[solved] Easy way to use a lot of photos?

Comments

• MaximeRosito

  September 2014 edited 7:36AM

  I have exactly the same request
  We have a set of 3500 fractals, and a lot of variation of the code.
  The hard drive is quite full...
• sebastiaan

  September 2014 edited 7:36AM

  Hi, I already posted this question a couple of days ago but I can't find the discussion anymore.

  Technical issues, sorry. But we're back online now!

  Yes, you can certainly read images from a folder. OpenSesame doesn't just accept files from the file pool, it also looks in the experiment folder, and accepts relative and absolute paths.

  In your case, what you could do is build a list of images at the start of your experiment. Something like this:

  import os
  import random
  # Specify the folder with your images
  my_image_folder = '/home/sebastiaan/Pictures'
  # Walk through all files in this folder ...
  my_image_list = []
  for fname in os.listdir(my_image_folder):
      # ... but only take those that end with .png. Change this to .jpg,
      # .bmp, etc. as required.
      if fname.endswith('.png'):
          # Add the full path (i.e. including the folder) of the image to
          # a list.
          path = os.path.join(my_image_folder, fname)
          my_image_list.append(path)
  # Shuffle the list
  random.shuffle(my_image_list)
  # Print it out so we can see in the debug window. This way you can check
  # if things work as expected.
  print my_image_list
  

  Once you have this list (my_image_list), you can simply take one item from the top at the beginning of every trial. Like so (in the prepare phase on an inline_script):

  # Take the last image from the list ('pop').
  image = my_image_list.pop()
  # And set it as an experimental variable.
  exp.set('image', image)
  

  Because you've set image as an experimental variable, you can log it, use it in a sketchpad, etc.

  See? Not that hard, really.

  Cheers!
  Sebastiaan

  
  
  Check out SigmundAI.eu for our OpenSesame AI assistant!
  
• MaximeRosito

  September 2014 edited 7:36AM

  Thank's a lot I'll try as soon as possible
• floor_30

  September 2014 edited 7:36AM

  Yes, thank you! This helps a lot.
  However, with each photo, there is also a correct keyboard response. Is there a way to link a correct keyboard response to a particular picture this way?
• floor_30

  September 2014 edited 7:36AM

  so, to clarify: can I implement something that says "if the filename ends with 3.jpg the correct keybord response is 3"?
• floor_30

  September 2014 edited 7:36AM

  Nevermind, I found a way which works! (after a little bit of testing).
  for those who are interested:
  after
  exp.set('image', image)
  write: if image.endswith('3.JPG'):
  exp.set("correct_answer", 3)

  Thanks again for your help Sebastiaan, you saved me a lot of work!