`_ (search
-for "matbench").
-
*Note: Larger datasets will take several minutes to load.*
\ No newline at end of file
diff --git a/docs/_static/documentation_options.js b/docs/_static/documentation_options.js
index f7f7e1a0..77127823 100644
--- a/docs/_static/documentation_options.js
+++ b/docs/_static/documentation_options.js
@@ -1,6 +1,6 @@
var DOCUMENTATION_OPTIONS = {
URL_ROOT: document.getElementById("documentation_options").getAttribute('data-url_root'),
- VERSION: '1.0.2.20191110',
+ VERSION: '1.0.3.20191111',
LANGUAGE: 'None',
COLLAPSE_INDEX: false,
FILE_SUFFIX: '.html',
diff --git a/docs/advanced.html b/docs/advanced.html
index 187a30a5..8fe5bf1c 100644
--- a/docs/advanced.html
+++ b/docs/advanced.html
@@ -4,7 +4,7 @@
Advanced Usage — Automatminer 1.0.2.20191110 documentation
+ Advanced Usage — Automatminer 1.0.3.20191111 documentation
automatminer.automl.config package — Automatminer 1.0.2.20191110 documentation
+ automatminer.automl.config package — Automatminer 1.0.3.20191111 documentation
automatminer.automl package — Automatminer 1.0.2.20191110 documentation
+ automatminer.automl package — Automatminer 1.0.3.20191111 documentation
automatminer.automl.tests package — Automatminer 1.0.2.20191110 documentation
+ automatminer.automl.tests package — Automatminer 1.0.3.20191111 documentation
automatminer.featurization package — Automatminer 1.0.2.20191110 documentation
+ automatminer.featurization package — Automatminer 1.0.3.20191111 documentation
Submodules) – If True, returns a multiindexed dataframe. Not
recommended for use in MatPipe.
- n_jobs (int
do_precheck (bool
n_jobs (int
The number of parallel jobs to use during featurization
+for each featurizer. Default is n_cores
+
+composition_col=”composition”,
+
+composition_col (str
structure_col (str
bandstructure (str
dos_col (str
automatminer.featurization.tests package — Automatminer 1.0.2.20191110 documentation
+ automatminer.featurization.tests package — Automatminer 1.0.3.20191111 documentation
automatminer package — Automatminer 1.0.2.20191110 documentation
+ automatminer package — Automatminer 1.0.3.20191111 documentation
automatminer.preprocessing package — Automatminer 1.0.2.20191110 documentation
+ automatminer.preprocessing package — Automatminer 1.0.3.20191111 documentation
automatminer.preprocessing.tests package — Automatminer 1.0.2.20191110 documentation
+ automatminer.preprocessing.tests package — Automatminer 1.0.3.20191111 documentation
automatminer.tests package — Automatminer 1.0.2.20191110 documentation
+ automatminer.tests package — Automatminer 1.0.3.20191111 documentation
automatminer.utils package — Automatminer 1.0.2.20191110 documentation
+ automatminer.utils package — Automatminer 1.0.3.20191111 documentation
automatminer.utils.tests package — Automatminer 1.0.2.20191110 documentation
+ automatminer.utils.tests package — Automatminer 1.0.3.20191111 documentation
Basic Usage — Automatminer 1.0.2.20191110 documentation
+ Basic Usage — Automatminer 1.0.3.20191111 documentation
MatBench benchmark — Automatminer 1.0.2.20191110 documentation
+ MatBench benchmark — Automatminer 1.0.3.20191111 documentation
Navigation
MatBench benchmark
Tutorials and Support MatBench benchmark Details on the benchmark are coming soon in a publication we have submitted.
@@ -85,11 +87,145 @@
Overview
We have made the MatBench benchmark publicly available via the matminer
datasets repository
-(and also via Figshare ).
-You can download the datasets with the matminer.datasets.load_dataset matbench-* *
-Here’s the MatBench benchmark for predicting refractive index (calculated with
+and through the Materials Project MPContribs-ML Deployment .
+All the Matbench datasets begin with matbench_
+Here’s a full list of the 13 datasets in Matbench v0.1:
+
+
+
+
+dataset name
target column
number of samples
task type
download link
+
+matbench_dielectric
n
4764
regression
link
matbench_expt_gap
gap expt
4604
regression
link
matbench_expt_is_metal
is_metal
4921
classification
link
matbench_glass
gfa
5680
classification
link
matbench_jdft2d
exfoliation_en
636
regression
link
matbench_log_gvrh
log10(G_VRH)
10987
regression
link
matbench_log_kvrh
log10(K_VRH)
10987
regression
link
matbench_mp_e_form
e_form
132752
regression
link
matbench_mp_gap
gap pbe
106113
regression
link
matbench_mp_is_metal
is_metal
106113
classification
link
matbench_perovskites
e_form
18928
regression
link
matbench_phonons
last phdos peak
1265
regression
link
matbench_steels
yield strength
312
regression
link
+
+
+
+
Getting dataset info
+
You can get more info (such as the meaning of column names, brief cleaning
+procedures, etc.) on a dataset with matminer.datasets.get_all_dataset_info
+
from matminer.datasets import get_all_dataset_info
+
+# Get dataset info from matminer
+info = get_all_dataset_info ( "matbench_steels" )
+
+# Check out the info about the dataset.
