Node classification on Film with 60%/20%/20% random splits for training/validation/test.

Dataset, splits, models, and scripts from the manuscript "When Do Quantum Mechanical Descriptors Help Graph Neural Networks Predict Chemical Properties?" are provided. The curated dataset includes 37 QM descriptors for 64,921 unique molecules across six levels of theory: wB97XD, B3LYP, M06-2X, PBE0, TPSS, and BP86. This dataset is stored in the data.tar.gz file, which also contains a file for multitask constraints applied to various atomic and bond properties. The data splits (training, validation, and test splits) for both random and scaffold-based divisions are saved as separate index files in splits.tar.gz. The trained D-MPNN models for predicting QM descriptors are saved in the models.tar.gz file. The scripts.tar.gz file contains ready-to-use scripts for training machine learning models to predict QM descriptors, as well as scripts for predicting QM descriptors using our trained models on unseen molecules and for applying radial basis function (RBF) expansion to QM atom and bond features.

Below are descriptions of the available scripts:

1. atom_bond_descriptors.sh: Trains atom/bond targets.
2. atom_bond_descriptors_predict.sh: Predicts atom/bond targets from pre-trained model.
3. dipole_quadrupole_moments.sh: Trains dipole and quadrupole moments.
4. dipole_quadrupole_moments_predict.sh: Predicts dipole and quadrupole moments from pre-trained model.
5. energy_gaps_IP_EA.sh: Trains energy gaps, ionization potential (IP), and electron affinity (EA).
6. energy_gaps_IP_EA_predict.sh: Predicts energy gaps, IP, and EA from pre-trained model.
7. get_constraints.py: Generates constraints file for testing dataset. This generated file needs to be provided before using our trained models to predict the atom/bond QM descriptors of your testing data.
8. csv2pkl.py: Converts QM atom and bond features to .pkl files using RBF expansion for use with Chemprop software.

Below is the procedure for running the ml-QM-GNN on your own dataset:

1. Use get_constraints.py to generate a constraint file required for predicting atom/bond QM descriptors with the trained ML models.
2. Execute atom_bond_descriptors_predict.sh to predict atom and bond properties. Run dipole_quadrupole_moments_predict.sh and energy_gaps_IP_EA_predict.sh to calculate molecular QM descriptors.
3. Utilize csv2pkl.py to convert the data from predicted atom/bond descriptors .csv file into separate atom and bond feature files (which are saved as .pkl files here).
4. Run Chemprop to train your models using the additional predicted features supported here.

Dataset

This dataset was created by pascalammeter

Released under CC BY-NC-SA 4.0

Contents

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.

Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms

Authors:

• Han Xiao, Kashif Rasul and Roland Vollgraf
• https://arxiv.org/abs/1708.07747

Dataset Obtained From: https://github.com/zalandoresearch/fashion-mnist

All images were sized 28x28 in the original dataset

Fashion-MNIST is a dataset of Zalando's article images—consisting of a training set of 60,000 examples and a test set of 10,000 examples. Each example is a 28x28 grayscale image, associated with a label from 10 classes. We intend Fashion-MNIST to serve as a direct drop-in replacement for the original MNIST dataset for benchmarking machine learning algorithms. It shares the same image size and structure of training and testing splits. * Source

Here's an example of how the data looks (each class takes three-rows): https://github.com/zalandoresearch/fashion-mnist/raw/master/doc/img/fashion-mnist-sprite.png" alt="Visualized Fashion MNIST dataset">

Version 1 (original-images_Original-FashionMNIST-Splits):

• Original images, with the original splits for MNIST: train (86% of images - 60,000 images) set and test (14% of images - 10,000 images) set only.
• This version was not trained

Version 3 (original-images_trainSetSplitBy80_20):

• Original, raw images, with the train set split to provide 80% of its images to the training set and 20% of its images to the validation set
• https://blog.roboflow.com/train-test-split/ https://i.imgur.com/angfheJ.png" alt="Train/Valid/Test Split Rebalancing">

Citation:

@online{xiao2017/online,
 author    = {Han Xiao and Kashif Rasul and Roland Vollgraf},
 title    = {Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms},
 date     = {2017-08-28},
 year     = {2017},
 eprintclass = {cs.LG},
 eprinttype  = {arXiv},
 eprint    = {cs.LG/1708.07747},
}

