Merge branch 'main' into pgvector-test

run-llama · Dec 9, 2024 · 4f1f454 · 4f1f454
2 parents 3094eae + 4032456
commit 4f1f454
Show file tree

Hide file tree

Showing 123 changed files with 4,016 additions and 2,189 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,5 +1,105 @@
 # ChangeLog
 
+## [2024-12-08]
+
+### `llama-index-core` [0.12.4]
+
+- Fix sync and async structured streaming (#17194)
+- unpin pydantic to allow 2.8 or greater (#17193)
+- Update core structured predict streaming, add ollama structured predict (#17188)
+- bump tenacity dependency in llama-index-core (#17178)
+
+### `llama-index-indices-managed-vectara` [0.3.1]
+
+- Add Verbose to Vectara `as_query_engine` (#17176)
+
+### `llama-index-llms-ollama` [0.5.0]
+
+- Update core structured predict streaming, add ollama structured predict (#17188)
+
+### `llama-index-llms-perplexity` [0.3.2]
+
+- Fix message format for perplexity (#17182)
+
+### `llama-index-readers-web` [0.3.1]
+
+- Add possibility to use URI as doc id in WholeSiteReader (#17187)
+
+### `llama-index-vector-stores-chroma` [0.4.1]
+
+- BUG FIX: llama-index-vectorstore-chromadb to work with chromadb v0.5.17 (#17184)
+
+## [2024-12-06]
+
+### `llama-index-core` [0.12.3]
+
+- cover SimpleDirectoryReader with unit tests (#17156)
+- docs: rewrite openai image reasoning example without multimodal LLM (#17148)
+- fix(metrics): fixed NDCG calculation and added comprehensive test cases (#17126)
+- feat: improve ImageBlock (#17111)
+- Remove forgotten print in ChatMemoryBuffer (#17114)
+- [FIX] Move JSONalyzeQueryEngine to experimental (#17110)
+
+### `llama-index-embeddings-clip` [0.3.1]
+
+- Unrestrict clip models to use (#17162)
+
+### `llama-index-embeddings-openai` [0.3.1]
+
+- fix/openai-embbeding-retry (#17072)
+
+### `llama-index-embeddings-text-embeddings-inference` [0.3.1]
+
+- proper auth token in TEI (#17158)
+
+### `llama-index-indices-managed-llama-cloud` [0.6.3]
+
+- chore: fix httpx_client typo in LlamaCloudRetriever (#17101)
+- fix: wrong project id variable in LlamaCloudRetriever (#17086)
+
+### `llama-index-llms-bedrock-converse` [0.4.1]
+
+- Adding AWS Nova models to Bedrock Converse (#17139)
+
+### `llama-index-llms-ollama` [0.4.2]
+
+- Ollama LLM: Added TypeError exception to `_get_response_token_counts` (#17150)
+
+### `llama-index-llms-sambanovasystems` [0.4.3]
+
+- changes in openai identification in url (#17161)
+
+### `llama-index-memory-mem0` [0.2.1]
+
+- Fix mem0 version check (#17159)
+
+### `llama-index-multi-modal-llms-openai` [0.4.0]
+
+- fix: make OpenAIMultiModal work with new ChatMessage (#17138)
+
+### `llama-index-postprocessor-bedrock-rerank` [0.3.0]
+
+- Add AWS Bedrock Reranker (#17134)
+
+### `llama-index-readers-file` [0.4.1]
+
+- update doc id for unstructured reader (#17160)
+
+### `llama-index-retrievers-duckdb-retriever` [0.4.0]
+
+- fix: use prepared statement in DuckDBRetriever (#17092)
+
+### `llama-index-vector-stores-postgres` [0.3.2]
+
+- Create tables for pgvector regardless of schema status (#17100)
+
+### `llama-index-vector-stores-weaviate` [1.2.4]
+
+- make alpha not none in weaviate (#17163)
+- Make Weaviate Vector Store integration work with complex properties (#17129)
+- Add support for `IS_EMPTY` metadata filters to Weaviate Vector Store integration (#17128)
+- Make Weaviate Vector Store integration support nested metadata filtering (#17107)
+
 ## [2024-11-26]
 
 ### `llama-index-core` [0.12.2]

diff --git a/docs/docs/CHANGELOG.md b/docs/docs/CHANGELOG.md
@@ -1,5 +1,105 @@
 # ChangeLog
 
