Advances in neuroimaging, genomic, motion tracking, eye-tracking and many other technology-based data collection methods have led to a torrent of high dimensional datasets, which commonly have a small number of samples because of the intrinsic high cost of data collection involving human participants. High dimensional data with a small number of samples is of critical importance for identifying biomarkers and conducting feasibility and pilot work, however it can lead to biased machine learning (ML) performance estimates. Our review of studies which have applied ML to predict autistic from non-autistic individuals showed that small sample size is associated with higher reported classification accuracy. Thus, we have investigated whether this bias could be caused by the use of validation methods which do not sufficiently control overfitting. Our simulations show that K-fold Cross-Validation (CV) produces strongly biased performance estimates with small sample sizes, and the bias is still evident with sample size of 1000. Nested CV and train/test split approaches produce robust and unbiased performance estimates regardless of sample size. We also show that feature selection if performed on pooled training and testing data is contributing to bias considerably more than parameter tuning. In addition, the contribution to bias by data dimensionality, hyper-parameter space and number of CV folds was explored, and validation methods were compared with discriminable data. The results suggest how to design robust testing methodologies when working with small datasets and how to interpret the results of other studies based on what validation method was used.

Overview

The data has been split into two groups:

• training set (train.csv)
• test set (test.csv)

The training set should be used to build your machine learning models. For the training set, we provide the outcome (also known as the “ground truth”) for each passenger. Your model will be based on “features” like passengers’ gender and class. You can also use feature engineering to create new features.

The test set should be used to see how well your model performs on unseen data. For the test set, we do not provide the ground truth for each passenger. It is your job to predict these outcomes. For each passenger in the test set, use the model you trained to predict whether or not they survived the sinking of the Titanic.

We also include gender_submission.csv, a set of predictions that assume all and only female passengers survive, as an example of what a submission file should look like.

Data Dictionary:

| Variable | Definition | Key | | --- | --- | | survival | Survival | 0 = No, 1 = Yes | | pclass | Ticket class | 1 = 1st, 2 = 2nd, 3 = 3rd | | sex | Sex | | | Age | Age in years | | | sibsp | # of siblings / spouses aboard the Titanic | | | parch | # of parents / children aboard the Titanic | | | ticket | Ticket number | | | fare | Passenger fare | | | cabin | Cabin number | | | embarked | Port of Embarkation | C = Cherbourg, Q = Queenstown, S = Southampton |

Variable Notes

pclass: A proxy for socio-economic status (SES) 1st = Upper 2nd = Middle 3rd = Lower

age: Age is fractional if less than 1. If the age is estimated, is it in the form of xx.5

sibsp: The dataset defines family relations in this way... Sibling = brother, sister, stepbrother, stepsister Spouse = husband, wife (mistresses and fiancés were ignored)

parch: The dataset defines family relations in this way... Parent = mother, father Child = daughter, son, stepdaughter, stepson Some children travelled only with a nanny, therefore parch=0 for them.

More - Find More Exciting🙀 Datasets Here - An Upvote👍 A Dayᕙ(`▿´)ᕗ , Keeps Aman Hurray Hurray..... ٩(˘◡˘)۶Hehe

Sure! I'd be happy to provide you with an introduction to machine learning basics for beginners. Machine learning is a subfield of artificial intelligence (AI) that focuses on enabling computers to learn and make predictions or decisions without being explicitly programmed. Here are some key concepts and terms to help you get started:

1. Supervised Learning: In supervised learning, the machine learning algorithm learns from labeled training data. The training data consists of input examples and their corresponding correct output or target values. The algorithm learns to generalize from this data and make predictions or classify new, unseen examples.

2. Unsupervised Learning: Unsupervised learning involves learning patterns and relationships from unlabeled data. Unlike supervised learning, there are no target values provided. Instead, the algorithm aims to discover inherent structures or clusters in the data.

3. Training Data and Test Data: Machine learning models require a dataset to learn from. The dataset is typically split into two parts: the training data and the test data. The model learns from the training data, and the test data is used to evaluate its performance and generalization ability.

4. Features and Labels: In supervised learning, the input examples are often represented by features or attributes. For example, in a spam email classification task, features might include the presence of certain keywords or the length of the email. The corresponding output or target values are called labels, indicating the class or category to which the example belongs (e.g., spam or not spam).

