TabFusionTransformer

This is a a Pytorch implementeation of TabTransformerTransformer created by Ignacio Oguiza (oguiza@timeseriesAI.co)

This implementation is inspired by:

Huang, X., Khetan, A., Cvitkovic, M., & Karnin, Z. (2020). TabTransformer: Tabular Data Modeling Using Contextual Embeddings. arXiv preprint https://arxiv.org/pdf/2012.06678

Official repo: https://github.com/awslabs/autogluon/tree/master/tabular/src/autogluon/tabular/models/tab_transformer

---

source

TabFusionTransformer

def TabFusionTransformer(
    classes, cont_names, c_out, d_model:int=32, n_layers:int=6, n_heads:int=8, d_k:NoneType=None, d_v:NoneType=None,
    d_ff:NoneType=None, res_attention:bool=True, attention_act:str='gelu', res_dropout:float=0.0,
    fc_mults:tuple=(4, 2), fc_dropout:float=0.0, fc_act:NoneType=None, fc_skip:bool=False, fc_bn:bool=False,
    bn_final:bool=False, init:bool=True
):

Class that allows you to pass one or multiple inputs

---

source

TabFusionBackbone

def TabFusionBackbone(
    classes, cont_names, d_model:int=32, n_layers:int=6, n_heads:int=8, d_k:NoneType=None, d_v:NoneType=None,
    d_ff:NoneType=None, init:bool=True, res_attention:bool=True, attention_act:str='gelu', res_dropout:float=0.0
):

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing them to be nested in a tree structure. You can assign the submodules as regular attributes::

import torch.nn as nn
import torch.nn.functional as F

class Model(nn.Module):
    def __init__(self) -> None:
        super().__init__()
        self.conv1 = nn.Conv2d(1, 20, 5)
        self.conv2 = nn.Conv2d(20, 20, 5)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        return F.relu(self.conv2(x))

Submodules assigned in this way will be registered, and will also have their parameters converted when you call :meth:to, etc.

.. note:: As per the example above, an __init__() call to the parent class must be made before assignment on the child.

:ivar training: Boolean represents whether this module is in training or evaluation mode. :vartype training: bool

---

source

Sequential

def Sequential(
    args:VAR_POSITIONAL
):

Class that allows you to pass one or multiple inputs

---

source

ifnone

def ifnone(
    a, b
):

b if a is None else a

from fastai.tabular.all import *

path = untar_data(URLs.ADULT_SAMPLE)
df = pd.read_csv(path/'adult.csv')
dls = TabularDataLoaders.from_csv(path/'adult.csv', path=path, y_names="salary",
    cat_names = ['workclass', 'education', 'marital-status', 'occupation', 'relationship', 'race'],
    cont_names = ['age', 'fnlwgt', 'education-num'],
    procs = [Categorify, FillMissing, Normalize])
x_cat, x_cont, yb = first(dls.train)
model = TabFusionTransformer(dls.classes, dls.cont_names, dls.c)
test_eq(model(x_cat, x_cont).shape, (dls.train.bs, dls.c))

---

source

TSTabFusionTransformer

def TSTabFusionTransformer(
    c_in, c_out, seq_len, classes, cont_names, d_model:int=32, n_layers:int=6, n_heads:int=8, d_k:NoneType=None,
    d_v:NoneType=None, d_ff:NoneType=None, res_attention:bool=True, attention_act:str='gelu', res_dropout:float=0.0,
    fc_mults:tuple=(1, 0.5), fc_dropout:float=0.0, fc_act:NoneType=None, fc_skip:bool=False, fc_bn:bool=False,
    bn_final:bool=False, init:bool=True
):

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing them to be nested in a tree structure. You can assign the submodules as regular attributes::

import torch.nn as nn
import torch.nn.functional as F

class Model(nn.Module):
    def __init__(self) -> None:
        super().__init__()
        self.conv1 = nn.Conv2d(1, 20, 5)
        self.conv2 = nn.Conv2d(20, 20, 5)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        return F.relu(self.conv2(x))

Submodules assigned in this way will be registered, and will also have their parameters converted when you call :meth:to, etc.

.. note:: As per the example above, an __init__() call to the parent class must be made before assignment on the child.

:ivar training: Boolean represents whether this module is in training or evaluation mode. :vartype training: bool

classes = {'education': ['#na#', '10th', '11th', '12th', '1st-4th', '5th-6th', '7th-8th', '9th', 'Assoc-acdm', 'Assoc-voc', 'Bachelors', 'Doctorate', 
                         'HS-grad', 'Masters', 'Preschool', 'Prof-school', 'Some-college'],
 'education-num_na': ['#na#', False, True],
 'marital-status': ['#na#', 'Divorced', 'Married-AF-spouse', 'Married-civ-spouse', 'Married-spouse-absent', 'Never-married', 'Separated', 'Widowed'],
 'occupation': ['#na#', '?', 'Adm-clerical', 'Armed-Forces', 'Craft-repair', 'Exec-managerial', 'Farming-fishing', 'Handlers-cleaners', 'Machine-op-inspct', 
                'Other-service', 'Priv-house-serv', 'Prof-specialty', 'Protective-serv', 'Sales', 'Tech-support', 'Transport-moving'],
 'race': ['#na#', 'Amer-Indian-Eskimo', 'Asian-Pac-Islander', 'Black', 'Other', 'White'],
 'relationship': ['#na#', 'Husband', 'Not-in-family', 'Other-relative', 'Own-child', 'Unmarried', 'Wife'],
 'workclass': ['#na#', '?', 'Federal-gov', 'Local-gov', 'Never-worked', 'Private', 'Self-emp-inc', 'Self-emp-not-inc', 'State-gov', 'Without-pay']}

cont_names = ['a', 'b', 'c']
c_out = 6
x_ts = torch.randn(64, 3, 10)
x_cat = torch.randint(0,3,(64,7))
x_cont = torch.randn(64,3)
model = TSTabFusionTransformer(x_ts.shape[1], c_out, x_ts.shape[-1], classes, cont_names)
x = (x_ts, (x_cat, x_cont))
test_eq(model(x).shape, (x_ts.shape[0], c_out))