Upload app.py with huggingface_hub

app.py

@@ -1,8 +1,5 @@
 from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
 from langchain_core.prompts import ChatPromptTemplate
-from langchain.prompts import PromptTemplate
-from langchain_core.output_parsers import StrOutputParser
-from langchain.memory import ConversationSummaryMemory
 from langchain_huggingface import HuggingFacePipeline
 from langchain_core.runnables import RunnableSequence
 import gradio as gr
@@ -17,28 +14,26 @@ generator = pipeline(
     "text-generation",
     model=model,
     tokenizer=tokenizer,
-    max_new_tokens=100,
+    max_new_tokens=200,
     do_sample=True,
     temperature=0.7
 )
 
-
-
 # LangChain wrapper
 llm = HuggingFacePipeline(pipeline=generator)
 
 # Prompt template
 prompt = ChatPromptTemplate.from_messages([
-     ("system", "You are a helpful assistant. Explain the following code clearly:\n\n{code}")
+    ("system", "You are a helpful assistant. Please respond to the user queries."),
+    ("user", "Question: {question}")
 ])
 
 # Runnable sequence instead of LLMChain
-chain = prompt | llm | StrOutputParser()
-
+chain = prompt | llm
 
 # Gradio interface
-def generate_answer(input_code):
-    result = chain.invoke({"code":input_code })
+def generate_answer(question):
+    result = chain.invoke({"question": question})
     return result
 
-gr.Interface(fn=generate_answer, inputs="text", outputs="text", title="Gemma 2B Code Explainer").launch()
+gr.Interface(fn=generate_answer, inputs="text", outputs="text", title="Gemma 2B Chat").launch()