@@ -193,6 +193,7 @@ class default_trainer(object):
Zum starten des LSTMS einfach in den ordner caption_lib navigieren und von dort aus [start.py](/caption_lib/start.py) ausführen.
### Evaluation
| Measure | Singlelabel | Multilabel | Show & Tell (2015) | SOTA (2017) |
@@ -206,6 +207,15 @@ Zum starten des LSTMS einfach in den ordner caption_lib navigieren und von dort
| Bleu-4 | 0.141 | 0.140 | 0.309 | **0.363** |
| SPICE | 0.092 | 0.089 | 0.182 | **0.213** |
### Attention
## visualisation
[Attention_good](/Demo/Attention_good.png)
Here we can see that the attention reacts on specific words and if it does not react at all then we can assume that there was a learning only from the captions not from the vector we provided.
## Bad visualisation
[Attention_bad](/Demo/Attention_bad.png)
The attention can't be maped correctly because we do not know the dimensions the pictures have and how the CNN changed the data during the learning.
