Initial Commit

app.py

@@ -0,0 +1,36 @@
+from transformers import AutoTokenizer
+import transformers
+import torch
+import streamlit as st
+
+prompt = st.text_input('Prompt', 'Hello, How you doing ?')
+
+model = "meta-llama/Llama-2-13b-chat-hf"
+
+#Not Working
+#tokenizer = AutoTokenizer.from_pretrained(model)
+
+
+pipeline = transformers.pipeline(
+    "text-generation",
+    model=model,
+    torch_dtype=torch.float16,
+    device_map="auto",
+    do_sample=True,
+)
+
+def get_llama_response(prompt):
+    
+    sequences = pipeline(
+        prompt,
+        do_sample=True,
+        top_k=10,
+        num_return_sequences=1,
+        max_length=256,
+    )
+    print(sequences[0]['generated_text'])
+
+#prompt="Can you help me to write rest api endpoints in python ?"
+response = get_llama_response(prompt)
+
+st.write('Answer: ',response)

requirements.txt

@@ -0,0 +1,3 @@
+streamlit
+transformers
+torch