Dataset documentation improvements (#116)

* datasets description and tables

* [github-action] formatting fixes

* Center statistics table

* Update Benchmark documentation

* [github-action] formatting fixes

* Add UCI link

---------

Co-authored-by: pereverges <pere.verges@estudiantat.upc.edu>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: mikeheddes <mikeheddes@gmail.com>

Author: 4 people

docs/_static/css/custom.css

@@ -48,10 +48,10 @@ a:hover {
 
 html.writer-html4 .rst-content dl:not(.docutils)>dt,
 html.writer-html5 .rst-content dl[class]:not(.option-list):not(.field-list):not(.footnote):not(.glossary):not(.simple)>dt {
-    background: #f3f4f7;
-    color: #6c6c6d;
-    border-top: none;
-    border-left: 3px solid #ee4c2c;
+    background: #f3f4f7 !important;
+    color: #6c6c6d !important;
+    border-top: none !important;
+    border-left: 3px solid #ee4c2c !important;
     padding: 0.5rem;
     padding-right: 100px;
     word-wrap: break-word;

docs/datasets.rst

@@ -7,139 +7,10 @@ The Torchhd library provides many popular built-in datasets to work with.
 
 .. currentmodule:: torchhd.datasets
 
-.. autosummary:: 
-    :toctree: generated/
-    :template: class_dataset.rst
+.. automodule:: torchhd.datasets
+    :members:
+    :exclude-members: download, process, processed_file_names, raw_file_names, num_classes, DatasetFourFold, DatasetTrainTest, CollectionDataset
 
-    BeijingAirQuality
-    ISOLET
-    EuropeanLanguages
-    UCIHAR
-    AirfoilSelfNoise
-    EMGHandGestures
-    PAMAP
-    CyclePowerPlant
-    Abalone
-    Adult
-    AcuteInflammation    
-    AcuteNephritis
-    Annealing
-    Arrhythmia
-    AudiologyStd
-    BalanceScale
-    Balloons
-    Bank
-    Blood
-    BreastCancer
-    BreastCancerWisc
-    BreastCancerWiscDiag
-    BreastCancerWiscProg
-    BreastTissue
-    Car
-    Cardiotocography3Clases
-    Cardiotocography10Clases
-    ChessKrvk
-    ChessKrvkp
-    CongressionalVoting
-    ConnBenchSonarMinesRocks
-    ConnBenchVowelDeterding
-    Connect4
-    Contrac
-    CreditApproval
-    CylinderBands
-    Dermatology
-    Echocardiogram
-    Ecoli
-    EnergyY1
-    EnergyY2
-    Fertility
-    Flags
-    Glass
-    HabermanSurvival
-    HayesRoth
-    HeartCleveland
-    HeartHungarian
-    HeartSwitzerland
-    HeartVa
-    Hepatitis
-    HillValley
-    HorseColic
-    IlpdIndianLiver
-    ImageSegmentation
-    Ionosphere
-    Iris
-    LedDisplay
-    Lenses
-    Letter
-    Libras
-    LowResSpect
-    LungCancer
-    Lymphography
-    Magic
-    Mammographic
-    Miniboone
-    MolecBiolPromoter
-    MolecBiolSplice
-    Monks1
-    Monks2
-    Monks3
-    Mushroom
-    Musk1
-    Musk2
-    Nursery
-    OocytesMerlucciusNucleus4d
-    OocytesMerlucciusStates2f
-    OocytesTrisopterusNucleus2f
-    OocytesTrisopterusStates5b
-    Optical
-    Ozone
-    PageBlocks
-    Parkinsons
-    Pendigits
-    Pima
-    PittsburgBridgesMaterial
-    PittsburgBridgesRelL
-    PittsburgBridgesSpan
-    PittsburgBridgesTOrD
-    PittsburgBridgesType
-    Planning
-    PlantMargin
-    PlantShape
-    PlantTexture
-    PostOperative
-    PrimaryTumor
-    Ringnorm
-    Seeds
-    Semeion
-    Soybean
-    Spambase
-    Spect
-    Spectf
-    StatlogAustralianCredit
-    StatlogGermanCredit
-    StatlogHeart
-    StatlogImage
-    StatlogLandsat
-    StatlogShuttle
-    StatlogVehicle
-    SteelPlates
-    SyntheticControl
-    Teaching
-    Thyroid
-    TicTacToe
-    Titanic
-    Trains
-    Twonorm
-    VertebralColumn2Clases
-    VertebralColumn3Clases
-    WallFollowing
-    Waveform
-    WaveformNoise
-    Wine
-    WineQualityRed
-    WineQualityWhite
-    Yeast
-    Zoo
 
 Base classes
 ------------------------
@@ -148,7 +19,6 @@ Base classes
     :toctree: generated/
     :template: class_dataset.rst
 
-    UCIClassificationBenchmark
     CollectionDataset
     DatasetFourFold
-    DatasetTrainTest
+    DatasetTrainTest

examples/UCI_benchmark_intRVFL.py

@@ -28,7 +28,6 @@ def classifier_ridge_regression(
     data_type: torch.dtype,
     device: torch.device,
 ):
-
     # Get number of training samples
     num_train = len(train_ld.dataset)
     # Collects high-dimensional represetations of data in the train data
@@ -49,7 +48,6 @@ def classifier_ridge_regression(
     with torch.no_grad():
         count = 0
         for samples, labels in tqdm(train_ld, desc="Training"):
-
             samples = samples.to(device)
             labels = labels.to(device)
             # Make one-hot encoding

examples/reghd.py

@@ -50,6 +50,7 @@ def target_transform(x):
 train_dl = data.DataLoader(train_ds, batch_size=1, shuffle=True)
 test_dl = data.DataLoader(test_ds, batch_size=1)
 
+
 # Model based on RegHD application for Single model regression
 class SingleModel(nn.Module):
     def __init__(self, num_classes, size):

torchhd/datasets/abalone.py

@@ -5,6 +5,20 @@
 class Abalone(DatasetFourFold):
     """`Abalone <https://archive.ics.uci.edu/ml/datasets/abalone>`_ dataset.
 
