Add the 'hit callback' func (#542)

Signed-off-by: SimFG <[email protected]>

gptcache/adapter/adapter.py

@@ -178,7 +178,10 @@ def adapt(llm_handler, cache_data_convert, update_cache_callback, *args, **kwarg
         cache_answers = sorted(cache_answers, key=lambda x: x[0], reverse=True)
         answers_dict = dict((d[1], d) for d in cache_answers)
         if len(cache_answers) != 0:
-
+            hit_callback = kwargs.pop("hit_callback", None)
+            if hit_callback and callable(hit_callback):
+                factor = max_rank - min_rank
+                hit_callback([(d[3].question, d[0] / factor if factor else d[0]) for d in cache_answers])
             def post_process():
                 if chat_cache.post_process_messages_func is temperature_softmax:
                     return_message = chat_cache.post_process_messages_func(

gptcache/utils/__init__.py

@@ -105,7 +105,7 @@ def import_huggingface_hub():
 
 
 def import_onnxruntime():
-    _check_library("onnxruntime")
+    _check_library("onnxruntime", package="onnxruntime==1.14.1")
 
 
 def import_faiss():

tests/unit_tests/adapter/test_api.py

@@ -42,7 +42,7 @@ def test_gptcache_api():
     put("api-hello2", "foo2", cache_obj=inner_cache)
     put("api-hello3", "foo3", cache_obj=inner_cache)
 
-    messages = get("hello", cache_obj=inner_cache, top_k=3)
+    messages = get("hello", cache_obj=inner_cache, top_k=3, hit_callback=lambda x: print("hit_callback", x))
     assert len(messages) == 3
     assert "foo1" in messages
     assert "foo2" in messages