5. Model Evaluation Metrics: To assess the performance of a machine learning model, various evaluation metrics are used. Common metrics include accuracy (the proportion of correctly predicted examples), precision (the proportion of true positives among all positive predictions), recall (the proportion of true positives predicted correctly), and F1 score (a combination of precision and recall).

6. Overfitting and Underfitting: Overfitting occurs when a model becomes too complex and learns to memorize the training data instead of generalizing well to unseen examples. On the other hand, underfitting happens when a model is too simple and fails to capture the underlying patterns in the data. Balancing the complexity of the model is crucial to achieve good generalization.

7. Feature Engineering: Feature engineering involves selecting or creating relevant features that can help improve the performance of a machine learning model. It often requires domain knowledge and creativity to transform raw data into a suitable representation that captures the important information.

8. Bias and Variance Trade-off: The bias-variance trade-off is a fundamental concept in machine learning. Bias refers to the errors introduced by the model's assumptions and simplifications, while variance refers to the model's sensitivity to small fluctuations in the training data. Reducing bias may increase variance and vice versa. Finding the right balance is important for building a well-performing model.

9. Supervised Learning Algorithms: There are various supervised learning algorithms, including linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), and neural networks. Each algorithm has its own strengths, weaknesses, and specific use cases.

10. Unsupervised Learning Algorithms: Unsupervised learning algorithms include clustering algorithms like k-means clustering and hierarchical clustering, dimensionality reduction techniques like principal component analysis (PCA) and t-SNE, and anomaly detection algorithms, among others.

These concepts provide a starting point for understanding the basics of machine learning. As you delve deeper, you can explore more advanced topics such as deep learning, reinforcement learning, and natural language processing. Remember to practice hands-on with real-world datasets to gain practical experience and further refine your skills.

Performances of models to predict unseen analysis windows [cross validation (CV) approach] of the Swedish, Swiss and Swedish+Swiss testing data sets.

Context The Caltech-256 dataset is a foundational benchmark for object recognition, containing 30,607 images across 257 categories (256 object categories + 1 clutter category).

The original dataset is typically provided as a collection of directories, one for each category. This version streamlines the machine learning workflow by providing:

A clean, pre-defined 80/20 train-test split.

Manifest files (train.csv, test.csv) that map image paths directly to their labels, allowing for easy use with data generators in frameworks like PyTorch and TensorFlow.

A flat directory structure (train/, test/) for simplified file access.

File Content The dataset is organized into a single top-level folder and two CSV files:

train.csv: A CSV file containing two columns: image_path and label. This file lists all images designated for the training set.

test.csv: A CSV file with the same structure as train.csv, listing all images designated for the testing set.

Caltech-256_Train_Test/: The primary data folder.

train/: This directory contains 80% of the images from all 257 categories, intended for model training.

test/: This directory contains the remaining 20% of the images from all categories, reserved for model evaluation.

Data Split The dataset has been thoroughly partitioned to create a standard 80% training and 20% testing split. This split is (or should be assumed to be) stratified, meaning that each of the 257 object categories is represented in roughly an 80/20 proportion in the respective sets.

Acknowledgements & Original Source This dataset is a derivative work created for convenience. The original data and images belong to the authors of the Caltech-256 dataset.

Original Dataset Link: https://www.kaggle.com/datasets/jessicali9530/caltech256/data

Citation: Griffin, G. Holub, A.D. Perona, P. (2007). Caltech-256 Object Category Dataset. California Institute of Technology.

