Can you provide "Real" command to test the Text evaluation please ?

Bartzi / see

Code for the AAAI 2018 publication "SEE: Towards Semi-Supervised End-to-End Scene Text Recognition"

GNU General Public License v3.0

573 stars 147 forks source link

Can you provide "Real" command to test the Text evaluation please ? #86

Open Sanix-Darker opened 4 years ago

Sanix-Darker commented 4 years ago

For the command :

python evaluate.py textrec 
<path to dir with specified items> 
<name of snapshot to evaluate> 
<path to ground truth file> 
<path to char map (e.g. ctc_char_map.json)> 23

Is it possible to provide real values please ? For a working example.

Bartzi commented 4 years ago

Hi,

let's assume you downloaded the file text_recognition_model.zip from the Website mentioned in the README. You extract the archive and put it in the directory ~/recognition. You cloned the repository into the directory ~/see and you are currently in that directory. Then the following should work:

python evaluate.py textrec ~/recognition/model model_190000.npz ~/recognition/small_dataset/gt_word.csv ~/recognition/small_dataset/ctc_char_map.json