The BUTTER Empirical Deep Learning Dataset represents an empirical study of the deep learning phenomena on dense fully connected networks, scanning across thirteen datasets, eight network shapes, fourteen depths, twenty-three network sizes (number of trainable parameters), four learning rates, six minibatch sizes, four levels of label noise, and fourteen levels of L1 and L2 regularization each. Multiple repetitions (typically 30, sometimes 10) of each combination of hyperparameters were preformed, and statistics including training and test loss (using a 80% / 20% shuffled train-test split) are recorded at the end of each training epoch. In total, this dataset covers 178 thousand distinct hyperparameter settings ("experiments"), 3.55 million individual training runs (an average of 20 repetitions of each experiments), and a total of 13.3 billion training epochs (three thousand epochs were covered by most runs). Accumulating this dataset consumed 5,448.4 CPU core-years, 17.8 GPU-years, and 111.2 node-years.

The runtime benchmarks were obtained by running each algorithm on the seed and full multi-MSAs Pfam-A.seed and Pfam-A.full on 2 cores with 8 GB RAM for the seed alignments and on 3 cores with 12 GB RAM for the full alignments. We did not compute the maximum runtime of the Blue algorithm; the algorithm failed to terminate within 6 days for 34 families.

Cleaned_Dataset.csv – The combined CSV files of all scraped documents from DABI, e-LiS, o-bib and Springer.

Data_Cleaning.ipynb – The Jupyter Notebook with python code for the analysis and cleaning of the original dataset.

ger_train.csv – The German training set as CSV file.

ger_validation.csv – The German validation set as CSV file.

en_test.csv – The English test set as CSV file.

en_train.csv – The English training set as CSV file.

en_validation.csv – The English validation set as CSV file.

splitting.py – The python code for splitting a dataset into train, test and validation set.

DataSetTrans_de.csv – The final German dataset as a CSV file.

DataSetTrans_en.csv – The final English dataset as a CSV file.

translation.py – The python code for translating the cleaned dataset.

As more hydrocarbon production from hydraulic fracturing and other methods produce large volumes of water, innovative methods must be explored for treatment and reuse of these waters. However, understanding the general water chemistry of these fluids is essential to providing the best treatment options optimized for each producing area. Machine learning algorithms can often be applied to datasets to solve complex problems. In this study, we used the U.S. Geological Survey’s National Produced Waters Geochemical Database (USGS PWGD) in an exploratory exercise to determine if systematic variations exist between produced waters and geologic environment that could be used to accurately classify a water sample to a given geologic province. Two datasets were used, one with fewer attributes (n = 7) but more samples (n = 58,541) named PWGD7, and another with more attributes (n = 9) but fewer samples (n = 33,271) named PWGD9. The attributes of interest were specific gravity, pH, HCO3, Na, Mg, Ca, Cl, SO4, and total dissolved solids. The two datasets, PWGD7 and PWGD9, contained samples from 20 and 19 geologic provinces, respectively. Outliers across all attributes for each province were removed at a 99% confidence interval. Both datasets were divided into a training and test set using an 80/20 split and a 90/10 split, respectively. Random forest, Naïve Bayes, and k-Nearest Neighbors algorithms were applied to the two different training datasets and used to predict on three different testing datasets. Overall model accuracies across the two datasets and three applied models ranged from 23.5% to 73.5%. A random forest algorithm (split rule = extratrees, mtry = 5) performed best on both datasets, producing an accuracy of 67.1% for a training set based on the PWGD7 dataset, and 73.5% for a training set based on the PWGD9 dataset. Overall, the three algorithms predicted more accurately on the PWGD7 dataset than PWGD9 dataset, suggesting that either a larger sample size and/or fewer attributes lead to a more successful predicting algorithm. Individual balanced accuracies for each producing province ranged from 50.6% (Anadarko) to 100% (Raton) for PWGD7, and from 44.5% (Gulf Coast) to 99.8% (Sedgwick) for PWGD9. Results from testing the model on recently published data outside of the USGS PWGD suggests that some provinces may be lacking information about their true geochemical diversity while others included in this dataset are well described. Expanding on this effort could lead to predictive tools that provide ranges of contaminants or other chemicals of concern within each province to design future treatment facilities to reclaim wastewater. We anticipate that this classification model will be improved over time as more diverse data are added to the USGS PWGD.

