Update app.py

app.py

@@ -2,4 +2,73 @@ import streamlit as st
 from transformers import pipeline
 from PIL import Image
 
-pipeline = pipeline(task="image-classification", model="julien-c/hotdog-not-hotdog")
+st.title("LLM Translate for ko->eng")
+
+# adding the text that will show in the text box as default
+
+text_default = """
+그를 중심으로 휘몰아치는 막대한마나. 허공에서 피어오른 다섯 개의불꽂이 크기를 부풀리고, 이내 포탄처럼 쏘아졌다.
+
+후우우우웅, 까앙!
+
+수만의 몬스터로 이루어진 검은 파도가 갈라졌다. 초고온의 열기가 살과 뼈를 태우고 지면을 녹였다."""
+
+
+from peft import AutoPeftModelForCausalLM
+from transformers import AutoTokenizer
+import torch
+
+attn_implementation = None
+if USE_FLASH_ATTENTION:
+  attn_implementation="flash_attention_2"
+
+
+model_id = "r1208/c4ai-command-r-v01-4bit_32r"
+
+model = AutoPeftModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.bfloat16, attn_implementation=attn_implementation,)
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+
+tokenizer_with_prefix_space = AutoTokenizer.from_pretrained(model_id, add_prefix_space=True)
+
+def get_tokens_as_list(word_list):
+    "Converts a sequence of words into a list of tokens"
+    tokens_list = []
+    for word in word_list:
+        tokenized_word = tokenizer_with_prefix_space([word], add_special_tokens=False).input_ids[0]
+        tokens_list.append(tokenized_word)
+    return tokens_list
+bad_words_ids = get_tokens_as_list( word_list=["\n", "\n\n", "\ ", " \ ", "\\", "'\n'"] )
+
+max_new_tokens = st.sidebar.slider("Max Length", value=100, min_value=10, max_value=1000)
+temperature = st.sidebar.slider("Temperature", value=0.3, min_value=0.0, max_value=1.0, step=0.05)
+top_k = st.sidebar.slider("Top-k", min_value=0, max_value=50, value=0)
+top_p = st.sidebar.slider("Top-p", min_value=0.75, max_value=1.0, step=0.05, value=0.9)
+
+def translate(text):
+    # Prepare the prompt
+    messages = f"Translate from Korean to English: {text}"
+    input_ids = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt")
+    
+    # Generate the translation
+    outputs = model.generate(input_ids, max_length=max_new_tokens, temperature=temperature, top_k=top_k, top_p=top_p, bad_words_ids = bad_words_ids)
+    
+    translation = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    
+    return translation
+
+def main():
+    st.subheader("Enter text to translate")
+    input_text = st.text_area("", height=300)
+    
+    if st.button("Translate"):
+        if input_text:
+            translation = translate(input_text)
+            st.text_area("Translated Text", value=translation, height=300)
+        else:
+            st.error("Please enter some text to translate.")
+
+if __name__ == "__main__":
+    main()
+
+
+