Update app.py

app.py

@@ -14,6 +14,7 @@ from scipy.sparse import csr_matrix, save_npz, load_npz, vstack
 from qdrant_client import QdrantClient, models
 from langchain_community.document_loaders import WikipediaLoader, WebBaseLoader
 from fastembed_ext import FastEmbedEmbeddingsLc
+from langchain_experimental.text_splitter import SemanticChunker
 from fastembed.sparse.splade_pp import supported_splade_models
 from fastembed import SparseTextEmbedding, SparseEmbedding
 from unstructured.partition.auto import partition
@@ -193,9 +194,13 @@ def load_models_and_documents():
             n_gpu_layers=32
         )
 
-        dense_model = HuggingFaceEncoder(name="mixedbread-ai/mxbai-embed-large-v1", device="cuda")
-
         provider = ['CPUExecutionProvider']
+
+        dense_model = FastEmbedEmbeddingsLc(
+            model_name="mixedbread-ai/mxbai-embed-large-v1", 
+            cache_dir=os.getenv('HF_HOME'),
+            providers=provider
+        )
         
         sparse_model = SparseTextEmbedding(
             'prithivida/Splade_PP_en_v2',