I'm trying to deploy a CNN model through the Google Cloud API. The model has already been uploaded to Google Cloud. I specified, while creating my Cloud Bucket, a Python version of 3.7, and Tensorflow 2.1.0. I'm trying to deploy by running the following code in my Anaconda terminal:
gcloud ai-platform predict --region europe-west1 --model <model name> --version <version name> --json-instances check_deployed_model/test.json
However, the following error keeps popping up, even after double checking my model was indeed trained in Tensorflow 2.1.0:
ERROR: (gcloud.ai-platform.predict) HTTP request failed. Response: { "error": { "code": 400, "message": "{ "error": "Error while reading resource variable conv2d/bias from Container: localhost. This could mean that the variable was uninitialized. Not found: Container localhost does not exist. (Could not find resource: localhost/conv2d/bias)\n\t [[{{node model/sequential/conv2d/BiasAdd/ReadVariableOp}}]]" }", "status": "INVALID_ARGUMENT" } }
I have already tried this advice, but I couldn't seem to make it work.
Any advice on how I could proceed?