pembot 0.0.9__py2.py3-none-any.whl → 0.1.0__py2.py3-none-any.whl

pembot/TextEmbedder/mongodb_embedder.py

@@ -69,22 +69,27 @@ def search_within_document(
     if document_belongs_to_a_type:
         project_dict['type']= 1
 
+    vectorSearchParams= {
+        'queryVector': aggregate_query_embedding,
+        'path': 'embedding',
+
+        #number of nearest neighbors to consider
+        'numCandidates': 100,
+        'limit': limit,
+        'index': index_name,
+
+    }
+
+    #filter (if a type or docid filter is given) to search only within the specified documents search space
+    if document_name_id:
+        vectorSearchParams['filter']= (
+            { "type": {"$in": [document_belongs_to_a_type ]} } if document_belongs_to_a_type else
+            { 'docId': document_name_id }
+        )
+
     pipeline = [
         {
-            '$vectorSearch': {
-                'queryVector': aggregate_query_embedding,
-                'path': 'embedding',
-
-                #number of nearest neighbors to consider
-                'numCandidates': 100,
-                'limit': limit,
-                'index': index_name,
-
-                #filter to search only within the specified document
-                'filter':
-                    { "type": {"$in": [document_belongs_to_a_type ]} } if document_belongs_to_a_type else
-                    { 'docId': document_name_id }
-            }
+            '$vectorSearch': vectorSearchParams
         },
 
         # to exclude the MongoDB internal _id
@@ -93,6 +98,8 @@ def search_within_document(
         }
     ]
 
+
+
     # print("sesraching now:")
     results = list(embeddings_collection.aggregate(pipeline))
     # print("search results: ", results)

pembot/__init__.py

@@ -1,6 +1,6 @@
 """
 A Python Package to convert PEM blog content to usseful information by leveraging LLMs
 """
-__version__ = '0.0.9'
+__version__ = '0.1.0'
 from .main import save_to_json_file, make_query
 __all__ = ["save_to_json_file", "make_query"]

pembot/config/config.yaml

@@ -2,4 +2,4 @@ OUTPUT_DIR: /home/cyto/dev/pembotdir
 PAGE_DELIMITER: ___________________________ NEXT PAGE ___________________________
 app:
   name: pembot
-  version: 0.0.9
+  version: 0.1.0

pembot/query.py

@@ -82,7 +82,13 @@ def multi_embedding_average(llm_client, inference_client, descriptions, model= "
 
 
 
-def rag_query_llm(db_client, llm_client, inference_client, user_query: str, document_id: str, required_fields_descriptions: list[str], model_name: str = "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B", ollama_base_url: str = "http://localhost:11434", no_of_fields= 4, embedding_model= "BAAI/bge-en-icl", llm_provider_name: PROVIDER_T= "novita", index_name: str= "test_search", embeddings_collection= "doc_chunks", document_belongs_to_a_type= ""):
+def rag_query_llm(db_client, llm_client, inference_client,
+    user_query: str, document_id: str, required_fields_descriptions: list[str],
+    model_name: str = "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
+    ollama_base_url: str = "http://localhost:11434", no_of_fields= 4,
+    embedding_model= "BAAI/bge-en-icl", llm_provider_name: PROVIDER_T= "novita",
+    index_name: str= "test_search", embeddings_collection= "doc_chunks",
+    document_belongs_to_a_type= "", prompt_prefix= ""):
     """
     Performs a RAG (Retrieval Augmented Generation) query using a Hugging Face
     embedding model, ChromaDB for retrieval, and a local Ollama model for generation.
@@ -139,6 +145,7 @@ def rag_query_llm(db_client, llm_client, inference_client, user_query: str, docu
     If you don't know the answer, just say that you don't know, don't try to make up an answer.
 
     Context:
+    {prompt_prefix}
     {context}
 
     Question: {user_query}

{pembot-0.0.9.dist-info → pembot-0.1.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pembot
-Version: 0.0.9
+Version: 0.1.0
 Summary: A Python Package to convert PEM blog content to usseful information by leveraging LLMs
 Author-email: cyto <aryan_sidhwani@protonmail.com>
 License-Expression: MIT