Overview

The Hard Hat dataset is an object detection dataset of workers in workplace settings that require a hard hat. Annotations also include examples of just "person" and "head," for when an individual may be present without a hard hart.

The original dataset has a 75/25 train-test split.

Example Image: https://i.imgur.com/7spoIJT.png" alt="Example Image">

Use Cases

One could use this dataset to, for example, build a classifier of workers that are abiding safety code within a workplace versus those that may not be. It is also a good general dataset for practice.

Using this Dataset

Use the fork or Download this Dataset button to copy this dataset to your own Roboflow account and export it with new preprocessing settings (perhaps resized for your model's desired format or converted to grayscale), or additional augmentations to make your model generalize better. This particular dataset would be very well suited for Roboflow's new advanced Bounding Box Only Augmentations.

Dataset Versions:

Image Preprocessing | Image Augmentation | Modify Classes * v1 (resize-416x416-reflect): generated with the original 75/25 train-test split | No augmentations * v2 (raw_75-25_trainTestSplit): generated with the original 75/25 train-test split | These are the raw, original images * v3 (v3): generated with the original 75/25 train-test split | Modify Classes used to drop person class | Preprocessing and Augmentation applied * v5 (raw_HeadHelmetClasses): generated with a 70/20/10 train/valid/test split | Modify Classes used to drop person class * v8 (raw_HelmetClassOnly): generated with a 70/20/10 train/valid/test split | Modify Classes used to drop head and person classes * v9 (raw_PersonClassOnly): generated with a 70/20/10 train/valid/test split | Modify Classes used to drop head and helmet classes * v10 (raw_AllClasses): generated with a 70/20/10 train/valid/test split | These are the raw, original images * v11 (augmented3x-AllClasses-FastModel): generated with a 70/20/10 train/valid/test split | Preprocessing and Augmentation applied | 3x image generation | Trained with Roboflow's Fast Model * v12 (augmented3x-HeadHelmetClasses-FastModel): generated with a 70/20/10 train/valid/test split | Preprocessing and Augmentation applied, Modify Classes used to drop person class | 3x image generation | Trained with Roboflow's Fast Model * v13 (augmented3x-HeadHelmetClasses-AccurateModel): generated with a 70/20/10 train/valid/test split | Preprocessing and Augmentation applied, Modify Classes used to drop person class | 3x image generation | Trained with Roboflow's Accurate Model * v14 (raw_HeadClassOnly): generated with a 70/20/10 train/valid/test split | Modify Classes used to drop person class, and remap/relabel helmet class to head

Choosing Between Computer Vision Model Sizes | Roboflow Train

About Roboflow

Roboflow makes managing, preprocessing, augmenting, and versioning datasets for computer vision seamless.

Developers reduce 50% of their code when using Roboflow's workflow, automate annotation quality assurance, save training time, and increase model reproducibility.

Runs from two papers exploring the use of mass conserving LSTM. Model results used in the papers are 1) model_outputs_for_analysis_extreme_events_paper.tar.gz, and 2) model_outputs_for_analysis_mass_balance_paper.tar.gz.

The models here are trained/calibrated on three different time periods. Standard Time Split (time split 1): test period(1989-1999) is the same period used by previous studies which allows us to confirm that the deep learning models (LSTM andMC-LSTM) trained for this project perform as expected relative to prior work. NWM Time Split (time split 2): The second test period (1995-2014) allows us to benchmark against the NWM-Rv2, which does not provide data prior to 1995. Return period split: The third test period (based on return periods) allows us to benchmark only on water years that contain streamflow events that are larger (per basin) than anything seen in the training data (<= 5-year return periods in training and > 5-year return periods in testing).

Also included are an ensemble of model runs for LSTM, MC-LSTM for the "standard" training period and two forcing products. These files are provided in the format "

IMPORTANT NOTE: This python environment should be used to extract and load the data: https://github.com/jmframe/mclstm_2021_extrapolate/blob/main/python_environment.yml, as the pickle files serialized the data with specific versions of python libraries. Specifically, the pickle serialization was done with xarray=0.16.1.

