replace 'shared_memory' with 'share_inputs' (#1600)

Author: openvino-dev-samples

notebooks/254-llm-chatbot/254-rag-chatbot.ipynb

@@ -76,7 +76,7 @@
     "\"git+https://github.com/openvinotoolkit/nncf@release_v280\"\\\n",
     "\"openvino-nightly\"\\\n",
     "\"gradio\"\\\n",
-    "\"onnx\" \"chromadb\" \"sentence_transformers\" \"langchain\" \"langchainhub\" \"transformers>=4.34.0\" \"unstructured\" \"scikit-learn\" \"python-docx\""
+    "\"onnx\" \"chromadb\" \"sentence_transformers\" \"langchain\" \"langchainhub\" \"transformers>=4.34.0\" \"unstructured\" \"scikit-learn\" \"python-docx\" \"pypdf\""
    ]
   },
   {

notebooks/254-llm-chatbot/ov_llm_model.py

@@ -101,7 +101,7 @@ def forward(
             inputs["attention_mask"] = np.array(attention_mask)
 
         # Run inference
-        self.request.start_async(inputs, shared_memory=True)
+        self.request.start_async(inputs, share_inputs=True)
         self.request.wait()
 
         logits = torch.from_numpy(self.request.get_tensor("logits").data).to(
@@ -248,7 +248,7 @@ def forward(
         if "token_type_ids" in self.input_names and attention_mask is not None:
             inputs["token_type_ids"] = np.array(attention_mask)
         # Run inference
-        self.request.start_async(inputs, shared_memory=True)
+        self.request.start_async(inputs, share_inputs=True)
         self.request.wait()
 
         logits = torch.from_numpy(self.request.get_tensor("logits").data).to(
@@ -397,7 +397,7 @@ def forward(
         if "position_ids" in self.input_names and attention_mask is not None:
             inputs["position_ids"] = np.array(position_ids)
         # Run inference
-        self.request.start_async(inputs, shared_memory=True)
+        self.request.start_async(inputs, share_inputs=True)
         self.request.wait()
 
         logits = torch.from_numpy(self.request.get_tensor("logits").data).to(