pygpt-net 2.4.47__py3-none-any.whl → 2.4.48__py3-none-any.whl

pygpt_net/provider/llms/google.py

@@ -6,11 +6,15 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.12.14 22:00:00                  #
+# Updated Date: 2025.01.16 01:00:00                  #
 # ================================================== #
 
+from typing import Optional, List, Dict
+
 from llama_index.llms.gemini import Gemini
 from llama_index.core.llms.llm import BaseLLM as LlamaBaseLLM
+from llama_index.core.base.embeddings.base import BaseEmbedding
+from llama_index.embeddings.gemini import GeminiEmbedding
 
 from pygpt_net.core.types import (
     MODE_LLAMA_INDEX,
@@ -30,7 +34,7 @@ class GoogleLLM(BaseLLM):
             - api_key: API key for Google API
         """
         self.id = "google"
-        self.type = [MODE_LLAMA_INDEX]
+        self.type = [MODE_LLAMA_INDEX, "embeddings"]
 
     def llama(
             self,
@@ -48,3 +52,22 @@ class GoogleLLM(BaseLLM):
         """
         args = self.parse_args(model.llama_index)
         return Gemini(**args)
+
+    def get_embeddings_model(
+            self,
+            window,
+            config: Optional[List[Dict]] = None
+    ) -> BaseEmbedding:
+        """
+        Return provider instance for embeddings
+
+        :param window: window instance
+        :param config: config keyword arguments list
+        :return: Embedding provider instance
+        """
+        args = {}
+        if config is not None:
+            args = self.parse_args({
+                "args": config,
+            })
+        return GeminiEmbedding(**args)

pygpt_net/provider/llms/openai.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.12.14 22:00:00                  #
+# Updated Date: 2025.01.16 01:00:00                  #
 # ================================================== #
 
 from typing import Optional, List, Dict

pygpt_net/provider/loaders/hub/github/issues.py

@@ -89,7 +89,7 @@ class GitHubRepositoryIssuesReader(BaseGitHubRepositoryIssuesReader):
                     doc_id=str(issue["number"]),
                     text=f"{title}\n{body}",
                 )
-                extra_info = {
+                metadata = {
                     "state": issue["state"],
                     "created_at": issue["created_at"],
                     # url is the API URL
@@ -98,12 +98,12 @@ class GitHubRepositoryIssuesReader(BaseGitHubRepositoryIssuesReader):
                     "source": issue["html_url"],
                 }
                 if issue["closed_at"] is not None:
-                    extra_info["closed_at"] = issue["closed_at"]
+                    metadata["closed_at"] = issue["closed_at"]
                 if issue["assignee"] is not None:
-                    extra_info["assignee"] = issue["assignee"]["login"]
+                    metadata["assignee"] = issue["assignee"]["login"]
                 if issue["labels"] is not None:
-                    extra_info["labels"] = ",".join([label["name"] for label in issue["labels"]])  # fix for labels
-                document.extra_info = extra_info
+                    metadata["labels"] = ",".join([label["name"] for label in issue["labels"]])  # fix for labels
+                document.metadata = metadata
                 documents.append(document)
 
             print_if_verbose(self._verbose, f"Resulted in {len(documents)} documents")

pygpt_net/provider/loaders/hub/json/base.py

@@ -68,11 +68,11 @@ class JSONReader(BaseReader):
             useful_lines = [
                 line for line in lines if not re.match(r"^[{}\\[\\],]*$", line)
             ]
-            return Document(text="\n".join(useful_lines), extra_info=extra_info or {})
+            return Document(text="\n".join(useful_lines), metadata=extra_info or {})
 
         else:
             lines = [*_depth_first_yield(json_data_object, self.levels_back, [])]
-            return Document(text="\n".join(lines), extra_info=extra_info or {})
+            return Document(text="\n".join(lines), metadata=extra_info or {})
 