+print ( info )
+
+
+
+
+
You can also view all the Matbench datasets on the matminer
+Dataset Summary page (search
+for “matbench”).
+
+
(Down)loading datasets
+
While you can download the zipped json datasets via the download links above, we
+recommend using matminer’s tools to load datasets. Matminer intelligently manages the
+dataset downloads in its central folder and provides methods for robustly loading dataframes containing
+pymatgen primitives such as structures.
+
You can load the datasets with the matminer.datasets.load_dataset
from matminer.datasets import load_dataset
@@ -101,6 +237,22 @@ Accessing MatBench
+
+
+
This loads the dataframe in this format:
df matbench_dielectric
@@ -127,9 +279,6 @@ Accessing MatBench (search
-for “matbench”).
Note: Larger datasets will take several minutes to load.
diff --git a/docs/genindex.html b/docs/genindex.html
index e210cf05..13dd2fa0 100644
--- a/docs/genindex.html
+++ b/docs/genindex.html
@@ -5,7 +5,7 @@
Index — Automatminer 1.0.2.20191110 documentation
+ Index — Automatminer 1.0.3.20191111 documentation
automatminer documentation — Automatminer 1.0.2.20191110 documentation
+ automatminer documentation — Automatminer 1.0.3.20191111 documentation
Tutorials and Support
diff --git a/docs/installation.html b/docs/installation.html
index c722a654..21ddbfc0 100644
--- a/docs/installation.html
+++ b/docs/installation.html
@@ -4,7 +4,7 @@
Installation — Automatminer 1.0.2.20191110 documentation
+ Installation — Automatminer 1.0.3.20191111 documentation
License — Automatminer 1.0.2.20191110 documentation
+ License — Automatminer 1.0.3.20191111 documentation
automatminer — Automatminer 1.0.2.20191110 documentation
+ automatminer — Automatminer 1.0.3.20191111 documentation
Python Module Index — Automatminer 1.0.2.20191110 documentation
+ Python Module Index — Automatminer 1.0.3.20191111 documentation
Search — Automatminer 1.0.2.20191110 documentation
+ Search — Automatminer 1.0.3.20191111 documentation
`_ for materials science; a
-set of 13 benchmarking ML problems for fair comparison, across a wide domain of
+set of 13 supervised ML tasks for benchmarking and fair comparison spanning a wide domain of
inorganic materials science applications.
.. image:: _static/matbench_pie_charts.png
@@ -20,18 +20,141 @@ For now, you can still access the benchmark datasets. See the "Accessing MatBenc
section for more info.
+
Accessing MatBench
------------------
We have made the MatBench benchmark publicly available via the `matminer
datasets repository `_
-(and also via `Figshare `_).
+and through the `Materials Project MPContribs-ML Deployment `_.
+All the Matbench datasets begin with :code:`matbench_`.
+
+
+Here's a full list of the 13 datasets in Matbench v0.1:
+
+.. list-table::
+ :align: left
+ :header-rows: 1
+
+ * - dataset name
+ - target column
+ - number of samples
+ - task type
+ - download link
+ * - :code:`matbench_dielectric`
+ - :code:`n`
+ - 4764
+ - regression
+ - `link `_
+ * - :code:`matbench_expt_gap`
+ - :code:`gap expt`
+ - 4604
+ - regression
+ - `link `_
+ * - :code:`matbench_expt_is_metal`
+ - :code:`is_metal`
+ - 4921
+ - classification
+ - `link `_
+ * - :code:`matbench_glass`
+ - :code:`gfa`
+ - 5680
+ - classification
+ - `link `_
+ * - :code:`matbench_jdft2d`
+ - :code:`exfoliation_en`
+ - 636
+ - regression
+ - `link `_
+ * - :code:`matbench_log_gvrh`
+ - :code:`log10(G_VRH)`
+ - 10987
+ - regression
+ - `link `_
+ * - :code:`matbench_log_kvrh`
+ - :code:`log10(K_VRH)`
+ - 10987
+ - regression
+ - `link `_
+ * - :code:`matbench_mp_e_form`
+ - :code:`e_form`
+ - 132752
+ - regression
+ - `link `_
+ * - :code:`matbench_mp_gap`
+ - :code:`gap pbe`
+ - 106113
+ - regression
+ - `link `_
+ * - :code:`matbench_mp_is_metal`
+ - :code:`is_metal`
+ - 106113
+ - classification
+ - `link `_
+ * - :code:`matbench_perovskites`
+ - :code:`e_form`
+ - 18928
+ - regression
+ - `link `_
+ * - :code:`matbench_phonons`
+ - :code:`last phdos peak`
+ - 1265
+ - regression
+ - `link `_
+ * - :code:`matbench_steels`
+ - :code:`yield strength`
+ - 312
+ - regression
+ - `link `_
+
+
+Getting dataset info
+--------------------
+
+You can get more info (such as the meaning of column names, brief cleaning
+procedures, etc.) on a dataset with :code:`matminer.datasets.get_all_dataset_info`:
+
+.. code-block:: python
+
+ from matminer.datasets import get_all_dataset_info
+
+ # Get dataset info from matminer
+ info = get_all_dataset_info("matbench_steels")
+
+ # Check out the info about the dataset.
