How to test classic sample flowers

I new in TF, I tried the classic flowers recognition sample:

I replicate this in PyCharm and it works but I dont’ understand how to test this script passing some real flowers picture to test this in real world.

Thank in advance


You can follow this tutorial to get an idea. Go through the particular section I linked and if there’s any doubt, let me know.


If you want to play with this on your Android device camera you could try to follow this tutorial:


Many thanks for info.
I’m trying to create a custom classification using my set of images but I’m getting lost in hundreds of examples all different each other. :grinning:



Hi @Paolo_Pini You can try tf.keras.Model.predict as in name_of_your_model.predict(...) (be mindful of tensor shapes).


  1. Image classification  |  TensorFlow Core

Predict on new data

Finally, let’s use our model to classify an image that wasn’t included in the training or validation sets.

sunflower_url = ""
sunflower_path = tf.keras.utils.get_file('Red_sunflower', origin=sunflower_url)

img = keras.preprocessing.image.load_img(
    sunflower_path, target_size=(img_height, img_width)
img_array = keras.preprocessing.image.img_to_array(img)
img_array = tf.expand_dims(img_array, 0) # Create a batch

predictions = model.predict(img_array)
  1. Writing your own callbacks  |  TensorFlow Core

Now, define a simple custom callback that logs:

  • When fit/evaluate/predict starts & ends
  • When each epoch starts & ends
  • When each training batch starts & ends
  • When each evaluation (test) batch starts & ends
  • When each inference (prediction) batch starts & ends
class CustomCallback(keras.callbacks.Callback):
res = model.predict(x_test, batch_size=128, callbacks=[CustomCallback()])