Update README.md

* **Restriction:**  

  Because extracting all vectors from images took too long we limited the dataset to 100 pictures per category, resulting in 8000 vectors total (which are used as input for the LSTM). We expect that this decision will get us inferior end results, but had to be done to stay within time limits.

* the vector for a single picture can be extracted with the following scripts for the respective models: [vector_from_image_singlelabel.py](/preprocessing/vector_from_image_singlelabel.py) and [vector_from_image_multilabel.py](/preprocessing/vector_from_image_multilabel.py)  

    example use:

    ```

    python vector_from_image_multilabel.py -image ../some_test_images/cat3.jpg -graph ../models/multilabel_default/retrained_graph.pb -label ../cocoapi/labels.txt

    ```

* to create an equal distribution without intersecting pictures we used the following [script](preprocessing/coco_preprocessing/100.py):

```

python 100.py