Continual-inference

Latest version: v1.2.3

Safety actively analyzes 706267 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 7 of 8

0.6.1

Changed
- `co.Residual` modules to be unnamed. This allows the module state dicts to be flattened.

0.6.0

Added
- Flattened state dict export and loading via a `flatten` argument. This feature improves interoperability complex modules, that were not originally constructed with the `co.Sequential` and `co.Parallel` building blocks.
- Context manager for triggering flattened state_dict export and loading.

0.5.0

Added
- Support for zero-delay in `co.Delay`
- Support for broadcasting in `co.Parallel`
- Mul (hadamard product) aggregation in `co.Parallel`
- Example of Squeeze and Excitation block

Changed
- `co._PoolNd` attribute naming: "temporal_*" removed as prefix for kernel_size, stride, dilation, and padding.

0.4.0

Added
- `co.Delay` handling for padding.
- Handling of initialization and strides in containers

Changed
- `co.Conv` `build_from` behavior to not change dilation and stride. Argument overload supported instead.
- `pad_start` and `pad_end` args to convolution and pooling modules `forward_steps`.
- Behavior of modules while they initialize. Now, a TensorPlaceholder is passed for initializing steps.

Removed
- Automatic unsqueeze in pooling.

0.3.1

Added
- Support for dropout.

0.3.0

Added
- Support for dilation and stride in pooling.

Changed
- Pooling API to match torch.nn better.
- `_ConvCoNd.forward_steps` doesn't invoke `clean_state` anymore.

Page 7 of 8

© 2025 Safety CLI Cybersecurity Inc. All Rights Reserved.