Expand description
Input validation utilities for Python bindings
Functionsยง
- validate_
betas - Validate beta parameters for Adam-like optimizers
- validate_
broadcast_ shapes - Validate that dimensions are compatible for broadcasting
- validate_
conv_ params - Validate convolution parameters
- validate_
dimension - Validate that a dimension index is valid for a tensor
- validate_
dropout_ probability - Validate dropout probability is in valid range [0, 1]
- validate_
epsilon - Validate that epsilon is positive
- validate_
finite - Validate that a value is finite (not NaN or infinity)
- validate_
index - Validate that an index is within bounds for a given dimension
- validate_
kernel_ size - Validate kernel size is positive
- validate_
learning_ rate - Validate that a learning rate is positive
- validate_
momentum - Validate that momentum is in valid range [0, 1]
- validate_
num_ features - Validate that number of features matches expected value
- validate_
parameters_ not_ empty - Validate that parameters list is not empty
- validate_
pooling_ output_ size - Validate pooling output size calculation
- validate_
range - Validate that a range is valid (start < end)
- validate_
shape - Validate that a shape is valid (all dimensions > 0)
- validate_
stride - Validate stride is positive
- validate_
tensor_ min_ ndim - Validate that input tensor has at least minimum number of dimensions
- validate_
tensor_ ndim - Validate that input tensor has expected number of dimensions
- validate_
tensor_ shapes_ match - Validate that tensor dimensions match for operations
- validate_
weight_ decay - Validate that weight decay is non-negative