     def load_data(
         self,

pygpt_net/provider/loaders/hub/pandas_excel/base.py

@@ -86,7 +86,7 @@ class PandasExcelReader(BaseReader):
                 documents.append(
                     Document(
                         text=text,
-                        extra_info=doc_extra_info,
+                        metadata=doc_extra_info,
                     )
                 )
             return documents
@@ -107,6 +107,6 @@ class PandasExcelReader(BaseReader):
             return [
                 Document(
                     text=text,
-                    extra_info=extra_info or {},
+                    metadata=extra_info or {},
                 )
             ]

pygpt_net/provider/loaders/hub/simple_csv/base.py

@@ -46,8 +46,8 @@ class SimpleCSVReader(BaseReader):
             for row in csv_reader:
                 text_list.append(", ".join(row))
         if self._concat_rows:
-            return [Document(text="\n".join(text_list), extra_info=extra_info or {})]
+            return [Document(text="\n".join(text_list), metadata=extra_info or {})]
         else:
             return [
-                Document(text=text, extra_info=extra_info or {}) for text in text_list
+                Document(text=text, metadata=extra_info or {}) for text in text_list
             ]

pygpt_net/provider/loaders/hub/yt/base.py

@@ -59,7 +59,7 @@ class YoutubeTranscriptReader(BasePydanticReader):
             transcript = "\n".join(chunk_text)
             results.append(
                 Document(
-                    text=transcript, id_=video_id, extra_info={"video_id": video_id}
+                    text=transcript, id_=video_id, metadata={"video_id": video_id}
                 )
             )
         return results

pygpt_net/provider/vector_stores/__init__.py

@@ -6,13 +6,12 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.12.14 22:00:00                  #
+# Updated Date: 2025.01.16 01:00:00                  #
 # ================================================== #
 
 import hashlib
 from typing import Optional, Tuple, List
 
-from llama_index.core.indices.service_context import ServiceContext
 from llama_index.core.indices.base import BaseIndex
 from llama_index.core.indices.vector_store.base import VectorStoreIndex
 
@@ -110,13 +109,15 @@ class Storage:
     def get(
             self,
             id: str,
-            service_context: Optional[ServiceContext]
+            llm: Optional = None,
+            embed_model: Optional = None,
     ) -> BaseIndex:
         """
         Get index instance
 
         :param id: index name
-        :param service_context: service context
+        :param llm: LLM instance
+        :param embed_model: Embedding model instance
         :return: index instance
         """
         storage = self.get_storage()
@@ -124,7 +125,8 @@ class Storage:
             raise Exception('Storage engine not found!')
         return storage.get(
             id=id,
-            service_context=service_context,
+            llm=llm,
+            embed_model=embed_model,
         )
 
     def store(
@@ -189,13 +191,15 @@ class Storage:
     def get_tmp(
             self,
             identifier: str,
-            service_context: Optional[ServiceContext] = None
+            llm: Optional = None,
+            embed_model: Optional = None,
     ) -> Tuple[str, BaseIndex]:
         """
         Get tmp index instance
 
         :param identifier: identifier
-        :param service_context: service context
+        :param llm: LLM instance
+        :param embed_model: Embedding model instance
         :return: index instance
         """
         # convert path to md5 hash
@@ -205,7 +209,8 @@ class Storage:
             raise Exception('Storage engine not found!')
         return id, storage.get(
             id=id,
-            service_context=service_context,
+            llm=llm,
+            embed_model=embed_model,
         )
 
     def store_tmp(
@@ -252,13 +257,15 @@ class Storage:
     def get_ctx_idx(
             self,
             path: str,
-            service_context: Optional[ServiceContext] = None
+            llm: Optional = None,
+            embed_model: Optional = None,
     ) -> BaseIndex:
         """
         Get context index instance
 