Bats play crucial ecological roles and provide valuable ecosystem services, yet many populations face serious threats from various ecological disturbances. The North American Bat Monitoring Program (NABat) aims to assess status and trends of bat populations while developing innovative and community-driven conservation solutions using its unique data and technology infrastructure. To support scalability and transparency in the NABat acoustic data pipeline, we developed a fully-automated machine-learning algorithm. This dataset includes audio files of bat echolocation calls that were considered to develop V1.0 of the NABat machine-learning algorithm, however the test set (i.e., holdout dataset) has been excluded from this release. These recordings were collected by various bat monitoring partners across North America using ultrasonic acoustic recorders for stationary acoustic and mobile acoustic surveys. For more information on how these surveys may be conducted, see Chapters 4 and 5 of “A Plan for the North American Bat Monitoring Program” (https://doi.org/10.2737/SRS-GTR-208). These data were then post-processed by bat monitoring partners to remove noise files (or those that do not contain recognizable bat calls) and apply a species label to each file. There is undoubtedly variation in the steps that monitoring partners take to apply a species label, but the steps documented in “A Guide to Processing Bat Acoustic Data for the North American Bat Monitoring Program” (https://doi.org/10.3133/ofr20181068) include first processing with an automated classifier and then manually reviewing to confirm or downgrade the suggested species label. Once a manual ID label was applied, audio files of bat acoustic recordings were submitted to the NABat database in Waveform Audio File format. From these available files in the NABat database, we considered files from 35 classes (34 species and a noise class). Files for 4 species were excluded due to low sample size (Corynorhinus rafinesquii, N=3; Eumops floridanus, N =3; Lasiurus xanthinus, N = 4; Nyctinomops femorosaccus, N =11). From this pool, files were randomly selected until files for each species/grid cell combination were exhausted or the number of recordings reach 1250. The dataset was then randomly split into training, validation, and test sets (i.e., holdout dataset). This data release includes all files considered for training and validation, including files that had been excluded from model development and testing due to low sample size for a given species or because the threshold for species/grid cell combinations had been met. The test set (i.e., holdout dataset) is not included. Audio files are grouped by species, as indicated by the four-letter species code in the name of each folder. Definitions for each four-letter code, including Family, Genus, Species, and Common name, are also included as a dataset in this release.

Many e-shops have started to mark-up product data within their HTML pages using the schema.org vocabulary. The Web Data Commons project regularly extracts such data from the Common Crawl, a large public web crawl. The Web Data Commons Training and Test Sets for Large-Scale Product Matching contain product offers from different e-shops in the form of binary product pairs (with corresponding label “match” or “no match”) for four product categories, computers, cameras, watches and shoes. In order to support the evaluation of machine learning-based matching methods, the data is split into training, validation and test sets. For each product category, we provide training sets in four different sizes (2.000-70.000 pairs). Furthermore there are sets of ids for each training set for a possible validation split (stratified random draw) available. The test set for each product category consists of 1.100 product pairs. The labels of the test sets were manually checked while those of the training sets were derived using shared product identifiers from the Web weak supervision. The data stems from the WDC Product Data Corpus for Large-Scale Product Matching - Version 2.0 which consists of 26 million product offers originating from 79 thousand websites. For more information and download links for the corpus itself, please follow the links below.

Estimated differences in TPR between the model indicated in the line header and the model indicated in the column header. Models in bold are considered as the best predictive models regarding their TPR.

Here we use Python to visualize how certain machine learning algorithms classify certain data points in the Iris dataset. Let's begin by importing the Iris dataset and splitting it into features and labels. We will use only the petal length and width for this analysis.

Models performances on training data sets for the Swiss, Swedish and Swedish+Swiss data sets.

Curated image privacy data sets

In the article, we trained and evaluated models on the Image Privacy Dataset (IPD) and the PrivacyAlert dataset. The datasets are originally provided by other sources and have been re-organised and curated for this work.

Our curation organises the datasets in a common structure. We updated the annotations and labelled the splits of the data in the annotation file. This avoids having separated folders of images for each data split (training, validation, testing) and allows a flexible handling of new splits, e.g. created with a stratified K-Fold cross-validation procedure. As for the original datasets (PicAlert and PrivacyAlert), we provide the link to the images in bash scripts to download the images. Another bash script re-organises the images in sub-folders with maximum 1000 images in each folder.

Both datasets refer to images publicly available on Flickr. These images have a large variety of content, including sensitive content, seminude people, vehicle plates, documents, private events. Images were annotated with a binary label denoting if the content was deemed to be public or private. As the images are publicly available, their label is mostly public. These datasets have therefore a high imbalance towards the public class. Note that IPD combines two other existing datasets, PicAlert and part of VISPR, to increase the number of private images already limited in PicAlert. Further details in our corresponding https://doi.org/10.48550/arXiv.2503.12464" target="_blank" rel="noopener">publication.

