For the package installation, first install all the requirements and then install the sa_forst package.
$ pip install -r requirements.txt
$ python setup.py install
Self-Attention Forest model has a scikit-learn like interface.
It is extended with optimize_weights method which can be executed with the same training data as used for an underlying forest training, or with a new data set.
Code example for model instantiation:
from sa_forest import (
SAFParams,
SelfAttentionForest,
ForestKind,
TaskType,
)
model = SelfAttentionForest(
SAFParams(
kind=ForestKind.EXTRA,
task=TaskType.REGRESSION,
eps=0.9,
tau=1.0,
gamma=0.9,
sa_tau=1.0,
sa_dist='y',
forest=dict(
n_estimators=200,
max_depth=None,
min_samples_leaf=5,
random_state=12345,
),
)
)
After the underlying forest should be trained:
model.fit(X_train, y_train)
And then weights are optimized:
model.optimize_weights(X_train, y_train)
In order to estimate weights optimization impact scores for model.predict_original(X_val) and model.predict(X_val) could be compared.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent