creating TFrecord file for multiple objects

Question

I am following the racoon detection tutorial from GitHub (https://github.com/datitran/raccoon_dataset) to detect animals using google object detection API. For this, I need to generate tfrecord file which is already generated here on line number 29 to 34 (https://github.com/datitran/raccoon_dataset/blob/master/generate_tfrecord.py).

But he has done the code for only one animal (racoon) line 29 to 34. I have multiple animals like a racoon, praying mantis, hermit crab etc how do I modify this tfrecord file for multiple animals. One way I found is making changes to line 29 to 34 in generatetfrecord file as follows

def class_text_to_int(row_label):
    if row_label == 'raccoon':
        return 1
    if row_label == 'prayingmantis':
        return 2
    else:
        None

Is this approach corect to include multiple if in same file or I need to generate multiple tfrecord files to train multiple objects

Answer 1

def class_text_to_int(row_label):
    if row_label == 'raccoon':
        return 1
    elif row_label == 'prayingmantis':
        return 2
    else:
        None

Making the above change in generatetfrecord is a correct approach to train multiple custom images simultaniously. Also, in object detection.pbtxt file, make the following changes:

item{
       id:1
       name:'racoon'
      }
item{
       id:2
       name: 'prayingmantis'
      }

And retrain the model from the first step.