List of datasets and their original source:

• PicAlert [Images occupy 2.4 GB]
• VISPR [Images occupy 49.7 GB]
• PrivacyAlert [Images occupy 1 GB]

Notes:

• For PicAlert and PrivacyAlert, only urls to the original locations in Flickr are available in the Zenodo record
• Collector and authors of the PrivacyAlert dataset selected the images from Flickr under Public Domain license
• Owners of the photos on Flick could have removed the photos from the social media platform
• Running the bash scripts to download the images can incur in the "429 Too Many Requests" status code

Pre-computed visual entitities

Some of the models run their pipeline end-to-end with the images as input, whereas other models require different or additional inputs. These inputs include the pre-computed visual entities (scene types and object types) represented in a graph format, e.g. for a Graph Neural Network. Re-using these pre-computed visual entities allows other researcher to build new models based on these features while avoiding re-computing the same on their own or for each epoch during the training of a model (faster training).

For each image of each dataset, namely PrivacyAlert, PicAlert, and VISPR, we provide the predicted scene probabilities as a .csv file , the detected objects as a .json file in COCO data format, and the node features (visual entities already organised in graph format with their features) as a .json file. For consistency, all the files are already organised in batches following the structure of the images in the datasets folder. For each dataset, we also provide the pre-computed adjacency matrix for the graph data.

Note: IPD is based on PicAlert and VISPR and therefore IPD refers to the scene probabilities and object detections of the other two datasets. Both PicAlert and VISPR must be downloaded and prepared to use IPD for training and testing.

Further details on downloading and organising data can be found in our GitHub repository: https://github.com/graphnex/privacy-from-visual-entities (see ARTIFACT-EVALUATION.md#pre-computed-visual-entitities-)

Enquiries, questions and comments

If you have any enquiries, question, or comments, or you would like to file a bug report or a feature request, use the issue tracker of our GitHub repository.

Dataset Description

Context and Methodology

• Research Domain/Project:
  This dataset was created for a machine learning experiment aimed at developing a classification model to predict outcomes based on a set of features. The primary research domain is disease prediction in patients. The dataset was used in the context of training, validating, and testing.

• Purpose of the Dataset:
  The purpose of this dataset is to provide training, validation, and testing data for the development of machine learning models. It includes labeled examples that help train classifiers to recognize patterns in the data and make predictions.

• Dataset Creation:
  Data preprocessing steps involved cleaning, normalization, and splitting the data into training, validation, and test sets. The data was carefully curated to ensure its quality and relevance to the problem at hand. For any missing values or outliers, appropriate handling techniques were applied (e.g., imputation, removal, etc.).

Technical Details

• Structure of the Dataset:
  The dataset consists of several files organized into folders by data type:

  • Training Data: Contains the training dataset used to train the machine learning model.

  • Validation Data: Used for hyperparameter tuning and model selection.

  • Test Data: Reserved for final model evaluation.

  Each folder contains files with consistent naming conventions for easy navigation, such as train_data.csv, validation_data.csv, and test_data.csv. Each file follows a tabular format with columns representing features and rows representing individual data points.

• Software Requirements:
  To open and work with this dataset, you need VS Code or Jupyter, which could include tools like:

  • Python (with libraries such as pandas, numpy, scikit-learn, matplotlib, etc.)

Further Details

• Reusability:
  Users of this dataset should be aware that it is designed for machine learning experiments involving classification tasks. The dataset is already split into training, validation, and test subsets. Any model trained with this dataset should be evaluated using the test set to ensure proper validation.

• Limitations:
  The dataset may not cover all edge cases, and it might have biases depending on the selection of data sources. It's important to consider these limitations when generalizing model results to real-world applications.

WIKI Paragraphs English

A multi-split dataset for machine learning research and evaluation, containing text samples in JSON Lines format.

  Features

Multiple splits for different use cases Random shuffle with Fisher-Yates algorithm Structured format with text and metadata Size-varied validation/test sets (100 to 10k samples)

  Splits Overview

Split Name Samples Typical Usage

train 1,000,000 Primary training data

validation 10,000 Standard validation… See the full description on the dataset page: https://huggingface.co/datasets/pere/wiki_paragraphs_english.