Code to interpret these runs can be found here: https://github.com/jmframe/mclstm_2021_extrapolate https://github.com/jmframe/mclstm_2021_mass_balance

Papers are available here: https://hess.copernicus.org/preprints/hess-2021-423/

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 via 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.

This dataset was used for training and validating the PredictONCO web tool, supporting decision-making in precision oncology by extending the bioinformatics predictions with advanced computing and machine learning. The dataset consists of 1073 single-point mutants of 42 proteins, whose effect was classified as Oncogenic (509 data points) and Benign (564 data points). All mutations were annotated with a clinically verified effect and were compiled from the ClinVar and OncoKB databases. The dataset was manually curated based on the available information in other precision oncology databases (The Clinical Knowledgebase by The Jackson Laboratory, Personalized Cancer Therapy Knowledge Base by MD Anderson Cancer Center, cBioPortal, DoCM database) or in the primary literature. To create the dataset, we also removed any possible overlaps with the data points used in the PredictSNP consensus predictor and its constituents. This was implemented to avoid any test set data leakage due to using the PredictSNP score as one of the features (see below).

The entire dataset (SEQ) was further annotated by the pipeline of PredictONCO. Briefly, the following six features were calculated regardless of the structural information available: essentiality of the mutated residue (yes/no), the conservation of the position (the conservation grade and score), the domain where the mutation is located (cytoplasmic, extracellular, transmembrane, other), the PredictSNP score, and the number of essential residues in the protein. For approximately half of the data (STR: 377 and 76 oncogenic and benign data points, respectively), the structural information was available, and six more features were calculated: FoldX and Rosetta ddg_monomer scores, whether the residue is in the catalytic pocket (identification of residues forming the ligand-binding pocket was obtained from P2Rank), and the pKa changes (the minimum and maximum changes as well as the number of essential residues whose pKa was changed – all values obtained from PROPKA3). For both STR and SEQ datasets, 20% of the data was held out for testing. The data split was implemented at the position level to ensure that no position from the test data subset appears in the training data subset.

For more details about the tool, please visit the help page or get in touch with us.

14-Dec-2023 update: the file with features PredictONCO-features.txt now includes UniProt IDs, transcripts, PDB codes, and mutations.

Adapting Large Language Models to Domains via Continual Pre-Training

This repo contains the ConvFinQA dataset used in our ICLR 2024 paper Adapting Large Language Models via Reading Comprehension. We explore continued pre-training on domain-specific corpora for large language models. While this approach enriches LLMs with domain knowledge, it significantly hurts their prompting ability for question answering. Inspired by human learning via reading comprehension, we propose a… See the full description on the dataset page: https://huggingface.co/datasets/AdaptLLM/ConvFinQA.

Cross-validation is a common method to validate a QSAR model. In cross-validation, some compounds are held out as a test set, while the remaining compounds form a training set. A model is built from the training set, and the test set compounds are predicted on that model. The agreement of the predicted and observed activity values of the test set (measured by, say, R2) is an estimate of the self-consistency of the model and is sometimes taken as an indication of the predictivity of the model. This estimate of predictivity can be optimistic or pessimistic compared to true prospective prediction, depending how compounds in the test set are selected. Here, we show that time-split selection gives an R2 that is more like that of true prospective prediction than the R2 from random selection (too optimistic) or from our analog of leave-class-out selection (too pessimistic). Time-split selection should be used in addition to random selection as a standard for cross-validation in QSAR model building.

Here we supply the training and test data as used in the prepared publication of "Convolutional Neural Network Applied for Nanoparticle Classification using Coherent Scaterometry Data" by D. Kolenov, D. Davidse, J. Le Cam, S.F. Pereira.

We present the "main dataset" samples in the pixel size of both 150x150 and 100x100, and for the three "fooling datasets" the pixel size is 100x100. On average each dataset contains 1100 images with the .mat extension. The .mat extension is straightforward with MatLab, but it could also be opened in Python or MS Excel. For the "main dataset" the pixels represent the sampling points, and the magnitude of these pixels represent the em field registered as the photocurrent on the split-detector. For the three types of "fooling data" the images of a 1) noisy and 2) mirrored set are also based on the photocurrent; 3) the elephant set is based on the open-source Animal-10 data.

Splits of aggregated data into testing and training subsets.

