Split HuggingFace embeddings in HuggingFace API and TextGenerationInf…

…erence packages (run-llama#14013)
sansmoraxz · Jun 12, 2024 · 7c512fb · 7c512fb
1 parent 49d5415
commit 7c512fb
Show file tree

Hide file tree

Showing 65 changed files with 2,583 additions and 27 deletions.
diff --git a/docs/docs/api_reference/embeddings/huggingface.md b/docs/docs/api_reference/embeddings/huggingface.md
@@ -2,4 +2,3 @@
     options:
       members:
         - HuggingFaceEmbedding
-        - HuggingFaceInferenceAPIEmbedding
diff --git a/docs/docs/api_reference/embeddings/huggingface_api.md b/docs/docs/api_reference/embeddings/huggingface_api.md
@@ -0,0 +1,4 @@
+::: llama_index.embeddings.huggingface_api
+    options:
+      members:
+        - HuggingFaceInferenceAPIEmbedding
diff --git a/docs/docs/api_reference/llms/huggingface.md b/docs/docs/api_reference/llms/huggingface.md
@@ -1,5 +1,4 @@
 ::: llama_index.llms.huggingface
     options:
       members:
-        - HuggingFaceInferenceAPI
         - HuggingFaceLLM
diff --git a/docs/docs/api_reference/llms/huggingface_api.md b/docs/docs/api_reference/llms/huggingface_api.md
@@ -0,0 +1,4 @@
+::: llama_index.llms.huggingface_api
+    options:
+      members:
+        - HuggingFaceInferenceAPI
diff --git a/docs/docs/api_reference/llms/text_generation_inference.md b/docs/docs/api_reference/llms/text_generation_inference.md
@@ -0,0 +1,4 @@
+::: llama_index.llms.text_generation_inference
+    options:
+      members:
+        - TextGenerationInference
diff --git a/docs/docs/examples/cookbooks/llama3_cookbook.ipynb b/docs/docs/examples/cookbooks/llama3_cookbook.ipynb
@@ -29,7 +29,8 @@
    "source": [
     "!pip install llama-index\n",
     "!pip install llama-index-llms-huggingface\n",
-    "!pip install llama-index-embeddings-huggingface"
+    "!pip install llama-index-embeddings-huggingface\n",
+    "!pip install llama-index-embeddings-huggingface-api"
    ]
   },
   {
@@ -166,7 +167,7 @@
    "source": [
     "## You can deploy the model on HF Inference Endpoint and use it\n",
     "\n",
-    "# from llama_index.llms.huggingface import HuggingFaceInferenceAPI\n",
+    "# from llama_index.llms.huggingface_api import HuggingFaceInferenceAPI\n",
     "\n",
     "# llm = HuggingFaceInferenceAPI(\n",
     "#     model_name=\"<HF Inference Endpoint>\",\n",

diff --git a/docs/docs/examples/cookbooks/prometheus2_cookbook.ipynb b/docs/docs/examples/cookbooks/prometheus2_cookbook.ipynb
@@ -53,7 +53,7 @@
    "outputs": [],
    "source": [
     "!pip install llama-index\n",
-    "!pip install llama-index-llms-huggingface"
+    "!pip install llama-index-llms-huggingface-api"
    ]
   },
   {
@@ -145,7 +145,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from llama_index.llms.huggingface import HuggingFaceInferenceAPI\n",
+    "from llama_index.llms.huggingface_api import HuggingFaceInferenceAPI\n",
     "\n",
     "HF_TOKEN = \"YOUR HF TOKEN\"\n",
     "HF_ENDPOINT_URL = \"YOUR HF ENDPOINT URL\"\n",

diff --git a/docs/docs/examples/embeddings/jina_embeddings.ipynb b/docs/docs/examples/embeddings/jina_embeddings.ipynb
@@ -26,6 +26,7 @@
    "outputs": [],
    "source": [
     "%pip install llama-index-embeddings-huggingface\n",
+    "%pip install llama-index-embeddings-huggingface-api\n",
     "%pip install llama-index-embeddings-openai"
    ]
   },
@@ -58,6 +59,8 @@
    "source": [
     "from llama_index.embeddings.huggingface import (\n",
     "    HuggingFaceEmbedding,\n",
+    ")\n",
+    "from llama_index.embeddings.huggingface_api import (\n",
     "    HuggingFaceInferenceAPIEmbedding,\n",
     ")\n",
     "from llama_index.embeddings.openai import OpenAIEmbedding\n",