         :param path: path to index directory
-        :param service_context: service context
+        :param llm: LLM instance
+        :param embed_model: Embedding model instance
         :return: index instance
         """
         # convert path to md5 hash
@@ -267,7 +274,8 @@ class Storage:
             raise Exception('Storage engine not found!')
         return storage.get(
             id="",
-            service_context=service_context,
+            llm=llm,
+            embed_model=embed_model,
         )
 
     def store_ctx_idx(

pygpt_net/provider/vector_stores/base.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.12.14 22:00:00                  #
+# Updated Date: 2025.01.16 01:00:00                  #
 # ================================================== #
 
 import os
@@ -14,7 +14,6 @@ import shutil
 from typing import Optional
 
 from llama_index.core.indices.base import BaseIndex
-from llama_index.core.indices.service_context import ServiceContext
 from llama_index.core import StorageContext
 from llama_index.core.indices.vector_store.base import VectorStoreIndex
 
@@ -36,20 +35,23 @@ class BaseStore:
             self,
             vector_store,
             storage_context: StorageContext,
-            service_context: Optional[ServiceContext] = None
+            llm: Optional = None,
+            embed_model: Optional = None,
     ):
         """
         Get index instance
 
         :param vector_store: vector store instance
         :param storage_context: StorageContext instance
-        :param service_context: ServiceContext instance
+        :param llm: LLM instance
+        :param embed_model: Embedding model instance
         :return: index instance
         """
         return VectorStoreIndex.from_vector_store(
             vector_store,
             storage_context=storage_context,
-            service_context=service_context,
+            llm=llm,
+            embed_model=embed_model,
         )
 
     def index_from_empty(self):
@@ -106,13 +108,15 @@ class BaseStore:
     def get(
             self,
             id: str,
-            service_context: Optional[ServiceContext] = None
+            llm: Optional = None,
+            embed_model: Optional = None,
     ) -> BaseIndex:
         """
         Get index instance
 
         :param id: index name
-        :param service_context: Service context
+        :param llm: LLM instance
+        :param embed_model: Embedding model instance
         :return: index instance
         """
         pass

pygpt_net/provider/vector_stores/chroma.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.12.14 22:00:00                  #
+# Updated Date: 2025.01.16 01:00:00                  #
 # ================================================== #
 
 import os.path
@@ -16,7 +16,6 @@ import chromadb
 from chromadb.config import Settings
 
 from llama_index.core.indices.base import BaseIndex
-from llama_index.core.indices.service_context import ServiceContext
 from llama_index.core import StorageContext
 from llama_index.vector_stores.chroma import ChromaVectorStore
 
@@ -69,13 +68,15 @@ class ChromaProvider(BaseStore):
     def get(
             self,
             id: str,
-            service_context: Optional[ServiceContext] = None
+            llm: Optional = None,
+            embed_model: Optional = None,
     ) -> BaseIndex:
         """
         Get index
 
         :param id: index name
-        :param service_context: Service context
+        :param llm: LLM instance
+        :param embed_model: Embedding model instance
         :return: index instance
         """
         if not self.exists(id):
@@ -89,7 +90,12 @@ class ChromaProvider(BaseStore):
         storage_context = StorageContext.from_defaults(
             persist_dir=path,
         )
-        self.indexes[id] = self.index_from_store(vector_store, storage_context, service_context)
+        self.indexes[id] = self.index_from_store(
+            vector_store=vector_store,
+            storage_context=storage_context,
+            llm=llm,
+            embed_model=embed_model,
+        )
         return self.indexes[id]
 
     def store(

pygpt_net/provider/vector_stores/ctx_attachment.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.12.14 22:00:00                  #
+# Updated Date: 2025.01.16 01:00:00                  #
 # ================================================== #
 
 import os.path
@@ -14,7 +14,6 @@ from typing import Optional
 
 from llama_index.core import StorageContext, load_index_from_storage
 from llama_index.core.indices.base import BaseIndex
-from llama_index.core.indices.service_context import ServiceContext
 
 from .base import BaseStore
 
@@ -76,13 +75,15 @@ class CtxAttachmentProvider(BaseStore):
     def get(
             self,
             id: str,
-            service_context: Optional[ServiceContext] = None
+            llm: Optional = None,
+            embed_model: Optional = None,
     ) -> BaseIndex:
         """
         Get index
 
