* For nearly all metrics (except to CIDEr) we have a difference of 0.1 to 0.2 - which is quite a margin but can be explained by
considering that we used only ~7% of the trainings data for the LSTM.
considering that we used only ~7% of the training data for the LSTM.
* More reasons for our lower results: the cnn wasn't fully trained (precision was still low), no extended hyperparameter tuning for LSTM, no beam search implemented
* Another effect which does not show in the evaluation metric is that even though the scores of *Singlelabel* and *Multilabel* are quite similar,
we observed some difference when we looked at it manually. *Singlelabel* often produced more gramatically sound sentences, but not fitting to the objects in the image,
@@ -370,7 +370,7 @@ At this step we implemented [attention](#attention), but ran into problems becau
input image proportions that varied to a high degree.
This is reflected in our [evaluation](#evaluation) scores. As automatic scores are unreliable for that task we would have loved to do a manual evaluation - but that is left for
future work, as well as proper implementation of attention, beam search, parameter tuning for the LSTM, extraction of all feature vectors from the dataset and extended
