llama_index SPARQL Notes 07
Ok, time to attack the bits listed at the end of yesterday's post.
Break time. Made some progress, but got a bit distracted starting a master plan. Pretty much than same as I'd already put in the GraphRAG doc, but expanded a bit, more background, hopefully a better explanation for folks that occupy other Venn diagrams.
export PYTHONPATH=$PYTHONPATH:/home/danny/AI/LIBS-under-dev/llama_index
Added extra logging to nebulagraph.py, just markers on functions defined in types.py are being called -
logging.basicConfig(filename='loggy.log', filemode='w', level=logging.INFO)
logger = logging.getLogger(__name__)
logger.info('nebulagraph HERE')
...
logger.info('#### nebulagraph client(self) called')
After running python graph-rag-nebulagraph-minimal.py, what was in loggy.log (in the dir I ran from) :
INFO:llama_index.graph_stores.nebulagraph:HERE
INFO:llama_index.indices.loading:Loading all indices.
INFO:llama_index.indices.loading:Loading all indices.
INFO:llama_index.graph_stores.nebulagraph:#### get_schema called
Ok, that's a start. For a bit more coverage, I'll do the same to simple.py and the skeletal sparql.py.
INFO:llama_index.graph_stores.nebulagraph:nebulagraph HERE
INFO:llama_index.graph_stores.simple:simple HERE
INFO:llama_index.indices.loading:Loading all indices.
INFO:llama_index.indices.loading:Loading all indices.
INFO:llama_index.graph_stores.nebulagraph:#### nebulagraph get_schema called