Execution failue in 'build_index_tool': Unsupported File Type

[WilliamWatsonEAI]

Please can you help me to resolve this issue when Executing the prompt flow bulk run in the chat_with_pdf_ci_dev_workflow. I have searched exhaustively for a solution and cannot seem to determine what is going wrong. I have followed the workshop guide, FAQ and "working with github flows".

Apologies if this is not the correct forum for such a request but there is no instructions provided for getting support (https://github.com/microsoft/llmops-promptflow-template/blob/main/SUPPORT.md) and I am really stumped/cannot successfully deploy the LLMOPS template.

Below is a summary of my issue, with sensitive information redacted:

Run Summary

Run Details

• Run name: chat_with_pdf_20240729_140552_chat_with_pdf
• Run status: Completed
• Start time: 2024-07-29 14:06:19.212003+00:00
• Duration: 0:00:14.572418
• Run URL: Run Details

Run Metadata

• Name: chat_with_pdf_20240729_140552_chat_with_pdf
• Created on: 2024-07-29T14:06:00.889680+00:00
• Status: Completed
• Display name: chat_with_pdf_20240729_140552_chat_with_pdf
• Description:
• Tags:
  • build_id: 10145832774
• Properties:
  • azureml.promptflow.inputs_mapping: "math_question":"$data.question"
  • azureml.promptflow.runtime_name: automatic
  • azureml.promptflow.disable_trace: false
  • azureml.promptflow.session_id: c830a1b9a46730b77d57609a907e090cd91115d88070f2ee
  • azureml.promptflow.definition_file_name: flow.dag.yaml
  • azureml.promptflow.flow_lineage_id: d3534c65e5ee3a6a559789ba7cebd698fd2a289546e6aec92b4f5d23c8c1fe34
  • azureml.promptflow.flow_definition_datastore_name: workspaceblobstore
  • azureml.promptflow.flow_definition_blob_path: LocalUpload/216811aaf8dfdf1f9ca6148b99925b0a/standard/flow.dag.yaml
  • _azureml.evaluation_run: promptflow.BatchRun
  • azureml.promptflow.snapshot_id: 929e78db-5284-423a-86a5-67bff901a3db
  • azureml.promptflow.runtime_version: 20240709.v1
  • _azureml.evaluate_artifacts: [{"path": "instance_results.jsonl", "type": "table"}]
  • azureml.promptflow.total_tokens: 308
  • azureml.promptflow.completion_tokens: 9
  • azureml.promptflow.prompt_tokens: 299

Creation Context

• User Object ID: [REDACTED]
• User PU ID: [REDACTED]
• User IDP: [REDACTED]
• User Alt Sec ID: [REDACTED]
• User Issuer: [REDACTED]
• User Tenant ID: [REDACTED]
• User Name: [REDACTED]
• UPN: [REDACTED]

Timing

• Start time: 2024-07-29T14:06:19.212003+00:00
• End time: 2024-07-29T14:06:33.784421+00:00
• Duration: 00:00:14.5724183

Portal URL

• Run URL: Run Details

Data

• Data: azureml:chat_with_pdf:2
• Output: azureml://locations/australiaeast/workspaces/[REDACTED]/data/azureml_chat_with_pdf_20240729_140552_chat_with_pdf_output_data_flow_outputs/versions/1

Error

• Code: UserError
• Severity:
• Message: Execution failure in 'build_index_tool': (BadRequestError) Unsupported data type
• Message Format: "totalChildRuns": 1, "userErrorChildRuns": 1, "systemErrorChildRuns": 0, "errorDetails": [{"code": "UserError/ToolExecutionError", "messageFormat": "Execution failure in 'node_name'.", "count": 1}]
• Message Parameters:
  • Node Name: build_index_tool
• Reference Code: Tool/__pf_main__
• Details URI:
• Target:
• Details: []
• Inner Error:
  • Code: ToolExecutionError
  • Inner Error:
• Debug Info:
  • Type: ToolExecutionError
  • Message: Execution failure in 'build_index_tool': (BadRequestError) Unsupported data type
  • Stack Trace:

    The above exception was the direct cause of the following exception:
    Traceback (most recent call last):
      File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/promptflow/_core/flow_execution_context.py", line 182, in _invoke_tool_inner
        return f(**kwargs)
      File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/promptflow/tracing/_trace.py", line 556, in wrapped
        output = func(*args, **kwargs)
      File "/mnt/host/service/app/37099/requests/chat_with_pdf_20240729_140552_chat_with_pdf/build_index_tool.py", line 9, in build_index_tool
        return create_faiss_index(pdf_path)
      File "/mnt/host/service/app/37099/requests/chat_with_pdf_20240729_140552_chat_with_pdf/chat_with_pdf/build_index.py", line 57, in create_faiss_index
        index.insert_batch(segments)
      File "/mnt/host/service/app/37099/requests/chat_with_pdf_20240729_140552_chat_with_pdf/chat_with_pdf/utils/index.py", line 39, in insert_batch
        vector = self.embedding.generate(text)
      File "/mnt/host/service/app/37099/requests/chat_with_pdf_20240729_140552_chat_with_pdf/chat_with_pdf/utils/retry.py", line 21, in wrapper
        return func(*args, **kwargs)
      File "/mnt/host/service/app/37099/requests/chat_with_pdf_20240729_140552_chat_with_pdf/chat_with_pdf/utils/oai.py", line 119, in generate
        return self.client.embeddings.create(
      File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/openai/resources/embeddings.py", line 114, in create
        return self._post(
      File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/openai/_base_client.py", line 1266, in post
        return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls))
      File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/openai/_base_client.py", line 942, in request
        return self._request(
      File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/openai/_base_client.py", line 1046, in _request
        raise self._make_status_error_from_response(err.response) from None
    openai.BadRequestError: Unsupported data type