk</p> <p>or join us on slack:&nbsp;<a href="https://join.slack.com/t/dcase/shared_invite/zt-12zfa5kw0-dD41gVaPU3EZTCAw1mHTCA">task-fewshot-bio-sed</a></p>

opencc-by-4.0Apr 2023View details →
zenodo32/100

The few-shot training dataset for CLOSURE

<p>This upload contains the questions that were used for few-shot experiments in the CLOSURE paper.</p> <p>The rest of the dataset is available here:</p> <p><a href="https://zenodo.org/record/3634090#.XyAaAHVKgUH">https://zenodo.org/record/3634090#.XyAaAHVKgUH</a></p>

opencc-by-4.0Jul 2020View details →
zenodo32/100

DIPROMATS 2024 - Shared Task 2: few-shot training data for narrative identification

<p>Narratives are causally connected sequences of events that are selected and evaluated as meaningful for a particular audience. They make sense of the world by identifying the significance of people, places, objects, and events in time. In international relations, international actors create strategic narratives to &ldquo;construct a shared meaning of the past, present, and future of international politics to shape the behavior of domestic and international actors&rdquo;</p> <p>DIPROMATS 2024 Task 2 is a multiclass multilabel classification problem. Given a series of predefined narratives of each international actor, systems must determine which narrative the tweets belong to. Systems will receive the description of each narrative and a few examples of tweets in both languages (English and Spanish) that belong to each of them (few-shot learning). A tweet may be associated with one, several or none of the narratives.</p> <p>These are the few-shot training datasets for Englsih and Spanish.</p> <p>These files don't contain the narratives description. You can find them in the testing dataset:</p> <p>Pe&ntilde;as, A., Fraile-Hern&aacute;ndez, J. M., Moral, P., Rodrigo, &Aacute;., Deriu, J., Sharma, R., Centeno, R., Rodr&iacute;guez-Garc&iacute;a, R., Giedemann, P., &amp; Reyes-Montesinos, J. (2024). DIPROMATS 2024 - Shared Task 2: testing data for narrative identification (1.0.0) [Data set]. Zenodo. <a href="https://doi.org/10.5281/zenodo.12663310" target="_blank" rel="noopener">https://doi.org/10.5281/zenodo.12663310</a></p>

opencc-by-4.0Mar 2024View details →
zenodo32/100

Meta-learning an Intermediate Representation for Few-shot Block-wise Prediction of Landslide Susceptibility

<p>This upload contains 1) the used thematic maps in the study,&nbsp;2) the samples to train and validate the proposed model, and 3) the samples used to predict landslide susceptibility of Fengjie County and Fuling District. The code related to the methods is available on the website:&nbsp;<a href="https://github.com/Young-Excavator/Meta_LSM">https://github.com/Young-Excavator/Meta_LSM</a>.</p>

opencc-by-4.0Jan 2022View details →
zenodo32/100

Test_S: a synthesised dataset for evaluating few-shot Bioacoustic events detection system (DCASE 2021 Task5)

<ul> <li>The database consists of : <ul> <li>DC class&nbsp; and MEL class&nbsp;</li> <li>dense bioacoustic event like official Evaluation Set in each audio</li> </ul> </li> </ul>

opencc-by-4.0Jan 2022View details →

ScienceDex guides

Understand access before you commit

These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.

Compare curated datasets

Allen Brain Atlas

Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.

allen-brain-atlas
neuroscienceopenDocumentation, web resources, and API references are available online.
Last verified 2026-04-30Open record

Annotated Behaviour and Observability Dataset (ABODe)

ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.

abode-home-cage
behavioral-neuroscienceopenThe DataShare record exposes download links for annotations, documentation, license text, and the zipped per-snippet data directory.
Last verified 2026-04-30Open record

DANDI Archive for NWB datasets

DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.

dandi-nwb
electrophysiologyopenPublished Dandiset metadata and archive endpoints are available through the production DANDI API.
Last verified 2026-04-30Open record

International Brain Laboratory public data

The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.

ibl
behavioral-neuroscienceopenPublic sessions can be searched and loaded from the IBL public data server through ONE.
Last verified 2026-04-29Open record

OpenNeuro

OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.

openneuro
neuroscienceopenPublished datasets are available on demand over the internet.
Last verified 2026-04-29Open record