2D Materials · foundry-ml
dataset_exfoliatione
A tabular dataset of calculated exfoliation energies for 636 materials to support discovery of 2D layered compounds.
What dataset_exfoliatione is
dataset_exfoliatione is a materials science dataset containing calculated exfoliation energies for 636 materials. Compiled as part of the Joint Automated Repository for Various Integrated Simulations (JARVIS) initiative by Kamal Choudhary and colleagues, the resource is distributed via Foundry-ML and the Materials Data Facility.
The dataset addresses the challenge of identifying bulk layered materials that can be mechanically or chemically isolated into stable two-dimensional monolayers. By providing quantified exfoliation energies across diverse chemical compositions, it enables computational researchers to develop predictive screening models for two-dimensional materials design.
How it was made
The dataset is structured as tabular records pairing chemical composition with calculated physical properties. Each entry contains a chemical formula as an input feature and a calculated exfoliation energy expressed in electronvolts per atom (eV/atom) as the target variable.
The collection is provided in a tabular format, such as Parquet, and is packaged with a single training split suitable for regression and classification tasks.
Key facts
- Dataset size
- 636 materials
- Data type
- Tabular
- Input feature
- formula (material composition)
- Target feature
- target (exfoliation energy in eV/atom)
- Task categories
- Tabular regression, tabular classification
- Data format
- Parquet
- Release year
- 2020
- DOI
- 10.18126/yx4h-ny59
What you can use it for
- Exfoliation energy regression
- Train machine learning regression models to predict exfoliation energy directly from stoichiometric material formulas.
- High-throughput 2D material screening
- Screen candidate inorganic compounds to identify layered structures with low exfoliation energies suitable for monolayer synthesis.
- Exfoliability classification
- Develop binary or multi-class classifiers to categorise layered materials based on whether their exfoliation energy falls below practical thresholds.
Explore the data
The schema, real rows, full-text search and a SQL filter — read in your browser, nothing downloaded.
Explore
Read the real rows without downloading anything
Reading rows…
LiveRead from foundry-ml/dataset_exfoliatione at the moment you asked. Nothing is cached or stored — every row above came from that request.
How the columns are distributed
Over 636 rows of default/train
formula
string_text- min
- 2
- median
- 6
- mean
- 6.64
- max
- 11
target
float- min
- 0.03
- median
- 79.80
- mean
- 112.70
- max
- 1604.04
Strengths and limits
Good at
- Provides direct quantitative exfoliation energies in eV/atom derived from JARVIS simulations.
- Formatted simply as chemical formulas and corresponding numerical targets, allowing quick integration into tabular machine learning pipelines.
Watch out for
- Contains only 636 entries, representing a relatively small sample for training complex deep learning architectures.
- Includes only stoichiometric formulas without accompanying structural coordinates or detailed simulation metadata in the tabular fields.
- Does not provide predefined validation or test splits, requiring practitioners to partition the single train split manually.
Questions
Can I look inside dataset_exfoliatione without downloading it?
Yes. This page reads its schema, real rows and per-column statistics in the browser, with full-text search and a SQL filter. Nothing is downloaded to your machine.
What licence is dataset_exfoliatione under?
Its authors declare other. Read the terms on the original page before reusing the data.
What target property is measured in this dataset?
The dataset records calculated exfoliation energy in units of electronvolts per atom (eV/atom).
What input features are available for modeling?
The input feature is the material composition, supplied as a chemical formula string.
How many compound records are included?
The dataset contains calculated values for 636 materials.
How to cite it
Choudhary, Kamal, Garrity, Kevin F., Reid, Andrew C. E., DeCost, Brian, Biacchi, Adam J., Walker, Angela R. Hight, Trautt, Zachary, Hattrick-Simpers, Jason, Kusne, A. Gilad, Centrone, Andrea, Davydov, Albert, Jiang, Jie, Pachter, Ruth, Cheon, Gowoon, Reed, Evan, Agrawal, Ankit, Qian, Xiaofeng, Sharma, Vinit, Zhuang, Houlong, Kalinin, Sergei V., Sumpter, Bobby G., Pilania, Ghanshyam, Acar, Pinar, Mandal, Subhasish, Haule, Kristjan, Vanderbilt, David, Rabe, Karin, Tavazza, Francesca. The joint automated repository for various integrated simulations (JARVIS) for data-driven materials design. Materials Data Facility (2020). https://doi.org/10.18126/yx4h-ny59
Written from dataset_exfoliatione's own dataset card. Statements citing a figure that does not appear there are removed before publication. Licence and running details come from our registry.