Torch split classes stratified

X = np.random.randn(1000, 2)
y = np.random.randint(0, 10, size=1000)

X_train, X_val, y_train, y_val = train_test_split(X, y, test_size=0.1, stratify=y)
np.unique(y_train, return_counts=True)
np.unique(y_val, return_counts=True)

train_dataset = Dataset(X_train, y_train, ...)
train_loader = DataLoader(train_dataset, ...)

Source: discuss.pytorch.org

Similar codes

0 vote342 views1 answer
Python mayusculas
Python
0 vote361 views2 answers
Use reshape in python with zeros
Python
0 vote382 views1 answer
Gcp functions save BQ
Python
0 vote349 views1 answer
Django builtin signals
Python Django
0 vote309 views1 answer
Django router multiple pk
Python
2 votes404 views3 answers
Check if value exists in list python
Python
0 vote381 views1 answer
App is not a registered namespace django
Python Django
8 votes359 views4 answers
Sklearn plot confusion matrix
Python
0 vote380 views1 answer
Python multiline comment using docstrings
Python
12 votes395 views5 answers
Pause program python
Python