diff --git a/docs/docs/examples/evaluation/prometheus_evaluation.ipynb b/docs/docs/examples/evaluation/prometheus_evaluation.ipynb
@@ -76,7 +76,7 @@
    "outputs": [],
    "source": [
     "%pip install llama-index-llms-openai\n",
-    "%pip install llama-index-llms-huggingface"
+    "%pip install llama-index-llms-huggingface-api"
    ]
   },
   {
@@ -146,7 +146,7 @@
     }
    ],
    "source": [
-    "from llama_index.llms.huggingface import HuggingFaceInferenceAPI\n",
+    "from llama_index.llms.huggingface_api import HuggingFaceInferenceAPI\n",
     "\n",
     "HF_TOKEN = \"YOUR HF TOKEN\"\n",
     "HF_ENDPOINT_URL = (\n",

diff --git a/...ples/finetuning/llm_judge/correctness/finetune_llm_judge_single_grading_correctness.ipynb b/...ples/finetuning/llm_judge/correctness/finetune_llm_judge_single_grading_correctness.ipynb
@@ -27,7 +27,7 @@
     "%pip install llama-index-finetuning\n",
     "%pip install llama-index-llms-openai\n",
     "%pip install llama-index-finetuning-callbacks\n",
-    "%pip install llama-index-llms-huggingface"
+    "%pip install llama-index-llms-huggingface-api"
    ]
   },
   {
@@ -265,7 +265,7 @@
    ],
    "source": [
     "from llama_index.core.query_engine import RetrieverQueryEngine\n",
-    "from llama_index.llms.huggingface import HuggingFaceInferenceAPI\n",
+    "from llama_index.llms.huggingface_api import HuggingFaceInferenceAPI\n",
     "\n",
     "llm = HuggingFaceInferenceAPI(\n",
     "    model_name=\"meta-llama/Llama-2-7b-chat-hf\",\n",

diff --git a/docs/docs/examples/finetuning/llm_judge/pairwise/finetune_llm_judge.ipynb b/docs/docs/examples/finetuning/llm_judge/pairwise/finetune_llm_judge.ipynb
@@ -28,7 +28,7 @@
     "%pip install llama-index-finetuning\n",
     "%pip install llama-index-llms-openai\n",
     "%pip install llama-index-finetuning-callbacks\n",
-    "%pip install llama-index-llms-huggingface"
+    "%pip install llama-index-llms-huggingface-api"
    ]
   },
   {
@@ -410,7 +410,7 @@
    "outputs": [],
    "source": [
     "from llama_index.core.query_engine import RetrieverQueryEngine\n",
-    "from llama_index.llms.huggingface import HuggingFaceInferenceAPI\n",
+    "from llama_index.llms.huggingface_api import HuggingFaceInferenceAPI\n",
     "\n",
     "\n",
     "def create_query_engine(\n",

diff --git a/docs/docs/examples/llm/huggingface.ipynb b/docs/docs/examples/llm/huggingface.ipynb
@@ -49,7 +49,8 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "%pip install llama-index-llms-huggingface"
+    "%pip install llama-index-llms-huggingface\n",
+    "%pip install llama-index-llms-huggingface-api"
    ]
   },
   {
@@ -99,10 +100,8 @@
     "import os\n",
     "from typing import List, Optional\n",
     "\n",
-    "from llama_index.llms.huggingface import (\n",
-    "    HuggingFaceInferenceAPI,\n",
-    "    HuggingFaceLLM,\n",
-    ")\n",
+    "from llama_index.llms.huggingface import HuggingFaceLLM\n",
+    "from llama_index.llms.huggingface_api import HuggingFaceInferenceAPI\n",
     "\n",
     "# SEE: https://huggingface.co/docs/hub/security-tokens\n",
     "# We just need a token with read permissions for this demo\n",
@@ -227,6 +226,16 @@
     "The new `TextGenerationInference` class allows to interface with endpoints running [`text-generation-inference`, TGI](https://huggingface.co/docs/text-generation-inference/index). In addition to blazingly fast inference, it supports `tool` usage starting from version `2.0.1`. "
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "46c5c06d",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%pip install llama-index-llms-text-generation-inference"
+   ]
+  },
   {
    "cell_type": "markdown",
    "id": "055ddcb1",
@@ -253,7 +262,7 @@
     "import os\n",
     "from typing import List, Optional\n",
     "\n",
-    "from llama_index.llms.huggingface import (\n",
+    "from llama_index.llms.text_generation_inference import (\n",
     "    TextGenerationInference,\n",
     ")\n",
     "\n",