+## [2024-12-08]
+
+### `llama-index-core` [0.12.4]
+
+- Fix sync and async structured streaming (#17194)
+- unpin pydantic to allow 2.8 or greater (#17193)
+- Update core structured predict streaming, add ollama structured predict (#17188)
+- bump tenacity dependency in llama-index-core (#17178)
+
+### `llama-index-indices-managed-vectara` [0.3.1]
+
+- Add Verbose to Vectara `as_query_engine` (#17176)
+
+### `llama-index-llms-ollama` [0.5.0]
+
+- Update core structured predict streaming, add ollama structured predict (#17188)
+
+### `llama-index-llms-perplexity` [0.3.2]
+
+- Fix message format for perplexity (#17182)
+
+### `llama-index-readers-web` [0.3.1]
+
+- Add possibility to use URI as doc id in WholeSiteReader (#17187)
+
+### `llama-index-vector-stores-chroma` [0.4.1]
+
+- BUG FIX: llama-index-vectorstore-chromadb to work with chromadb v0.5.17 (#17184)
+
+## [2024-12-06]
+
+### `llama-index-core` [0.12.3]
+
+- cover SimpleDirectoryReader with unit tests (#17156)
+- docs: rewrite openai image reasoning example without multimodal LLM (#17148)
+- fix(metrics): fixed NDCG calculation and added comprehensive test cases (#17126)
+- feat: improve ImageBlock (#17111)
+- Remove forgotten print in ChatMemoryBuffer (#17114)
+- [FIX] Move JSONalyzeQueryEngine to experimental (#17110)
+
+### `llama-index-embeddings-clip` [0.3.1]
+
+- Unrestrict clip models to use (#17162)
+
+### `llama-index-embeddings-openai` [0.3.1]
+
+- fix/openai-embbeding-retry (#17072)
+
+### `llama-index-embeddings-text-embeddings-inference` [0.3.1]
+
+- proper auth token in TEI (#17158)
+
+### `llama-index-indices-managed-llama-cloud` [0.6.3]
+
+- chore: fix httpx_client typo in LlamaCloudRetriever (#17101)
+- fix: wrong project id variable in LlamaCloudRetriever (#17086)
+
+### `llama-index-llms-bedrock-converse` [0.4.1]
+
+- Adding AWS Nova models to Bedrock Converse (#17139)
+
+### `llama-index-llms-ollama` [0.4.2]
+
+- Ollama LLM: Added TypeError exception to `_get_response_token_counts` (#17150)
+
+### `llama-index-llms-sambanovasystems` [0.4.3]
+
+- changes in openai identification in url (#17161)
+
+### `llama-index-memory-mem0` [0.2.1]
+
+- Fix mem0 version check (#17159)
+
+### `llama-index-multi-modal-llms-openai` [0.4.0]
+
+- fix: make OpenAIMultiModal work with new ChatMessage (#17138)
+
+### `llama-index-postprocessor-bedrock-rerank` [0.3.0]
+
+- Add AWS Bedrock Reranker (#17134)
+
+### `llama-index-readers-file` [0.4.1]
+
+- update doc id for unstructured reader (#17160)
+
+### `llama-index-retrievers-duckdb-retriever` [0.4.0]
+
+- fix: use prepared statement in DuckDBRetriever (#17092)
+
+### `llama-index-vector-stores-postgres` [0.3.2]
+
+- Create tables for pgvector regardless of schema status (#17100)
+
+### `llama-index-vector-stores-weaviate` [1.2.4]
+
+- make alpha not none in weaviate (#17163)
+- Make Weaviate Vector Store integration work with complex properties (#17129)
+- Add support for `IS_EMPTY` metadata filters to Weaviate Vector Store integration (#17128)
+- Make Weaviate Vector Store integration support nested metadata filtering (#17107)
+
 ## [2024-11-26]
 
 ### `llama-index-core` [0.12.2]

diff --git a/docs/docs/api_reference/callbacks/argilla.md b/docs/docs/api_reference/callbacks/argilla.md
@@ -0,0 +1,4 @@
+::: llama_index.callbacks.argilla
+    options:
+      members:
+        - argilla_callback_handler
diff --git a/docs/docs/api_reference/postprocessor/bedrock_rerank.md b/docs/docs/api_reference/postprocessor/bedrock_rerank.md
@@ -0,0 +1,4 @@
+::: llama_index.postprocessor.bedrock_rerank
+    options:
+      members:
+        - BedrockRerank
diff --git a/docs/docs/examples/llm/bedrock_converse.ipynb b/docs/docs/examples/llm/bedrock_converse.ipynb
@@ -274,7 +274,7 @@
     "from llama_index.llms.bedrock_converse import BedrockConverse\n",
     "\n",
     "llm = BedrockConverse(\n",
-    "    model=\"anthropic.claude-3-haiku-20240307-v1:0\",\n",
+    "    model=\"us.amazon.nova-lite-v1:0\",\n",
     "    aws_access_key_id=\"AWS Access Key ID to use\",\n",
     "    aws_secret_access_key=\"AWS Secret Access Key to use\",\n",
     "    aws_session_token=\"AWS Session Token to use\",\n",