This is training and testing data for the detection of hunting pits in airborne laser data. The data is split into three parts. 1: Data for transfer learning with radar imagery and impact craters on the moon. 2. Data for training and testing of the machine learning model. 3: Data from a separate demonstration area used to evaluate the model.

The lunar data (1) were used to pre-train a machine learning model before training on the real data of hunting pits from earth (2). The demonstration data was used to visually evaluate the result of the final model. All code used to create this dataset and train the machine learning models can be found here: https://github.com/williamlidberg/Detection-of-hunting-pits-using-airborne-laser-scanning-and-deep-learning The code is also included in the file "code.zip"

The MLCommons Dollar Street Dataset is a collection of images of everyday household items from homes around the world that visually captures socioeconomic diversity of traditionally underrepresented populations. It consists of public domain data, licensed for academic, commercial and non-commercial usage, under CC-BY and CC-BY-SA 4.0. The dataset was developed because similar datasets lack socioeconomic metadata and are not representative of global diversity.

This is a subset of the original dataset that can be used for multiclass classification with 10 categories. It is designed to be used in teaching, similar to the widely used, but unlicensed CIFAR-10 dataset.

These are the preprocessing steps that were performed:

1. Only take examples with one imagenet_synonym label
2. Use only examples with the 10 most frequently occuring labels
3. Downscale images to 64 x 64 pixels
4. Split data in train and test
5. Store as numpy array

This is the label mapping:

Checkout https://github.com/carpentries-lab/deep-learning-intro/blob/main/instructors/prepare-dollar-street-data.ipynb" target="_blank" rel="noopener">this notebook to see how the subset was created.

The original dataset was downloaded from https://www.kaggle.com/datasets/mlcommons/the-dollar-street-dataset. See https://mlcommons.org/datasets/dollar-street/ for more information.

The methodology is the core component of any research-related work. The methods used to gain the results are shown in the methodology. Here, the whole research implementation is done using python. There are different steps involved to get the entire research work done which is as follows:

1. Acquire Personality Dataset

The kaggle machine learning dataset is a collection of datasets, data generators which are used by machine learning community for analysis purpose. The personality prediction dataset is acquired from the kaggle website. This dataset was collected (2016-2018) through an interactive on-line personality test. The personality test was constructed from the IPIP. The personality prediction dataset can be downloaded in zip file format just by clicking on the link available. The personality prediction file consists of two subject CSV files (test.csv & train.csv). The test.csv file has 0 missing values, 7 attributes, and final label output. Also, the dataset has multivariate characteristics. Here, data-preprocessing is done for checking inconsistent behaviors or trends.

2. Data preprocessing

After, Data acquisition the next step is to clean and preprocess the data. The Dataset available has numerical type features. The target value is a five-level personality consisting of serious,lively,responsible,dependable & extraverted. The preprocessed dataset is further split into training and testing datasets. This is achieved by passing feature value, target value, test size to the train-test split method of the scikit-learn package. After splitting of data, the training data is sent to the following Logistic regression & SVM design is used for training the artificial neural networks then test data is used to predict the accuracy of the trained network model.

3. Feature Extraction

The following items were presented on one page and each was rated on a five point scale using radio buttons. The order on page was EXT1, AGR1, CSN1, EST1, OPN1, EXT2, etc. The scale was labeled 1=Disagree, 3=Neutral, 5=Agree

        EXT1 I am the life of the party.
        EXT2  I don't talk a lot.
        EXT3  I feel comfortable around people.
        EXT4  I am quiet around strangers.
        EST1  I get stressed out easily.
        EST2  I get irritated easily.
        EST3  I worry about things.
        EST4  I change my mood a lot.
        AGR1  I have a soft heart.
        AGR2  I am interested in people.
        AGR3  I insult people.
        AGR4  I am not really interested in others.
        CSN1  I am always prepared.
        CSN2  I leave my belongings around.
        CSN3  I follow a schedule.
        CSN4  I make a mess of things.
        OPN1  I have a rich vocabulary.
        OPN2  I have difficulty understanding abstract ideas.
        OPN3  I do not have a good imagination.
        OPN4  I use difficult words.

