Hello there guys. First time making a post here.

So I am trying to make an AI chatbot using Python, in a Pycharm IDLE. While trying to start training the neural network that would enable the chatbot to work, I ran into this error and was unable to find any resources I could use to help me solve this:

Traceback (most recent call last):
File "C:\Users\Owner\PycharmProjects\ai\main.py", line 71, in model.compile(loss='categorical_crossentropy', optimiser=sgd, metrics=['accuracy'])

File "C:\Users\Owner\AppData\Roaming\Python\Python39\site-packages\keras\utils\traceback_utils.py", line 67, in error_handler raise e.with_traceback(filtered_tb) from None

File "C:\Users\Owner\AppData\Roaming\Python\Python39\site-packages\keras\engine\training.py", line 2983, in _validate_compile raise TypeError('Invalid keyword argument(s) in `compile()`: '

TypeError: Invalid keyword argument(s) in `compile()`: ({'optimiser'},). Valid keyword arguments include "cloning", "experimental_run_tf_function", "distribute", "target_tensors", or "sample_weight_mode".

Original Code

ie the code where the error occured

sgd = gradient_descent_v2.SGD(learning_rate=0.01, decay=1e-6, momentum=0.9, nesterov=True)
model.compile(loss='categorical_crossentropy', optimiser=sgd, metrics=['accuracy'])

I also had to import SGD like this because otherwise it couldn't be found:

from keras.optimizers import gradient_descent_v2

If anybody knows how to solve this, please tell me!

This is my first ever post too.

The problem is:

model.compile(loss='categorical_crossentropy', optimiser=sgd, metrics=['accuracy'])

The order of arguments should be changed. And the sgd should be included in single quotes. Also, the optimiser should be optimizer with a z.

So it should look like this:

model.compile(optimizer='sgd', loss='categorical_crossentropy', metrics=['accuracy'])

Also, in regard to importing, try this:

from tensorflow.keras.optimizers import SGD And make sure you imported the categorical_crossentropy before. A helpful document from keras website: https://www.tensorflow.org/api_docs/python/tf/keras/Model. And here you could find 30 examples that might be helpful: https://www.programcreek.com/python/example/97109/keras.losses.categorical_crossentropy.

Good luck with your project!