fine tune LLM

app.py

@@ -2,11 +2,6 @@ import gradio as gr
 # from langchain.vectorstores import Chroma
 
 
-import chromadb
-client = chromadb.PersistentClient(path="chroma.db")
-
-db = client.get_collection(name="banks")
-
 '''
 https://huggingface.co/spaces/kevinhug/clientX
 
@@ -17,31 +12,53 @@ counter="""
 <div class="elfsight-app-5f3e8eb9-9103-490e-9999-e20aa4157dc7" data-elfsight-app-lazy></div>
 """
 
+'''
+SIMILAR VECTOR DB SEARCH
+'''
+import chromadb
+client = chromadb.PersistentClient(path="chroma.db")
+
+db = client.get_collection(name="banks")
+
 def similar(issue):
   global db
   docs = db.query(query_texts=issue, n_results=5)
   return docs
 
-
+'''
+FINE TUNE LLM LIKE SCORE
+'''
+from fastai.vision.all import *
+learn = load_learner('banks_txt_like.pkl')
+def like(issue):
+  pred,idx,probs = learn.predict(issue)
+  return pred
 '''
 https://www.gradio.app/docs/interface
 '''
-iface = gr.Interface(fn=similar, inputs="text", outputs="json",
-                     title="Enhancing Customer Engagement and Operational Efficiency with Semantic Similarity Document Search (SSDS)",
-                     examples=[["having bad client experience"],
-  ["having credit card problem"],
-  ["late payment fee"],
-  ["credit score dropping"]],
-                     description="""
+
+
+
+
+with gr.Blocks() as demo:
+  gr.Markdown("""Enhancing Customer Engagement and Operational Efficiency with NLP
+  =========
+  LLM
+  Semantic Similarity Document Search (SSDS)
+  
   Data Scientist: Kevin Wong, objectdeveloper@gmail.com, 416-903-7937
-  ============
-  open source ml bank dataset
+
+  Open source ml bank dataset
 https://www.kaggle.com/datasets/trainingdatapro/20000-customers-reviews-on-banks/?select=Banks.csv
+  """)
+  with gr.Tab("Semantic Similarity Document Search (SSDS)"):
+    in_similar = gr.Textbox(placeholder="having credit card problem")
+    out_similar = gr.JSON()
 
-  Using Sentence Embedding to inject Public ML Banks Text Dataset @ https://github.com/kevinwkc/analytics/blob/master/ai/vectorDB.py""",
-                       custom_js=counter,
-                     article="""
-      
+    btn_similar = gr.Button("Find Similar Verbatim")
+    btn_similar.click(fn=similar, inputs=in_similar, outputs=out_similar)
+
+    gr.Markdown("""
 Description:
 =======
 In today's dynamic financial landscape, the Semantic Similarity Document Search (SSDS) capability is a practical innovation to improve client experience, marketing leads, and sentiment analysis. As a Data Scientist with a decades in the financial industry, I see the value of SSDS in action.
@@ -71,15 +88,21 @@ With no need for jargon, SSDS delivers tangible value to our fintech operations.
 ### issue:   
   - upset customer
   
-Future Improvement
-============
-tuning the distance for use case
- 
- <script src="https://static.elfsight.com/platform/platform.js" data-use-service-core defer></script>
-<div class="elfsight-app-5f3e8eb9-9103-490e-9999-e20aa4157dc7" data-elfsight-app-lazy></div>
- """)
-iface.launch()
+Using Sentence Embedding to inject Public ML Banks Text Dataset @ https://github.com/kevinwkc/analytics/blob/master/ai/vectorDB.py
+    """)
+
+  with gr.Tab("Fine Tune LLM")
+    in_like = gr.Textbox(placeholder="having credit card problem")
+    out_like = gr.Textbox(placeholder="like score")
+
+    btn_like = gr.Button("Find Like Score")
+    btn_like.click(fn=like, inputs=in_like, outputs=out_like)
+
+
+  with gr.Accordion("Future Improvement"):
+    gr.Markdown("""
+  tuning the distance for use case
+    """)
+
+demo.launch()
 
-iface2 = gr.Interface(fn=similar, inputs="text", outputs="json",
-                     title="testing")
-iface.launch()

banks_txt_like.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3e0db2b1e176931dde3f5172bb57aac30df4fe3521b80dbf330be74f0dde368
+size 130662474

requirements.txt

@@ -1 +1,2 @@
-chromadb
+chromadb
+fastai