4. Training the Model

Train/Test is a method to measure the accuracy of your model. It is called Train/Test because you split the the data set into two sets: a training set and a testing set. 80% for training, and 20% for testing. You train the model using the training set.In this model we trained our dataset using linear_model.LogisticRegression() & svm.SVC() from sklearn Package

5. Personality Prediction Output

After the training of the designed neural network, the testing of Logistic Regression & SVM is performed using Cohen_kappa_score & Accuracy Score.

Dataset with 72000 pins from 117 users in Pinterest. Each pin contains a short raw text and an image. The images are processed using a pretrained Convolutional Neural Network and transformed into a vector of 4096 features.

This dataset was used in the paper "User Identification in Pinterest Through the Refinement of a Cascade Fusion of Text and Images" to idenfity specific users given their comments. The paper is publishe in the Research in Computing Science Journal, as part of the LKE 2017 conference. The dataset includes the splits used in the paper.

There are nine files. text_test, text_train and text_val, contain the raw text of each pin in the corresponding split of the data. imag_test, imag_train and imag_val contain the image features of each pin in the corresponding split of the data. train_user and val_test_users contain the index of the user of each pin (between 0 and 116). There is a correspondance one-to-one among the test, train and validation files for images, text and users. There are 400 pins per user in the train set, and 100 pins per user in the validation and test sets each one.

If you have questions regarding the data, write to: jc dot gomez at ugto dot mx

The Ice Cream Selling dataset is a simple and well-suited dataset for beginners in machine learning who are looking to practice polynomial regression. It consists of two columns: temperature and the corresponding number of units of ice cream sold.

The dataset captures the relationship between temperature and ice cream sales. It serves as a practical example for understanding and implementing polynomial regression, a powerful technique for modeling nonlinear relationships in data.

The dataset is designed to be straightforward and easy to work with, making it ideal for beginners. The simplicity of the data allows beginners to focus on the fundamental concepts and steps involved in polynomial regression without overwhelming complexity.

By using this dataset, beginners can gain hands-on experience in preprocessing the data, splitting it into training and testing sets, selecting an appropriate degree for the polynomial regression model, training the model, and evaluating its performance. They can also explore techniques to address potential challenges such as overfitting.

With this dataset, beginners can practice making predictions of ice cream sales based on temperature inputs and visualize the polynomial regression curve that represents the relationship between temperature and ice cream sales.

Overall, the Ice Cream Selling dataset provides an accessible and practical learning resource for beginners to grasp the concepts and techniques of polynomial regression in the context of analyzing ice cream sales data.

We used the human genome reference sequence in its GRCh38.p13 version in order to have a reliable source of data in which to carry out our experiments. We chose this version because it is the most recent one available in Ensemble at the moment. However, the DNA sequence by itself is not enough, the specific TSS position of each transcript is needed. In this section, we explain the steps followed to generate the final dataset. These steps are: raw data gathering, positive instances processing, negative instances generation and data splitting by chromosomes.

First, we need an interface in order to download the raw data, which is composed by every transcript sequence in the human genome. We used Ensembl release 104 (Howe et al., 2020) and its utility BioMart (Smedley et al., 2009), which allows us to get large amounts of data easily. It also enables us to select a wide variety of interesting fields, including the transcription start and end sites. After filtering instances that present null values in any relevant field, this combination of the sequence and its flanks will form our raw dataset. Once the sequences are available, we find the TSS position (given by Ensembl) and the 2 following bases to treat it as a codon. After that, 700 bases before this codon and 300 bases after it are concatenated, getting the final sequence of 1003 nucleotides that is going to be used in our models. These specific window values have been used in (Bhandari et al., 2021) and we have kept them as we find it interesting for comparison purposes. One of the most sensitive parts of this dataset is the generation of negative instances. We cannot get this kind of data in a straightforward manner, so we need to generate it synthetically. In order to get examples of negative instances, i.e. sequences that do not represent a transcript start site, we select random DNA positions inside the transcripts that do not correspond to a TSS. Once we have selected the specific position, we get 700 bases ahead and 300 bases after it as we did with the positive instances.

