[BUG] Running locally with Ollama 3.1 does not do research

[oscmejia]

Describe the bug I'm running 0.2.4 with local Ollama and there is no outline

To Reproduce Report following things I'm calling it from the root directry:

python examples/run_storm_wiki_ollama.py \
    --output-dir test_output \
    --retriever serper \
    --do-research \
    --do-generate-outline \
    --do-generate-article \
    --do-polish-article \
    --url http://localhost \
    --port 11434 \
    --model llama3.1:latest 

Output:

Topic: Dogs
knowledge_storm.interface : INFO     : run_knowledge_curation_module executed in 126.1719 seconds
knowledge_storm.interface : INFO     : run_outline_generation_module executed in 7.0749 seconds
sentence_transformers.SentenceTransformer : INFO     : Use pytorch device_name: cuda
sentence_transformers.SentenceTransformer : INFO     : Load pretrained SentenceTransformer: paraphrase-MiniLM-L6-v2
/home/user1/miniconda3/envs/storm/lib/python3.11/site-packages/transformers/tokenization_utils_base.py:1601: FutureWarning: `clean_up_tokenization_spaces` was not set. It will be set to `True` by default. This behavior will be depracted in transformers v4.45, and will be then set to `False` by default. For more details check this issue: https://github.com/huggingface/transformers/issues/31884
  warnings.warn(
root : ERROR    : No outline for Dogs. Will directly search with the topic.
knowledge_storm.interface : INFO     : run_article_generation_module executed in 5.5393 seconds
knowledge_storm.interface : INFO     : run_article_polishing_module executed in 4.2038 seconds
***** Execution time *****
run_knowledge_curation_module: 126.1719 seconds
run_outline_generation_module: 7.0749 seconds
run_article_generation_module: 5.5393 seconds
run_article_polishing_module: 4.2038 seconds
***** Token usage of language models: *****
run_knowledge_curation_module
run_outline_generation_module
run_article_generation_module
run_article_polishing_module
***** Number of queries of retrieval models: *****
run_knowledge_curation_module: {'SerperRM': 36}
run_outline_generation_module: {'SerperRM': 0}
run_article_generation_module: {'SerperRM': 0}
run_article_polishing_module: {'SerperRM': 0}

The outline is empty

Environment:

• Ubuntu 22LTS
• llama3.1:latest

[Yucheng-Jiang]

Would you please also provide “ All output files generated for this topic as a zip file” as requested in the bug report template? It will greatly help us triage the problem.

[oscmejia]

sure: test_output 2.zip

[Yucheng-Jiang]

@oscmejia

in your provided output log file, we believe this issue is the same as issue #139. Please refer to this solution.