Update app.py

app.py

@@ -13,7 +13,6 @@ from numpy import ndarray
 from qdrant_client import QdrantClient, models
 from llama_cpp import Llama, GGML_TYPE_I8
 from optimum_encoder import OptimumEncoder
-from fastembed_encoder import FastEmbedEncoder
 from unstructured.partition.auto import partition
 from statistical_chunker import StatisticalChunker
 from fastembed import SparseEmbedding, SparseTextEmbedding
@@ -138,15 +137,9 @@ def load_models_and_documents():
             type_v=GGML_TYPE_I8
         )
 
-        #dense_model = OptimumEncoder(
-            #device="cuda",
-            #cache_dir=os.getenv('HF_HOME')
-        #)
-
-        dense_model = FastEmbedEncoder(
-            name='mixedbread-ai/mxbai-embed-large-v1',
-            cache_dir=os.getenv('HF_HOME'),
-            providers=['CPUExecutionProvider']
+        dense_model = OptimumEncoder(
+            device="cuda",
+            cache_dir=os.getenv('HF_HOME')
         )
         
         sparse_model = SparseTextEmbedding(