Add configurable retries, make flask optional, usable in Mech (#20)

* Making it usable in Mech * make flask optional * Add date to serper * configurable retries * Update serper_retriever.py
Libr-AI · May 12, 2024 · af71b0f · af71b0f
1 parent f620f6e
commit af71b0f
Show file tree

Hide file tree

Showing 4 changed files with 369 additions and 367 deletions.
diff --git a/factcheck/__init__.py b/factcheck/__init__.py
@@ -29,6 +29,7 @@ def __init__(
         evidence_retrieval_model: str = None,
         claim_verify_model: str = None,
         api_config: dict = None,
+        num_seed_retries: int = 3,
     ):
         self.encoding = tiktoken.get_encoding("cl100k_base")
 
@@ -63,6 +64,7 @@ def __init__(
         self.query_generator = QueryGenerator(llm_client=self.query_generator_model, prompt=self.prompt)
         self.evidence_crawler = retriever_mapper(retriever_name=retriever)(api_config=self.api_config)
         self.claimverify = ClaimVerify(llm_client=self.claim_verify_model, prompt=self.prompt)
+        self.num_seed_retries = num_seed_retries
 
         logger.info("===Sub-modules Init Finished===")
 
@@ -73,15 +75,15 @@ def load_config(self, api_config: dict) -> None:
     def check_response(self, response: str):
         st_time = time.time()
         # step 1
-        claims = self.decomposer.getclaims(doc=response)
+        claims = self.decomposer.getclaims(doc=response, num_retries=self.num_seed_retries)
         for i, claim in enumerate(claims):
             logger.info(f"== response claims {i}: {claim}")
 
         # step 2
         (
             checkworthy_claims,
             pairwise_checkworthy,
-        ) = self.checkworthy.identify_checkworthiness(claims)
+        ) = self.checkworthy.identify_checkworthiness(claims, num_retries=self.num_seed_retries)
         for i, claim in enumerate(checkworthy_claims):
             logger.info(f"== Check-worthy claims {i}: {claim}")
 

diff --git a/factcheck/core/Retriever/serper_retriever.py b/factcheck/core/Retriever/serper_retriever.py
@@ -67,6 +67,7 @@ def _retrieve_evidence_4_all_claim(self, query_list: list[str], top_k: int = 5,
 
         # get the results for queries with an answer box
         query_url_dict = {}
+        url_to_date = {}  # TODO: decide whether to use date
         _snippet_to_check = []
         for i, (query, result) in enumerate(zip(query_list, serper_response.json())):
             if query != result.get("searchParameters").get("q"):
@@ -92,6 +93,8 @@ def _retrieve_evidence_4_all_claim(self, query_list: list[str], top_k: int = 5,
                     "url": "Multiple",
                 }
 
+                # Save date for each url
+                url_to_date.update({result.get("link"): result.get("date") for result in results})
                 # Save query-url pair, 1 query may have multiple urls
                 query_url_dict.update({query: [result.get("link") for result in results]})
                 _snippet_to_check += [result["snippet"] for result in results]