ValueError
builtins.ValueError
Stack trace
Traceback (most recent call last):
File "app.py", line 42, in <module>
reranked_docs = reranker.rerank(documents, query, top_n=10) # triggers error
File "/usr/local/lib/python3.9/site-packages/reranking_lib/core.py", line 88, in rerank
raise ValueError(f"top_n={top_n} cannot be greater than number of documents={len(documents)}")
ValueError: top_n=10 cannot be greater than number of documents=5 Why it happens
The reranking function expects the top_n parameter to specify how many top documents to return after reranking. If top_n is larger than the number of documents provided, the function raises a ValueError to prevent invalid indexing or empty results.
Detection
Validate the top_n parameter against the length of the documents list before calling rerank, or catch ValueError exceptions and log the mismatch details.
Causes & fixes
top_n parameter is set larger than the number of documents available for reranking
Ensure top_n is less than or equal to the length of the documents list before calling rerank, e.g., top_n = min(top_n, len(documents))
Documents list is empty or smaller than expected due to upstream filtering or data loading issues
Add checks to confirm documents list is populated correctly before reranking and handle empty or small lists gracefully
Hardcoded top_n value without dynamic adjustment based on input size
Make top_n a dynamic parameter that adjusts to the input documents count or add validation logic to prevent invalid values
Code: broken vs fixed
documents = ['doc1', 'doc2', 'doc3', 'doc4', 'doc5']
query = 'example query'
# This line triggers the ValueError because top_n=10 > len(documents)=5
reranked_docs = reranker.rerank(documents, query, top_n=10)documents = ['doc1', 'doc2', 'doc3', 'doc4', 'doc5']
query = 'example query'
# Fix: Adjust top_n to not exceed number of documents
top_n = 10
top_n = min(top_n, len(documents))
reranked_docs = reranker.rerank(documents, query, top_n=top_n)
print('Reranking succeeded with top_n:', top_n)Workaround
Wrap the rerank call in try/except ValueError, and if caught, reduce top_n to the documents count and retry reranking.
Prevention
Always validate or dynamically set top_n based on the input documents length before reranking to avoid invalid parameter errors.