Model Assessment/Convergence #53
Description
Hello, wonderful tool! I had some questions about assessing the quality of a model in the context of hyper-parameter tuning (my first guess was the model's loss but without having included verbose=True when running is there any way to access it in post?). Are there any metrics used/recommended to assess how good SPECTRA's results are? I have attached some measures I have created but I would appreciate some insight, especially given the lack of change with increased epochs.
