Update app.py
Browse files
app.py
CHANGED
@@ -807,6 +807,21 @@ def get_llm_suggested_settings(file, num_chunks=1):
|
|
807 |
9. Whether to optimize vocabulary
|
808 |
10. Whether to apply phonetic matching
|
809 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
810 |
Text chunks:
|
811 |
{' '.join(sample_chunks)}
|
812 |
|
|
|
807 |
9. Whether to optimize vocabulary
|
808 |
10. Whether to apply phonetic matching
|
809 |
|
810 |
+
Expected output format:
|
811 |
+
{{
|
812 |
+
"embedding_models": "embedding_model_type:embedding_model_name",
|
813 |
+
"split_strategy": "token or recursive",
|
814 |
+
"chunk_size": 250,
|
815 |
+
"overlap_size": 50,
|
816 |
+
"vector_store_type": "FAISS or Chroma",
|
817 |
+
"search_type": "similarity, mmr, or custom",
|
818 |
+
"top_k": 5,
|
819 |
+
"apply_preprocessing": True,
|
820 |
+
"optimize_vocab": True,
|
821 |
+
"apply_phonetic": False,
|
822 |
+
"phonetic_weight": 0.3 # Default value, as it's not in the LLM suggestions
|
823 |
+
}}
|
824 |
+
|
825 |
Text chunks:
|
826 |
{' '.join(sample_chunks)}
|
827 |
|