Update app.py

app.py

@@ -16,7 +16,7 @@ dataset = load_dataset("Yoxas/statistical_literacyv2")
 
 data = dataset["train"]
 
-# Convert the string embeddings to numerical arrays and ensure they are 2D
+# Convert string embeddings to numpy arrays and ensure they are 2D
 def convert_and_ensure_2d_embeddings(example):
     embedding_str = example['embedding']
     embedding_str = embedding_str.replace('\n', ' ').replace('...', '')
@@ -39,8 +39,8 @@ def flatten_embeddings(example):
 
 data = data.map(flatten_embeddings)
 
-# Ensure embeddings are in the correct shape for FAISS
-embeddings = np.array([example['embedding'] for example in data], dtype=np.float32)
+# Extract embeddings and convert to numpy array
+embeddings = np.vstack([example['embedding'] for example in data])
 
 # Add FAISS index
 data = data.add_faiss_index_from_external_arrays("embedding", embeddings)
@@ -131,11 +131,11 @@ TITLE = "# RAG"
 
 DESCRIPTION = """
 A RAG pipeline with a chatbot feature
-Resources used to build this project :
-* Embedding model : https://huggingface.co/mixedbread-ai/mxbai-embed-large-v1
-* Dataset : https://huggingface.co/datasets/not-lain/wikipedia
-* FAISS docs : https://huggingface.co/docs/datasets/v2.18.0/en/package_reference/main_classes#datasets.Dataset.add_faiss_index 
-* Chatbot : https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct
+Resources used to build this project:
+* Embedding model: https://huggingface.co/mixedbread-ai/mxbai-embed-large-v1
+* Dataset: https://huggingface.co/datasets/not-lain/wikipedia
+* FAISS docs: https://huggingface.co/docs/datasets/v2.18.0/en/package_reference/main_classes#datasets.Dataset.add_faiss_index 
+* Chatbot: https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct
 """
 
 demo = gr.ChatInterface(