Regarding the positive to negative ratio, in a similar problem, but studying TIS instead of TSS (Zhang135
et al., 2017), a ratio of 10 negative instances to each positive one was found optimal. Following this136
idea, we select 10 random positions from the transcript sequence of each positive codon and label them137
as negative instances. After this process, we end up with 1,122,113 instances: 102,488 positive and 1,019,625 negative sequences. In order to validate and test our models, we need to split this dataset into three parts: train, validation and test. We have decided to make this differentiation by chromosomes, as it is done in (Perez-Rodriguez et al., 2020). Thus, we use chromosome 16 as validation because it is a good example of a chromosome with average characteristics. Then we selected samples from chromosomes 1, 3, 13, 19 and 21 to be part of the test set and used the rest of them to train our models. Every step of this process can be replicated using the scripts available in https://github.com/JoseBarbero/EnsemblTSSPrediction.

Offline reinforcement learning (RL) is a promising direction that allows RL agents to be pre-trained from large datasets avoiding recurrence of expensive data collection. To advance the field, it is crucial to generate large-scale datasets. Compositional RL is particularly appealing for generating such large datasets, since 1) it permits creating many tasks from few components, and 2) the task structure may enable trained agents to solve new tasks by combining relevant learned components. This submission provides four offline RL datasets for simulated robotic manipulation created using the 256 tasks from CompoSuite Mendez et al., 2022. In every task in CompoSuite, a robot arm is used to manipulate an object to achieve an objective all while trying to avoid an obstacle. There are for components for each of these four axes that can be combined arbitrarily leading to a total of 256 tasks. The component choices are * Robot: IIWA, Jaco, Kinova3, Panda* Object: Hollow box, box, dumbbell, plate* Objective: Push, pick and place, put in shelf, put in trashcan* Obstacle: None, wall between robot and object, wall between goal and object, door between goal and object The four included datasets are collected using separate agents each trained to a different degree of performance, and each dataset consists of 256 million transitions. The degrees of performance are expert data, medium data, warmstart data and replay data: * Expert dataset: Transitions from an expert agent that was trained to achieve 90% success on every task.* Medium dataset: Transitions from a medium agent that was trained to achieve 30% success on every task.* Warmstart dataset: Transitions from a Soft-actor critic agent trained for a fixed duration of one million steps.* Medium-replay-subsampled dataset: Transitions that were stored during the training of a medium agent up to 30% success. These datasets are intended for the combined study of compositional generalization and offline reinforcement learning. Methods The datasets were collected by using several deep reinforcement learning agents trained to the various degrees of performance described above on the CompoSuite benchmark (https://github.com/Lifelong-ML/CompoSuite) which builds on top of robosuite (https://github.com/ARISE-Initiative/robosuite) and uses the MuJoCo simulator (https://github.com/deepmind/mujoco). During reinforcement learning training, we stored the data that was collected by each agent in a separate buffer for post-processing. Then, after training, to collect the expert and medium dataset, we run the trained agents for 2000 trajectories of length 500 online in the CompoSuite benchmark and store the trajectories. These add up to a total of 1 million state-transitions tuples per dataset, totalling a full 256 million datapoints per dataset. The warmstart and medium-replay-subsampled dataset contain trajectories from the stored training buffer of the SAC agent trained for a fixed duration and the medium agent respectively. For medium-replay-subsampled data, we uniformly sample trajectories from the training buffer until we reach more than 1 million transitions. Since some of the tasks have termination conditions, some of these trajectories are trunctated and not of length 500. This sometimes results in a number of sampled transitions larger than 1 million. Therefore, after sub-sampling, we artificially truncate the last trajectory and place a timeout at the final position. This can in some rare cases lead to one incorrect trajectory if the datasets are used for finite horizon experimentation. However, this truncation is required to ensure consistent dataset sizes, easy data readability and compatibility with other standard code implementations. The four datasets are split into four tar.gz folders each yielding a total of 12 compressed folders. Every sub-folder contains all the tasks for one of the four robot arms for that dataset. In other words, every tar.gz folder contains a total of 64 tasks using the same robot arm and four tar.gz files form a full dataset. This is done to enable people to only download a part of the dataset in case they do not need all 256 tasks. For every task, the data is separately stored in an hdf5 file allowing for the usage of arbitrary task combinations and mixing of data qualities across the four datasets. Every task is contained in a folder that is named after the CompoSuite elements it uses. In other words, every task is represented as a folder named