Upload 2 files

app.py

@@ -0,0 +1,52 @@
+import streamlit as st
+from unsloth import FastLanguageModel
+import torch
+max_seq_length = 2048 # Choose any! We auto support RoPE Scaling internally!
+dtype = None # None for auto detection. Float16 for Tesla T4, V100, Bfloat16 for Ampere+
+load_in_4bit = True 
+
+
+model, tokenizer = FastLanguageModel.from_pretrained(
+    model_name = "Mudditha/test-phi-3", # YOUR MODEL YOU USED FOR TRAINING
+    max_seq_length = max_seq_length,
+    dtype = dtype,
+    load_in_4bit = load_in_4bit,
+)
+FastLanguageModel.for_inference(model)
+
+
+inf_prompt_1 = """
+
+### Input:
+{}
+
+### Response:
+{}"""
+
+
+def print_output(input):
+    inputs = tokenizer(
+    [
+        inf_prompt_1.format(
+            input, # instruction
+            "", # output - leave this blank for generation!
+        ),
+    ], return_tensors = "pt").to("cuda")
+    
+    # outputs = model.generate(**inputs, max_new_tokens = 64, use_cache = True)
+    # tokenizer.batch_decode(outputs)
+    
+    outputs = model.generate(**inputs, max_new_tokens = 100, use_cache = True)
+    return tokenizer.batch_decode(outputs)
+
+
+# Streamlit app UI
+st.title("Echo Input Example")
+
+# Text box for user input
+user_input = st.text_input("Type something here:")
+
+# Button to submit input
+if st.button("Submit"):
+    # Reprint the user input
+    st.write(f"You typed: {print_output(user_input)}")

requirements.txt

@@ -0,0 +1,6 @@
+torch
+unsloth
+
+
+# This is only needed for local deployment
+streamlit