{pembot-0.0.9.dist-info → pembot-0.1.0.dist-info}/RECORD

@@ -1,11 +1,11 @@
 pembot/.gitignore,sha256=_7FTsZokJ_pzEyyPjOsGw5x5Xx3gUBFaafs7UlPsv9E,98
 pembot/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
-pembot/__init__.py,sha256=1DmlCS-ZGr0pEr9CQ6lxoYtHSpRMghjac-_MbV29DhU,211
+pembot/__init__.py,sha256=oq-JydA8x2prvUKvfrjXWp-GeFLh_qZPesfmB4ad6HE,211
 pembot/gartner.py,sha256=3ALknQ5mSXIimmwCa3JFDzB_EW2hHEcQO1T2odyBquk,5408
 pembot/main.py,sha256=lZLIV8XPonvNoY4LVS-5fct1y9URMXWoSGJUKMw3Yg8,9667
 pembot/output_structure_local.py,sha256=YfpHzfTNeLMSsB_CjAamha9D6Iz7E1IC-tW9xPCMWFc,3000
 pembot/pem.py,sha256=mv6iGcN1peSY7z2dtCQ_BKj31EFBNfczBhps_d-0XDo,6377
-pembot/query.py,sha256=d6K2PyDDGoIOqwn7A_KIBr83w0zjMAHjhmx1S9VlVgg,8642
+pembot/query.py,sha256=0RBnLDBz8pctbsw9GA2UHG-uZPippKYjBBfBBKe2Oec,8705
 pembot/requirements.txt,sha256=vnjZ3uO4Dd8qPFDq1F2qGromJ5jrEBfBeiDwz7rKijk,1360
 pembot/.git/COMMIT_EDITMSG,sha256=iRLy0abINakd4FBRWJuD-QQ_WWbXZusnr_9e3wo4d90,38
 pembot/.git/HEAD,sha256=KNJb-Cr0wOK3L1CVmyvrhZ4-YLljCl6MYD2tTdsrboA,21
@@ -124,10 +124,10 @@ pembot/AnyToText/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,
 pembot/AnyToText/convertor.py,sha256=gqvhwFssUsAeirfO4n0Ztwga1hn8zHbdG96sMTjYrpE,17188
 pembot/TextEmbedder/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 pembot/TextEmbedder/gemini_embedder.py,sha256=P679-2mmQESlYKML1vcrwx_-CSgWJgIQk7NL4F7BLQE,677
-pembot/TextEmbedder/mongodb_embedder.py,sha256=n-xQwCxSCpGC9jvxHoCVWhBhtm4fVmeg_Gvcn0FDJQ4,10699
+pembot/TextEmbedder/mongodb_embedder.py,sha256=i5M56jaC2t3yUinqlXfC70GaTos4kSR_1LGsbljpEhU,10762
 pembot/TextEmbedder/mongodb_index_creator.py,sha256=kopqdVYJii_wExVrXGZjMfqWZ2dD42b3PeNWo71weHI,5354
 pembot/TextEmbedder/vector_query.py,sha256=Kh1uhx9CatB-oQlQtnW-1I2Qz7MGHI20n2h_8peAChM,1986
-pembot/config/config.yaml,sha256=9Ysd58bptzE9_od_DQQj1ebzrYsyLrOrlzN52dTA3zQ,156
+pembot/config/config.yaml,sha256=-6niASpCQzNynkHSTHWi3MawUWUHpOVuNg0Fhmun30M,156
 pembot/pdf2markdown/LICENSE,sha256=1JTJhQjUYDqJzFJhNtitm7mHyE71PRHgetIqRRWg6Pk,1068
 pembot/pdf2markdown/README.md,sha256=jitM1pwI69oa0N4mXv5-SY1ka9Sz3jsRNCDdpW-50kY,4545
 pembot/pdf2markdown/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -183,7 +183,7 @@ pembot/pdf2markdown/config/config.yaml,sha256=w75W2Eg4-tu8rRk_23PqxWDh0010kRKLmP
 pembot/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 pembot/utils/inference_client.py,sha256=jeURmY2P5heVlH1dCV0XSgiX3U2qYGEmrnUv0KFpdww,5380
 pembot/utils/string_tools.py,sha256=gtRa5rBR0Q7GspTu2WtCnvhJQLFjPfWLvhmyiPkyStU,1883
-pembot-0.0.9.dist-info/licenses/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
-pembot-0.0.9.dist-info/WHEEL,sha256=Dyt6SBfaasWElUrURkknVFAZDHSTwxg3PaTza7RSbkY,100
-pembot-0.0.9.dist-info/METADATA,sha256=Ba3uu0dcTllzXwQqeTwRIJx3--FKl_eXCCorQAPp2pU,313
-pembot-0.0.9.dist-info/RECORD,,
+pembot-0.1.0.dist-info/licenses/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
+pembot-0.1.0.dist-info/WHEEL,sha256=Dyt6SBfaasWElUrURkknVFAZDHSTwxg3PaTza7RSbkY,100
+pembot-0.1.0.dist-info/METADATA,sha256=3tLhZEY9302ZgIQ2itAMbcM354jWyEcZ_Z3h2FeD1-8,313
+pembot-0.1.0.dist-info/RECORD,,