search

HazyResearch / evaporate

This repo contains data and code for the paper "Language Models Enable Simple Systems for Generating Structured Views of Heterogeneous Data Lakes"
475 stars 45 forks source link

when i run bash run.sh, i got a issue"Failed to apply prompt". What should I do? #14

Closed jy-101361-1810897 closed 1 year ago

simran-arora commented 1 year ago

Hi! Thanks for trying the code. Can you include some information from the error message backtrace?

jy-101361-1810897 commented 1 year ago

Thanks.I got 'Failed to apply prompt' like this. Data lake Chunking files: 100%|███████████████████████| 100/100 [00:00<00:00, 1230.00it/s] Data-lake: fda_510ks, Train size: 10 Chunks in sample file ./data/evaporate/fda-ai-pmas/510k/K161714.txt: 8 Chunks in sample file ./data/evaporate/fda-ai-pmas/510k/K172913.txt: 11 Chunks in sample file ./data/evaporate/fda-ai-pmas/510k/K190219.txt: 15 Chunks in sample file ./data/evaporate/fda-ai-pmas/510k/K171641.txt: 10 Chunks in sample file ./data/evaporate/fda-ai-pmas/510k/K163260.txt: 23 Chunks in sample file ./data/evaporate/fda-ai-pmas/510k/K151923.txt: 7 Chunks in sample file ./data/evaporate/fda-ai-pmas/510k/K181288.txt: 9 Chunks in sample file ./data/evaporate/fda-ai-pmas/510k/K173927.txt: 9 Chunks in sample file ./data/evaporate/fda-ai-pmas/510k/K162705.txt: 4 Chunks in sample file ./data/evaporate/fda-ai-pmas/510k/K180886.txt: 5 Directly extracting metadata from chunks: 0%| | 0/10 [00:00<?, ?it/s]Failed to apply prompt to chunk. Failed to apply prompt to chunk. Failed to apply prompt to chunk.

I find it in line 55 in schema_identification.py. I debug to find why it got the issue. For example, I enter apply_prompt function and the next function. And I cannot find the cause of the problem.