how would you describe a machine learning model in a scientific report? It should be detailed but I just listed the hyperparameters... Have you got more important properties?
Some other details you could mention are:
total number of model parameters (e.g. 1.2M or 0.15M) & depth of the network (e.g. 38-layered network)
family/style of the network architecture (e.g. encoder-decoder arch., LSTM)
- specifics of connections between network layers (e.g. residual-, dense-, skip-connections)
- specifics of individual components of the network structure (e.g. dilated-convs. (CNNs), attention (LSTMs))
- description/reasoning of why you chose a particular structure/sequence of connections in your deep learning model
- specifics of training/validation/testing procedures (e.g. augmented training data, cross-validation, test-time-augmentation (TTA), frozen network weights)
- other specific details/caveats that allow the results of your deep learning model be easily reproduced from the scientific report
For more info on the best kinds of details to be included in the report, refer to "Methodology"/ "Training"/ "Implementation"/ "Proposed Architecture" sections of the deep learning research papers in your relevant area.