         :param id: index name (not used)
-        :param service_context: Service context
+        :param llm: LLM instance
+        :param embed_model: Embedding model instance
         :return: index instance
         """
         if not self.exists():
@@ -93,7 +94,8 @@ class CtxAttachmentProvider(BaseStore):
         )
         self.index = load_index_from_storage(
             storage_context,
-            service_context=service_context,
+            llm=llm,
+            embed_model=embed_model,
         )
 
         return self.index

pygpt_net/provider/vector_stores/elasticsearch.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.12.14 22:00:00                  #
+# Updated Date: 2025.01.16 01:00:00                  #
 # ================================================== #
 
 import datetime
@@ -14,7 +14,6 @@ import os.path
 from typing import Optional
 
 from llama_index.core.indices.base import BaseIndex
-from llama_index.core.indices.service_context import ServiceContext
 from llama_index.core import StorageContext
 from llama_index.vector_stores.elasticsearch import ElasticsearchStore
 
@@ -71,13 +70,15 @@ class ElasticsearchProvider(BaseStore):
     def get(
             self,
             id: str,
-            service_context: Optional[ServiceContext] = None
+            llm: Optional = None,
+            embed_model: Optional = None,
     ) -> BaseIndex:
         """
         Get index
 
         :param id: index name
-        :param service_context: service context
+        :param llm: LLM instance
+        :param embed_model: Embedding model instance
         :return: index instance
         """
         if not self.exists(id):
@@ -86,7 +87,12 @@ class ElasticsearchProvider(BaseStore):
         storage_context = StorageContext.from_defaults(
             vector_store=vector_store,
         )
-        self.indexes[id] = self.index_from_store(vector_store, storage_context, service_context)
+        self.indexes[id] = self.index_from_store(
+            vector_store=vector_store,
+            storage_context=storage_context,
+            llm=llm,
+            embed_model=embed_model,
+        )
         return self.indexes[id]
 
     def store(

pygpt_net/provider/vector_stores/pinecode.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.12.14 22:00:00                  #
+# Updated Date: 2025.01.16 01:00:00                  #
 # ================================================== #
 
 import datetime
@@ -17,7 +17,6 @@ from pinecone import Pinecone, ServerlessSpec
 
 from llama_index.core import StorageContext
 from llama_index.core.indices.base import BaseIndex
-from llama_index.core.indices.service_context import ServiceContext
 from llama_index.vector_stores.pinecone import PineconeVectorStore
 
 from pygpt_net.utils import parse_args
@@ -126,13 +125,15 @@ class PinecodeProvider(BaseStore):
     def get(
             self,
             id: str,
-            service_context: Optional[ServiceContext] = None
+            llm: Optional = None,
+            embed_model: Optional = None,
     ) -> BaseIndex:
         """
         Get index
 
         :param id: index name
-        :param service_context: service context
+        :param llm: LLM instance
+        :param embed_model: Embedding model instance
         :return: index instance
         """
         if not self.exists(id):
@@ -141,7 +142,12 @@ class PinecodeProvider(BaseStore):
         storage_context = StorageContext.from_defaults(
             vector_store=vector_store,
         )
-        self.indexes[id] = self.index_from_store(vector_store, storage_context, service_context)
+        self.indexes[id] = self.index_from_store(
+            vector_store=vector_store,
+            storage_context=storage_context,
+            llm=llm,
+            embed_model=embed_model,
+        )
         return self.indexes[id]
 
     def store(

pygpt_net/provider/vector_stores/redis.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.12.14 22:00:00                  #
+# Updated Date: 2025.01.16 01:00:00                  #
 # ================================================== #
 
 import datetime
@@ -15,7 +15,6 @@ from typing import Optional
 
 from llama_index.core import StorageContext
 from llama_index.core.indices.base import BaseIndex
-from llama_index.core.indices.service_context import ServiceContext
 from llama_index.vector_stores.redis import RedisVectorStore
 
 from pygpt_net.utils import parse_args
@@ -71,13 +70,15 @@ class RedisProvider(BaseStore):
     def get(
             self,
             id: str,
-            service_context: Optional[ServiceContext] = None
+            llm: Optional = None,
+            embed_model: Optional = None,
     ) -> BaseIndex:
         """
         Get index
 