diff --git a/docs/docs/examples/node_postprocessor/rankGPT.ipynb b/docs/docs/examples/node_postprocessor/rankGPT.ipynb
@@ -32,6 +32,7 @@
    "source": [
     "%pip install llama-index-postprocessor-rankgpt-rerank\n",
     "%pip install llama-index-llms-huggingface\n",
+    "%pip install llama-index-llms-huggingface-api\n",
     "%pip install llama-index-llms-openai\n",
     "%pip install llama-index-llms-ollama"
    ]
@@ -484,10 +485,8 @@
     "from llama_index.core import QueryBundle\n",
     "import pandas as pd\n",
     "from IPython.display import display, HTML\n",
-    "from llama_index.llms.huggingface import (\n",
-    "    HuggingFaceInferenceAPI,\n",
-    "    HuggingFaceLLM,\n",
-    ")\n",
+    "from llama_index.llms.huggingface_api import HuggingFaceInferenceAPI\n",
+    "from llama_index.llms.huggingface import HuggingFaceLLM\n",
     "\n",
     "from llama_index.postprocessor.rankgpt_rerank import RankGPTRerank\n",
     "\n",

diff --git a/docs/mkdocs.yml b/docs/mkdocs.yml
@@ -754,6 +754,7 @@ nav:
           - ./api_reference/embeddings/google.md
           - ./api_reference/embeddings/gradient.md
           - ./api_reference/embeddings/huggingface.md
+          - ./api_reference/embeddings/huggingface_api.md
           - ./api_reference/embeddings/huggingface_itrex.md
           - ./api_reference/embeddings/huggingface_openvino.md
           - ./api_reference/embeddings/huggingface_optimum.md
@@ -839,6 +840,7 @@ nav:
           - ./api_reference/llms/gradient.md
           - ./api_reference/llms/groq.md
           - ./api_reference/llms/huggingface.md
+          - ./api_reference/llms/huggingface_api.md
           - ./api_reference/llms/index.md
           - ./api_reference/llms/ipex_llm.md
           - ./api_reference/llms/konko.md
@@ -877,6 +879,7 @@ nav:
           - ./api_reference/llms/rungpt.md
           - ./api_reference/llms/sagemaker_endpoint.md
           - ./api_reference/llms/solar.md
+          - ./api_reference/llms/text_generation_inference.md
           - ./api_reference/llms/together.md
           - ./api_reference/llms/unify.md
           - ./api_reference/llms/upstage.md
@@ -1964,6 +1967,9 @@ plugins:
             - ../llama-index-integrations/readers/llama-index-readers-azure-devops
             - ../llama-index-integrations/retrievers/llama-index-retrievers-duckdb-retriever
             - ../llama-index-packs/llama-index-packs-zenguard
+            - ../llama-index-integrations/embeddings/llama-index-embeddings-huggingface-api
+            - ../llama-index-integrations/llms/llama-index-llms-text-generation-inference
+            - ../llama-index-integrations/llms/llama-index-llms-huggingface-api
   - redirects:
       redirect_maps:
         ./api/llama_index.vector_stores.MongoDBAtlasVectorSearch.html: api_reference/storage/vector_store/mongodb.md

diff --git a/llama-index-core/llama_index/core/embeddings/loading.py b/llama-index-core/llama_index/core/embeddings/loading.py
@@ -25,7 +25,7 @@
     pass
 
 try:
-    from llama_index.embeddings.huggingface import (
+    from llama_index.embeddings.huggingface_api import (
         HuggingFaceInferenceAPIEmbedding,
     )  # pants: no-infer-dep
 

diff --git a/llama-index-core/llama_index/core/ingestion/transformations.py b/llama-index-core/llama_index/core/ingestion/transformations.py
@@ -285,7 +285,7 @@ def build_configured_transformation(
         pass
 
     try:
-        from llama_index.embeddings.huggingface import (
+        from llama_index.embeddings.huggingface_api import (
             HuggingFaceInferenceAPIEmbedding,
         )  # pants: no-infer-dep
 

diff --git a/llama-index-core/llama_index/core/llms/loading.py b/llama-index-core/llama_index/core/llms/loading.py
@@ -25,7 +25,7 @@
     pass
 
 try:
-    from llama_index.llms.huggingface import (
+    from llama_index.llms.huggingface_api import (
         HuggingFaceInferenceAPI,
     )  # pants: no-infer-dep