I implemented an AI prediction model(GRU) and by running the code, the prediction result will show in Google Sheets. I customer the menu of Google sheet and added a run prediction button and want the user by choosing this option, can see the prediction result.
I need to deploy my AI model in GC, I created a function, defined environment variables, created a bucket, and uploaded models. but when I choose run prediction in Google Sheets, I get a format error, while the format of my model is Keras.
Error loading Day 2 model: File format not supported: filepath=/tmp/tmpci6d3rvv. Keras 3 only supports V3 `.keras` files and legacy H5 format files (`.h5` extension). Note that the legacy SavedModel format is not supported by `load_model()` in Keras 3. In order to reload a TensorFlow SavedModel as an inference-only layer in Keras 3, use `keras.layers.TFSMLayer(/tmp/tmpci6d3rvv, call_endpoint='serving_default')` (note that your `call_endpoint` might have a different name).
LinkedIn
Twitter
