search

jeffheaton / t81_558_deep_learning

T81-558: Keras - Applications of Deep Neural Networks @Washington University in St. Louis
https://sites.wustl.edu/jeffheaton/t81-558/
Other
5.71k stars 3.04k forks source link

T81-558: Applications of Deep Neural Networks Module 13: Advanced/Other Topics #82

Closed strategiepilot closed 3 years ago

strategiepilot commented 3 years ago

Hi Jeff, first of all, thanks a ton for your wonderful course, the hands-on coding and giving people the opportunity to access them for free online!

I've been trying to follow along your tutorial in module 18 to deploy a pretrained model with AWS Sagemaker. In chapter 13.2 Step 2: Load the Keras model using the JSON and weights file i am running into the followin value error:

_ValueError: Tensor-typed variable initializers must either be wrapped in an init_scope or callable (e.g., tf.Variable(lambda : tf.truncated_normal([10, 40]))) when building functions. Please file a feature request if this restriction inconveniences you._

I tried to wrap it into a tf.Variable(lambda ...) but that gets me some new errors. Not sure, how to resolve this but really would like to be able to process further in this tutorial. thx for looking into this. Andreas from Germany

jeffheaton commented 3 years ago

I will have to try to run through this example again. AWS may have changed something since when I first created it.

jeffheaton commented 3 years ago

I am replacing this material with other topics for the course.