Skip to content

The agent engineering platform.

fix RAG with quantized embeddings notebook (#24422)

rbrugaro37b89fb7fc9c
diff --git a/cookbook/rag_with_quantized_embeddings.ipynb b/cookbook/rag_with_quantized_embeddings.ipynb
index a071ec21b1..001f2b4727 100644
--- a/cookbook/rag_with_quantized_embeddings.ipynb
+++ b/cookbook/rag_with_quantized_embeddings.ipynb
@@ -370,13 +370,14 @@
],
"source": [
"import torch\n",
- "from langchain.llms.huggingface_pipeline import HuggingFacePipeline\n",
- "from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline\n",
+ "from langchain_huggingface.llms import HuggingFacePipeline\n",
+ "from optimum.intel.ipex import IPEXModelForCausalLM\n",
+ "from transformers import AutoTokenizer, pipeline\n",
"\n",
"model_id = \"Intel/neural-chat-7b-v3-3\"\n",
"tokenizer = AutoTokenizer.from_pretrained(model_id)\n",
- "model = AutoModelForCausalLM.from_pretrained(\n",
- " model_id, device_map=\"auto\", torch_dtype=torch.bfloat16\n",
+ "model = IPEXModelForCausalLM.from_pretrained(\n",
+ " model_id, torch_dtype=torch.bfloat16, export=True\n",
")\n",
"\n",
"pipe = pipeline(\"text-generation\", model=model, tokenizer=tokenizer, max_new_tokens=100)\n",
@@ -581,7 +582,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.9.18"
+ "version": "3.10.14"
}
},
"nbformat": 4,