diff --git a/docs/docs/examples/llm/ollama.ipynb b/docs/docs/examples/llm/ollama.ipynb
@@ -349,7 +349,7 @@
    "source": [
     "## Structured Outputs\n",
     "\n",
-    "We can also attach a pyndatic class to the LLM to ensure structured outputs"
+    "We can also attach a pyndatic class to the LLM to ensure structured outputs. This will use Ollama's builtin structured output capabilities for a given pydantic class."
    ]
   },
   {
@@ -360,7 +360,6 @@
    "outputs": [],
    "source": [
     "from llama_index.core.bridge.pydantic import BaseModel\n",
-    "from llama_index.core.tools import FunctionTool\n",
     "\n",
     "\n",
     "class Song(BaseModel):\n",
@@ -392,11 +391,13 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "{\"name\": \"Yesterday\", \"artist\": \"The Beatles\"}\n"
+      "{\"name\":\"Radioactive\",\"artist\":\"Imagine Dragons\"}\n"
      ]
     }
    ],
    "source": [
+    "from llama_index.core.llms import ChatMessage\n",
+    "\n",
     "response = sllm.chat([ChatMessage(role=\"user\", content=\"Name a random song!\")])\n",
     "print(response.message.content)"
    ]
@@ -419,7 +420,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "{\"name\": \"Happy Birthday to You\", \"artist\": \"Traditional\"}\n"
+      "{\"name\":\"Lose Yourself\",\"artist\":\"Eminem\"}\n"
      ]
     }
    ],
@@ -432,10 +433,54 @@
   },
   {
    "cell_type": "markdown",
-   "id": "c4b224c6",
+   "id": "cdad7904",
    "metadata": {},
    "source": [
-    "Currently, Ollama does not support streaming structured objects. But hopefully soon!"
+    "You can also stream structured outputs! Streaming a structured output is a little different than streaming a normal string. It will yield a generator of the most up to date structured object."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "d5c40157",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "{\"name\":null,\"artist\":null}\n",
+      "{\"name\":null,\"artist\":null}\n",
+      "{\"name\":null,\"artist\":null}\n",
+      "{\"name\":null,\"artist\":null}\n",
+      "{\"name\":null,\"artist\":null}\n",
+      "{\"name\":null,\"artist\":null}\n",
+      "{\"name\":null,\"artist\":null}\n",
+      "{\"name\":null,\"artist\":\"\"}\n",
+      "{\"name\":null,\"artist\":\"The\"}\n",
+      "{\"name\":null,\"artist\":\"The Black\"}\n",
+      "{\"name\":null,\"artist\":\"The Black Keys\"}\n",
+      "{\"name\":null,\"artist\":\"The Black Keys\"}\n",
+      "{\"name\":null,\"artist\":\"The Black Keys\"}\n",
+      "{\"name\":null,\"artist\":\"The Black Keys\"}\n",
+      "{\"name\":null,\"artist\":\"The Black Keys\"}\n",
+      "{\"name\":null,\"artist\":\"The Black Keys\"}\n",
+      "{\"name\":\"\",\"artist\":\"The Black Keys\"}\n",
+      "{\"name\":\"Lon\",\"artist\":\"The Black Keys\"}\n",
+      "{\"name\":\"Lonely\",\"artist\":\"The Black Keys\"}\n",
+      "{\"name\":\"Lonely Boy\",\"artist\":\"The Black Keys\"}\n",
+      "{\"name\":\"Lonely Boy\",\"artist\":\"The Black Keys\"}\n",
+      "{\"name\":\"Lonely Boy\",\"artist\":\"The Black Keys\"}\n",
+      "{\"name\":\"Lonely Boy\",\"artist\":\"The Black Keys\"}\n"
+     ]
+    }
+   ],
+   "source": [
+    "response_gen = sllm.stream_chat(\n",
+    "    [ChatMessage(role=\"user\", content=\"Name a random song!\")]\n",
+    ")\n",
+    "for r in response_gen:\n",
+    "    print(r.message.content)"
    ]
   }
  ],