When you train a neural network, you usually use 3 sets: one for training, one for development, one for testing.
Your training set is here for (obviously) training your model: the performance of your model on your training set reflects how well your model learnt "by heart" what you showed it.
Your development set is used jointly during training: ...
The network is able to overfit the dataset, given enough epochs. So even if it performs great it doesn't mean that it will behave this way with unknown data.
Assuming that the accuracy is the right metric for your problem, you have to take into account only the results on the test set. These show the performance of your model in unknown data.
This is the ...
If I understand correctly in this specific case I agree with you, it's reasonable to count such a sample as both TP and FP. However this must be explained clearly when you describe the evaluation method, since it's not the standard behaviour of precision/recall.
The alternative more standard version would be to apply a true multi-label evaluation, i.e. ...
The training error shouldn't be too far from test error, otherwise it is a high deviance scenario and you could be in an overfitting situation in production.
However, having a higher deviance could be normal by increasing depth, but it shouldn't happen if you have enough data.
Consequently, if you haven't a lot of data, the depth of 1 seems better, and you ...
Primarily, go for CV for the training and test set. If you still get the same type of result, then choose the second model.
The first model has a very large difference in accuracy between the training and test set.
It is a very specific model.
There is a chance that the high accuracy on the test set appeared due to data leakage.
The second model is a more ...
Before the deep learning wave, the the UCI dataset repository was widely used.
It contains classic (and rather small) datasets that were very relevant in the old days, like the Iris dataset for classification.
In each dataset page, you can find papers citing the dataset.
The prediction algorithm resulted in only positives, and no negatives, i.e. it predicted that every member of the roster was a member of the target class, with the target class being "your pick".
Of the positive results predicted by the algorithm, there was 1 true positive and N-1 false positives, from a roster of size N.
In that case, the recall ...
The loss is the sum of errors based on your train and validation datasets, whereas the accuracy is the percentage of good results obtained with the validation dataset.
So, if your loss is decreasing and your accuracy is increasing, it means that it has worse results in your train data sets but better ones in your validation dataset.
There is probably a way ...