* Add dummy batch dimension to meet the requirements of the batch norm ([`0f499f0`](https://github.com/es-ude/elastic-ai.creator/commit/0f499f048c0606de3e14163f16e8bf049708e6f1))
0.55.1
Fix
* Fix non existing in_channels variable and remove unused import ([`0e73c2d`](https://github.com/es-ude/elastic-ai.creator/commit/0e73c2dc6876772d0caba46639af77bd5ac53b62))
* Typing and errors ([`af6f859`](https://github.com/es-ude/elastic-ai.creator/commit/af6f85913fd6111bcc7164a106a9cbb8d4b7b9a0))
0.54.0
Feature
* Use binary values instead of hex values to fill the rom template ([`af56c02`](https://github.com/es-ude/elastic-ai.creator/commit/af56c02da42433c2db1a9a2a6ddb3705d213d765))
Fix
* Use same bit width for all rom values ([`cd609e6`](https://github.com/es-ude/elastic-ai.creator/commit/cd609e65f306e62110fbdc4113f4bb330f960f19))
0.53.0
Feature
* Implement fixed point one dimensional convolution ([`2ea9389`](https://github.com/es-ude/elastic-ai.creator/commit/2ea9389a37eac7be62e26a9727b8824b47fc2085))
Fix
* Fix missing parameter in tests for conv1d ([`d8f8d4c`](https://github.com/es-ude/elastic-ai.creator/commit/d8f8d4c40ec1576c5dc58a38b2b80d9d4130b4fd))
0.52.0
Feature
* Added Swish activation function precomputed ([`26d292e`](https://github.com/es-ude/elastic-ai.creator/commit/26d292e83183ac0b6bee7afa70f3d616e42b2438)) * Added Swish activation function precomputed ([`fd487b5`](https://github.com/es-ude/elastic-ai.creator/commit/fd487b57bb7d3e7525f935f1533f815e58f1dc0d)) * Added nn module for Swish activation function ([`b7579c9`](https://github.com/es-ude/elastic-ai.creator/commit/b7579c9f1111521064a7fd0366647da7a45e2d7a)) * Implemeted base module for SiLU aka Swish activation function ([`93b5954`](https://github.com/es-ude/elastic-ai.creator/commit/93b59544c1d164de2a9f9362f0aefe1aaae8d7d8))