         :param id: index name
-        :param service_context: service context
+        :param llm: LLM instance
+        :param embed_model: Embedding model instance
         :return: index instance
         """
         if not self.exists(id):
@@ -86,7 +87,12 @@ class RedisProvider(BaseStore):
         storage_context = StorageContext.from_defaults(
             vector_store=vector_store,
         )
-        self.indexes[id] = self.index_from_store(vector_store, storage_context, service_context)
+        self.indexes[id] = self.index_from_store(
+            vector_store=vector_store,
+            storage_context=storage_context,
+            llm=llm,
+            embed_model=embed_model,
+        )
         return self.indexes[id]
 
     def store(

pygpt_net/provider/vector_stores/simple.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.12.14 22:00:00                  #
+# Updated Date: 2025.01.16 01:00:00                  #
 # ================================================== #
 
 import os.path
@@ -14,7 +14,6 @@ from typing import Optional
 
 from llama_index.core import StorageContext, load_index_from_storage
 from llama_index.core.indices.base import BaseIndex
-from llama_index.core.indices.service_context import ServiceContext
 
 from .base import BaseStore
 
@@ -50,13 +49,15 @@ class SimpleProvider(BaseStore):
     def get(
             self,
             id: str,
-            service_context: Optional[ServiceContext] = None
+            llm: Optional = None,
+            embed_model: Optional = None,
     ) -> BaseIndex:
         """
         Get index
 
         :param id: index name
-        :param service_context: Service context
+        :param llm: LLM instance
+        :param embed_model: Embedding model instance
         :return: index instance
         """
         if not self.exists(id):
@@ -67,7 +68,8 @@ class SimpleProvider(BaseStore):
         )
         self.indexes[id] = load_index_from_storage(
             storage_context,
-            service_context=service_context,
+            llm=llm,
+            embed_model=embed_model,
         )
         return self.indexes[id]
 

pygpt_net/provider/vector_stores/temp.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.12.14 22:00:00                  #
+# Updated Date: 2025.01.16 01:00:00                  #
 # ================================================== #
 
 import os.path
@@ -14,7 +14,6 @@ from typing import Optional
 
 from llama_index.core import StorageContext, load_index_from_storage
 from llama_index.core.indices.base import BaseIndex
-from llama_index.core.indices.service_context import ServiceContext
 
 from .base import BaseStore
 
@@ -107,13 +106,15 @@ class TempProvider(BaseStore):
     def get(
             self,
             id: str,
-            service_context: Optional[ServiceContext] = None
+            llm: Optional = None,
+            embed_model: Optional = None,
     ) -> BaseIndex:
         """
         Get index
 
         :param id: tmp idx id
-        :param service_context: Service context
+        :param llm: LLM instance
+        :param embed_model: Embedding model instance
         :return: index instance
         """
         if not self.exists(id):
@@ -126,7 +127,8 @@ class TempProvider(BaseStore):
             )
             self.indexes[id] = load_index_from_storage(
                 storage_context,
-                service_context=service_context,
+                llm=llm,
+                embed_model=embed_model,
             )
 
         return self.indexes[id]

pygpt_net/ui/layout/chat/output.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.12.14 19:00:00                  #
+# Updated Date: 2025.01.16 01:00:00                  #
 # ================================================== #
 
 from PySide6.QtCore import Qt
@@ -160,7 +160,6 @@ class Output:
         self.window.ui.nodes['inline.vision'] = HelpLabel(trans('inline.vision'))
         self.window.ui.nodes['inline.vision'].setVisible(False)
         self.window.ui.nodes['inline.vision'].setContentsMargins(0, 0, 0, 0)
-        self.window.ui.nodes['inline.vision'].setToolTip(trans('vision.checkbox.tooltip'))
 
         opts_layout = QHBoxLayout()
         # opts_layout.setSpacing(2)  #