Hi, I’m making a fully connected model.
I’ve trained the model with same data set that is generated by ‘DataGenerator’ library and set same seed.
The hyperparameters of the model were also fixed.
However, the performance of the model was very different. some of training cases were trained well, and some of others were not at all. I don’t know how it is possible and how I interpret it. There I can’t go further steps like tuning the hyperparameters because I am not sure about my model works well.
Please give me advise. Thanks and have a nice day.