+    .. list-table::
+       :widths: 10 10 10 10
+       :align: center
+       :header-rows: 1
+
+       * - Instances
+         - Attributes
+         - Task
+         - Area
+       * - 4177
+         - 8
+         - Classification
+         - Life
+
     Args:
         root (string): Root directory containing the files of the dataset.
         train (bool, optional): If True, returns training (sub)set from the file storing training data as further determined by fold and hyper_search variables.
@@ -22,6 +36,7 @@ class Abalone(DatasetFourFold):
         download (bool, optional): If True, downloads the dataset from the internet and
             puts it in root directory. If dataset is already downloaded, it is not
             downloaded again.
+
     """
 
     name = "abalone"

torchhd/datasets/acute_inflammation.py

@@ -5,6 +5,20 @@
 class AcuteInflammation(DatasetFourFold):
     """`Acute Inflammation of urinary bladder <https://archive.ics.uci.edu/ml/datasets/Acute+Inflammations>`_ dataset.
 
+    .. list-table::
+       :widths: 10 10 10 10
+       :align: center
+       :header-rows: 1
+
+       * - Instances
+         - Attributes
+         - Task
+         - Area
+       * - 120
+         - 6
+         - Classification
+         - Life
+
     Args:
         root (string): Root directory containing the files of the dataset.
         train (bool, optional): If True, returns training (sub)set from the file storing training data as further determined by fold and hyper_search variables.
@@ -22,6 +36,7 @@ class AcuteInflammation(DatasetFourFold):
         download (bool, optional): If True, downloads the dataset from the internet and
             puts it in root directory. If dataset is already downloaded, it is not
             downloaded again.
+
     """
 
     name = "acute-inflammation"

torchhd/datasets/acute_nephritis.py

@@ -5,6 +5,20 @@
 class AcuteNephritis(DatasetFourFold):
     """`Acute Nephritis of renal pelvis origin <https://archive.ics.uci.edu/ml/datasets/Acute+Inflammations>`_ dataset.
 
+    .. list-table::
+       :widths: 10 10 10 10
+       :align: center
+       :header-rows: 1
+
+       * - Instances
+         - Attributes
+         - Task
+         - Area
+       * - 120
+         - 6
+         - Classification
+         - Social
+
     Args:
         root (string): Root directory containing the files of the dataset.
         train (bool, optional): If True, returns training (sub)set from the file storing training data as further determined by fold and hyper_search variables.
@@ -22,6 +36,7 @@ class AcuteNephritis(DatasetFourFold):
         download (bool, optional): If True, downloads the dataset from the internet and
             puts it in root directory. If dataset is already downloaded, it is not
             downloaded again.
+
     """
 
     name = "acute-nephritis"

torchhd/datasets/adult.py

@@ -5,6 +5,20 @@
 class Adult(DatasetTrainTest):
     """`Adult <https://archive.ics.uci.edu/ml/datasets/adult>`_ dataset.
 
+    .. list-table::
+       :widths: 10 10 10 10
+       :align: center
+       :header-rows: 1
+
+       * - Instances
+         - Attributes
+         - Task
+         - Area
+       * - 48842
+         - 14
+         - Classification
+         - Social
+
     Args:
         root (string): Root directory containing the files of the dataset.
         train (bool, optional): If True, returns training (sub)set from the file storing training data as further determined by hyper_search variable.
@@ -18,6 +32,7 @@ class Adult(DatasetTrainTest):
         download (bool, optional): If True, downloads the dataset from the internet and
             puts it in root directory. If dataset is already downloaded, it is not
             downloaded again.
+
     """
 
     name = "adult"

torchhd/datasets/airfoil_self_noise.py

@@ -13,6 +13,19 @@ class AirfoilSelfNoise(data.Dataset):
     Dataset is obtained from a series of aerodynamic and acoustic tests of two and three-dimensional
     airfoil blade sections conducted in an anechoic wind tunnel.
 
+    .. list-table::
+       :widths: 10 10 10 10
+       :align: center
+       :header-rows: 1
+
+       * - Instances
+         - Attributes
+         - Task
+         - Area
+       * - 1503
+         - 6
+         - Regression
+         - Physical
 
     Args:
         root (string): Root directory of dataset where ``airfoil_self_noise.dat`` exists
@@ -23,6 +36,7 @@ class AirfoilSelfNoise(data.Dataset):
             and returns a transformed version.
         target_transform (callable, optional): A function/transform that takes in the
             target and transforms it.
+
     """
 
     def __init__(

torchhd/datasets/annealing.py

@@ -5,6 +5,20 @@
 class Annealing(DatasetTrainTest):
     """`Annealing <https://archive.ics.uci.edu/ml/datasets/Annealing>`_ dataset.
 
+    .. list-table::
+       :widths: 10 10 10 10
+       :align: center
+       :header-rows: 1
+
+       * - Instances
+         - Attributes
+         - Task
+         - Area
+       * - 798
+         - 38
+         - Classification
+         - Physical
+
     Args:
         root (string): Root directory containing the files of the dataset.
         train (bool, optional): If True, returns training (sub)set from the file storing training data as further determined by hyper_search variable.
@@ -18,6 +32,7 @@ class Annealing(DatasetTrainTest):
         download (bool, optional): If True, downloads the dataset from the internet and
             puts it in root directory. If dataset is already downloaded, it is not
             downloaded again.
+
     """
 
     name = "annealing"