This benchmark data is comprised of 50 different datasets for materials properties obtained from 16 previous publications. The data contains both experimental and computational data, data suited for regression as well as classification, sizes ranging from 12 to 6354 samples, and materials systems spanning the diversity of materials research. In addition to cleaning the data where necessary, each dataset was split into train, validation, and test splits.

For datasets with more than 100 values, train-val-test splits were created, either with a 5-fold or 10-fold cross-validation method, depending on what each respective paper did in their studies. Datasets with less than 100 values had train-test splits created using the Leave-One-Out cross-validation method.

For further information, as well as directions on how to access the data, please go to the corresponding GitHub repository: https://github.com/anhender/mse_ML_datasets/tree/v1.0

This is the training and testing data used to train a Residual Attention UNet for segmentation and detection of road culverts. The data consists of pairs of images with the size 256x256 pixels where one image is a labeled mask and the other a image with four channels containing the remote sensing data. The remote sensing data is a combination of topographical data extracted from arial laser scanning and ortophotos from arial imagery. An extensive culvert survey was conducted in 25 watersheds in central Sweden by the Swedish Forest Agency during the snow-free periods of 2014–2017. A total of 24,083 culverts were mapped with a handheld GPS with a horizontal accuracy of 0.3 m. Densely populated urban areas with underground drainage systems were excluded from the survey (0.3% of the combined area). The coordinates of both ends of each culvert were measured, and metrics such as diameter, length, material, working condition, and sediment accumulation were collected for most of the culverts. Additional metrics, such as the elevation difference between the outlet and stream water level, were manually measured with a ruler. The inventoried watersheds were split up into training and testing data, where 20 watersheds (23,304 culverts) were used for training, and five watersheds (5,208 culverts) were used for testing. A compact laser-based system (Leica ALS80-HP-8236) was used to collect the ALS data from an aircraft flying at 2888–3000 m. The ALS point clouds had a point density of 1–2 points m-2 and were divided into tiles with a size of 2.5 x 2.5 km each. A DEM with 0.5 m resolution was created from the ALS point clouds using a TIN gridding approach implemented in Whitebox tools 2.2.0. The topographical index max downslope elevation change was calculated from the DEM using Whitebox Tools . Max downslope elevation change represents the maximum elevation drop between each grid cell and its neighbouring cells within a DEM. This typically resulted in values between 0 and 10. Orthophotos from aerial imagery captured at the same time as the lidar data is also included. The orthophotos had three bands (red, green and blue) in 8-bit color depth and had a resolution of 0.5 m. The LiDAR data and orthophotos were downloaded from the Swedish mapping, cadastral and land registration authority. The topographical data and the ortophotos were merged into 8-bit four band images where the first three band is red, green and blue, and the last band is max downslope elevation change. The merged images where then split into smaller tiles with the size 256x256 pixels. The trained model was used to predict culverts in Sweden and the file PredictedCulvertsByIsobasins.zip contains the predicted culverts stored as shapefiles split by the watersheds in the file "isobasins.zip".

This repository accompanies the manuscript "Spatially resolved uncertainties for machine learning potentials" by E. Heid, J. Schörghuber, R. Wanzenböck, and G. K. H. Madsen. The following files are available:

• mc_experiment.ipynb is a Jupyter notebook for the Monte Carlo experiment described in the study (artificial model with only variance as error source).

• aggregate_cut_relax.py contains code to cut and relax boxes for the water active learning cycle.

• data_t1x.tar.gz contains reaction pathways for 10,073 reactions from a subset of the Transition1x dataset, split into training, validation and test sets. The training and validation sets contain the indices 1, 2, 9, and 10 from a 10-image nudged-elastic band search (40k datapoints), while the test set contains indices 3-8 (60k datapoints). The test set is ordered according to the reaction and index, i.e. rxn1_index3, rxn1_index4, [...] rxn1_index8, rxn2_index3, [...].

• data_sto.tar.gz contains surface reconstructions of SrTiO3, randomly split into a training and validation set, as well as a test set.

• data_h2o.tar.gz contains:

  • full_db.extxyz: The full dataset of 1.5k structures.

  • iter00_train.extxyz and iter00_validation.extxyz: The initial training and validation set for the active learning cycle.

  • the subfolders in the folders random and uncertain contain the training and validation sets for the random and uncertainty-based active learning loops.