Error testing out the sample.py

[zairm21]

Hi,
Thanks for sharing this amazing work. However I can't run the sample code to try it out.
I exported all the required exports but I get this error when I try the sample.py code: openai.error.InvalidRequestError: The API deployment for this resource does not exist. If you created the deployment within the last 5 minutes, please wait a moment and try again.
Does it mean this deployment_name: gpt-35-turbo-version-0301 is not available anymore? If so what is the available one?

Thanks!

[omer-erkek]

Faced the same problem. You need to deploy a model and write its name as deployment_name. See here and click Deployments. From there you can deploy a model

[zairm21]

Thanks @omer-erkek , that worked!

[zairm21]

Hi @omer-erkek ,
Were you able to test the code successfully? Because when I call output = chain.conversation("https://www.oracle-dba-online.com/sql/weekly_sales_table.png"), it apparently gets stuck with this message "Entering new AgentExecutor chain..." and killed after a while.

I also tried it with a text message as prompt to test the APIs:
output = chain.conversation("Describe this image")
the output is:
Entering new AgentExecutor chain...
1. There is an image in the input
2. The image is new in the conversation

<|im_sep|>AI
I am sorry, I need an image to describe. Please provide an image.

Finished chain.

However when I pass the image to chain.conversation, gets stuck with "> Entering new AgentExecutor chain...". Is any specific message format required?

Thanks!