ankanghosh commited on
Commit
c056c01
·
verified ·
1 Parent(s): 60a3d8c

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +16 -3
app.py CHANGED
@@ -381,16 +381,29 @@ if st.session_state.last_answer is not None:
381
  # Add helpful information
382
  st.markdown("---")
383
 
 
384
  # About section with enhanced explanations
385
  st.markdown("""
386
  ### About this app
387
- This application uses a Retrieval-Augmented Generation (RAG) system to answer questions about spirituality based on insights from Indian
388
  spiritual texts. It searches through a database of texts to find relevant passages and generates answers based on those passages.
389
 
390
- Techical Information (Optional): The RAG pipeline is powered by an underlying large embedding model that converts texts into searchable
391
  vector representations (using FAISS indices) and a Large Language Model (LLM) that summarizes and presents the findings of the system.
392
 
393
- **Important to note:**
 
 
 
 
 
 
 
 
 
 
 
 
394
  - This is not a general chatbot. It is specifically designed to answer spiritual questions based on referenced texts, as against generating historical information or reproducing stories of saints or spiritual leaders.
395
  - You may receive slightly different answers when asking the same question multiple times. This variation is intentional and reflects the nuanced nature of spiritual teachings across different traditions.
396
  - While you can select a specific number of citations and word limit, the actual response may contain fewer citations based on relevance and availability of information. Similarly, explanations may be shorter than the selected word limit if the retrieved information is concise.
 
381
  # Add helpful information
382
  st.markdown("---")
383
 
384
+ # About section with enhanced explanations
385
  # About section with enhanced explanations
386
  st.markdown("""
387
  ### About this app
388
+ **General:** This application uses a Retrieval-Augmented Generation (RAG) system to answer questions about spirituality based on insights from Indian
389
  spiritual texts. It searches through a database of texts to find relevant passages and generates answers based on those passages.
390
 
391
+ **Techical**: The RAG pipeline is powered by an underlying large embedding model that converts texts into searchable
392
  vector representations (using FAISS indices) and a Large Language Model (LLM) that summarizes and presents the findings of the system.
393
 
394
+ #### Goal
395
+ The path and journey to the SELF is designed to be undertaken alone. The all-encompassing knowledge is internal and not external.
396
+ Not all find the SatGuru - the true, real Guru, for guidance.
397
+
398
+ It has been observed that all of the necessary knowledge for spiritual progress is available in Indian spiritual texts.
399
+ Additionally, great saints, sages, mystics, yogis, sadhus, rishis, and siddhas have imparted spiritual wisdom, knowledge, and guidance to beings
400
+ for ages in the Indian subcontinent and beyond.
401
+
402
+ Our goal is to make a small contribution to the journey of beings towards self-discovery by making this knowledge available and
403
+ accessible, within ethical, moral, and resource-based constraints. **We do not have any commercial or for-profit interests and this application is
404
+ for educational purposes only.** We genuinely hope this humble offering will be of help towards your spiritual progress. God Bless.
405
+
406
+ #### Important to note
407
  - This is not a general chatbot. It is specifically designed to answer spiritual questions based on referenced texts, as against generating historical information or reproducing stories of saints or spiritual leaders.
408
  - You may receive slightly different answers when asking the same question multiple times. This variation is intentional and reflects the nuanced nature of spiritual teachings across different traditions.
409
  - While you can select a specific number of citations and word limit, the actual response may contain fewer citations based on relevance and availability of information. Similarly, explanations may be shorter than the selected word limit if the retrieved information is concise.