+ print(info)
-You can download the datasets with the :code:`matminer.datasets.load_dataset`
-function; the names of the datasets are named :code:`matbench-*` where :code:`*`
-is the name of the benchmark problem.
-Here's the MatBench benchmark for predicting refractive index (calculated with
+.. code-block:: text
+
+ Dataset: matbench_steels
+ Description: Matbench v0.1 dataset for predicting steel yield strengths from chemical composition alone. Retrieved from Citrine informatics. Deduplicated.
+ Columns:
+ composition: Chemical formula.
+ yield strength: Target variable. Experimentally measured steel yield strengths, in GPa.
+ Num Entries: 312
+ Reference: https://citrination.com/datasets/153092/
+ Bibtex citations: ['@misc{Citrine Informatics,\ntitle = {Mechanical properties of some steels},\nhowpublished = {\\url{https://citrination.com/datasets/153092/},\n}']
+ File type: json.gz
+ Figshare URL: https://ml.materialsproject.org/matbench_steels.json.gz
+
+
+You can also view all the Matbench datasets on the matminer
+`Dataset Summary page `_ (search
+for "matbench").
+
+
+(Down)loading datasets
+-----------------------
+
+While you can download the zipped json datasets via the download links above, we
+recommend using matminer's tools to load datasets. Matminer intelligently manages the
+dataset downloads in its central folder and provides methods for robustly loading dataframes containing
+pymatgen primitives such as structures.
+
+You can load the datasets with the :code:`matminer.datasets.load_dataset`
+function; the function accepts the dataset name as an argument.
+Here's an example of loading the Matbench task for predicting refractive index (calculated with
DFPT) from crystal structure.
.. code-block:: python
@@ -46,6 +169,25 @@ DFPT) from crystal structure.
print(df)
+.. code-block:: text
+
+ structure n
+ 0 [[4.29304147 2.4785886 1.07248561] S, [4.2930... 1.752064
+ 1 [[3.95051434 4.51121437 0.28035002] K, [4.3099... 1.652859
+ 2 [[-1.78688104 4.79604117 1.53044621] Rb, [-1... 1.867858
+ 3 [[4.51438064 4.51438064 0. ] Mn, [0.133... 2.676887
+ 4 [[-4.36731958 6.8886097 0.50929706] Li, [-2... 1.793232
+ ... ...
+ 4759 [[ 2.79280881 0.12499663 -1.84045389] Ca, [-2... 2.136837
+ 4760 [[0. 5.50363806 3.84192106] O, [4.7662... 2.690619
+ 4761 [[0. 0. 0.] Ba, [ 0.23821924 4.32393487 -0.35... 2.811494
+ 4762 [[0. 0.18884638 0. ] K, [0. ... 1.832887
+ 4763 [[0. 0. 0.] Cs, [2.80639641 2.80639641 2.80639... 2.559279
+ [4764 rows x 2 columns]
+
+
+This loads the dataframe in this format:
+
:code:`df` (:code:`matbench_dielectric`)
.. list-table::
@@ -64,8 +206,4 @@ DFPT) from crystal structure.
- ...
-Find all the MatBench problem names and info
-`here `_ (search
-for "matbench").
-
*Note: Larger datasets will take several minutes to load.*
\ No newline at end of file
diff --git a/docs/tutorials.html b/docs/tutorials.html
index a4fc7031..45d57164 100644
--- a/docs/tutorials.html
+++ b/docs/tutorials.html
@@ -4,7 +4,7 @@
Tutorials and Support — Automatminer 1.0.2.20191110 documentation
+ Tutorials and Support — Automatminer 1.0.3.20191111 documentation
