CODEWITHSUNDEEP
pythontensorflowkeras
shini
shini

Reputation: 141

my input np array turns into different shape when input in the keras model

The image has a shape of (512, 2048, 3) but I get a ValueError when running preds_train = new_model.predict(img, batch_size=1) :

ValueError: Could not find matching function to call loaded from the SavedModel. Got:
  Positional arguments (3 total):
    * Tensor("inputs:0", shape=(1, 2048, 3), dtype=float32)
    * False
    * None
  Keyword arguments: {}

Expected these arguments to match one of the following 4 option(s):

Option 1:
  Positional arguments (3 total):
    * TensorSpec(shape=(None, 512, 2048, 3), dtype=tf.float32, name='inputs')
    * True
    * None
  Keyword arguments: {}

Option 2:
  Positional arguments (3 total):
    * True
    * None
  Keyword arguments: {}

Option 3:
  Positional arguments (3 total):
    * TensorSpec(shape=(None, 512, 2048, 3), dtype=tf.float32, name='input_1')
    * False
    * None
  Keyword arguments: {}

Option 4:
  Positional arguments (3 total):
    * TensorSpec(shape=(None, 512, 2048, 3), dtype=tf.float32, name='inputs')
    * False
    * None
  Keyword arguments: {}

I've printed the initial dimensions of the shape so I'm sure it's (512, 2048, 3) and additionally when i trained the model i did so using images of that shape. I don't know why I can't test the model.

Upvotes: 0

Views: 60

Answers (2)

Bernad Peter
Bernad Peter

Reputation: 514

You probably need to reshape with one more dimension since your model is trained in that way, for example try this:

width,height,ch=img.shape
img_input=img.reshape((1,width,height,ch))

Upvotes: 0

Vijeth Rai
Vijeth Rai

Reputation: 320

Try using model.evaluate(x_test)

Upvotes: 2

Related Questions