String concatenation error when using local LLM with llama-cpp-python server

[shivam-saway]

When using LIDA with a local LLM hosted via llama-cpp-python server, I encounter a string concatenation error during the data summarization process. Despite this error, the summarization appears to work correctly and returns valid JSON output.

Environment
LIDA Version: [latest]
Local LLM: Microsoft Phi-3-mini-4k-instruct (GGUF format)
LLM Server: llama-cpp-python (ghcr.io/abetlen/llama-cpp-python:v0.3.5)
Setup: Docker Compose with OpenAI-compatible API endpoint

ERROR:

Error processing file: can only concatenate str (not "dict") to str

2025-09-27 15:01:48.891 | INFO:     Will watch for changes in these directories: ['/app']
2025-09-27 15:01:48.891 | INFO:     Uvicorn running on http://0.0.0.0:8080 (Press CTRL+C to quit)
2025-09-27 15:01:48.891 | INFO:     Started reloader process [1] using StatReload
2025-09-27 15:01:50.512 | INFO:     Started server process [29]
2025-09-27 15:01:50.512 | INFO:     Waiting for application startup.
2025-09-27 15:01:50.512 | INFO:     Application startup complete.
2025-09-27 15:09:34.747 | Error processing file: can only concatenate str (not "dict") to str
2025-09-27 15:08:22.880 | INFO:     172.18.0.1:39926 - "GET /api/models HTTP/1.1" 200 OK
2025-09-27 15:09:34.747 | <|assistant|> {
2025-09-27 15:09:34.747 |   "name": "mock_sales_data_2020_2021_limited.csv",
2025-09-27 15:09:34.747 |   "file_name": "mock_sales_data_2020_2021_limited.csv",
2025-09-27 15:09:34.747 |   "dataset_description": "",
2025-09-27 15:09:34.747 |   "fields": [
2025-09-27 15:09:34.747 |     {
2025-09-27 15:09:34.747 |       "column": "Year",
2025-09-27 15:09:34.747 |       "properties": {
2025-09-27 15:09:34.747 |         "dtype": "number",
2025-09-27 15:09:34.747 |         "std": 0,
2025-09-27 15:09:34.747 |         "min": 2020,
2025-09-27 15:09:34.747 |         "max": 2021,
2025-09-27 15:09:34.747 |         "samples": [2021, 2020],
2025-09-27 15:09:34.747 |         "num_unique_values": 2,
2025-09-27 15:09:34.747 |         "semantic_type": "",
2025-09-27 15:09:34.747 |         "description": ""
2025-09-27 15:09:34.747 |       }
2025-09-27 15:09:34.747 |     },
2025-09-27 15:09:34.747 |     {
2025-09-27 15:09:34.747 |       "column": "Date",
2025-09-27 15:09:34.747 |       "properties": {
2025-09-27 15:09:34.747 |         "dtype": "date",
2025-09-27 15:09:34.747 |         "min": "2020-01-01",
2025-09-27 15:09:34.747 |         "max": "2021-01-03",
2025-09-27 15:09:34.747 |         "samples": ["2020-01-01", "2020-01-02"],
2025-09-27 15:09:34.747 |         "num_unique_values": 6,
2025-09-27 15:09:34.747 |         "semantic_type": "",
2025-09-27 15:09:34.747 |         "description": ""
2025-09-27 15:09:34.747 |       }
2025-09-27 15:09:34.747 |     },
2025-09-27 15:09:34.747 |     {
2025-09-27 15:09:34.747 |       "column": "Category",
2025-09-27 15:09:34.747 |       "properties": {
2025-09-27 15:09:34.747 |         "dtype": "category",
2025-09-27 15:09:34.747 |         "samples": ["Electronics", "Clothing"],
2025-09-27 15:09:34.747 |         "num_unique_values": 6,
2025-09-27 15:09:34.747 |         "semantic_type": "",
2025-09-27 15:09:34.747 |         "description": ""
2025-09-27 15:09:34.747 |       }
2025-09-27 15:09:34.747 |     },
2025-09-27 15:09:34.747 |     {
2025-09-27 15:09:34.747 |       "column": "Segment",
2025-09-27 15:09:34.747 |       "properties": {
2025-09-27 15:09:34.747 |         "dtype": "category",
2025-09-27 15:09:34.747 |         "samples": ["Up Market", "Mid Market"],
2025-09-27 15:09:34.747 |         "num_unique_values": 3,
2025-09-27 15:09:34.747 |         "semantic_type": "",
2025-09-27 15:09:34.747 |         "description": ""
2025-09-27 15:09:34.747 |       }
2025-09-27 15:09:34.747 |     },
2025-09-27 15:09:34.747 |     {
2025-09-27 15:09:34.747 |       "column": "Sales_Count",
2025-09-27 15:09:34.747 |       "properties": {
2025-09-27 15:09:34.747 |         "dtype": "number",
2025-09-27 15:09:34.747 |         "std": 51,
2025-09-27 15:09:34.747 |         "min": 13,
2025-09-27 15:09:34.747 |         "max": 197,
2025-09-27 15:09:34.747 |         "samples": [17, 181],
2025-09-27 15:09:34.747 |         "num_unique_values": 78,
2025-09-27 15:09:34.747 |         "semantic_type": "",
2025-09-27 15:09:34.747 |         "description": ""
2025-09-27 15:09:34.747 |       }
2025-09-27 15:09:34.747 |     },
2025-09-27 15:09:34.747 |     {
2025-09-27 15:09:34.747 |       "column": "Sales_Amount",
2025-09-27 15:09:34.747 |       "properties": {
2025-09-27 15:09:34.747 |         "dtype": "number",
2025-09-27 15:09:34.747 |         "std": 14396.958068756981,
2025-09-27 15:09:34.747 |         "min": 1479.73,
2025-09-27 15:09:34.747 |         "max": 49023.57,
2025-09-27 15:09:34.747 |         "samples": [3096.54, 30281.98],
2025-09-27 15:09:34.747 |         "num_unique_values": 100,
2025-09-27 15:09:34.747 |         "semantic_type": "",
2025-09-27 15:09:34.747 |         "description": ""
2025-09-27 15:09:34.747 |       }
2025-09-27 15:09:34.747 |     }
2025-09-27 15:09:34.747 |   ]
2025-09-27 15:09:34.747 | }
2025-09-27 15:09:34.747 | INFO:     172.18.0.1:39926 - "POST /api/summarize HTTP/1.1" 200 OK
2025-09-27 15:09:34.749 | INFO:     172.18.0.1:60386 - "GET /api/docs HTTP/1.1" 200 OK
2025-09-27 15:09:34.749 | INFO:     172.18.0.1:39930 - "GET /api/models HTTP/1.1" 200 OK
2025-09-27 15:09:34.888 | INFO:     172.18.0.1:39930 - "GET /api/openapi.json HTTP/1.1" 200 OK
2025-09-27 15:09:38.587 | INFO:     172.18.0.1:39930 - "GET /api/models HTTP/1.1" 200 OK