add

.gitignore

@@ -0,0 +1,2 @@
+.venv
+gradio_chached_examples/

app.py

@@ -0,0 +1,49 @@
+import gradio as gr
+from peft import PeftModel, PeftConfig
+from transformers import AutoModelForCausalLM, AutoTokenizer
+import torch
+import re
+
+model_name = "google/gemma-2b"
+peft_model = "kazuma313/gemma-dokter-ft"
+device_map = "auto"
+
+base_model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    low_cpu_mem_usage=True,
+    return_dict=True,
+    torch_dtype=torch.float16,
+    device_map=device_map,
+)
+model = PeftModel.from_pretrained(base_model, peft_model)
+model = model.merge_and_unload()
+
+# Reload tokenizer to save it
+tokenizer = AutoTokenizer.from_pretrained(model_name, 
+                                          trust_remote_code=True)
+tokenizer.pad_token = tokenizer.eos_token
+tokenizer.padding_side = "right"
+
+def echo(message, history, tokens):
+    pattern = r'Step \d+/\d+|^\d+\.\s*'
+    input_ids = tokenizer(message, return_tensors="pt")
+    outputs = model.generate(**input_ids, max_length=tokens)
+    answer = tokenizer.decode(outputs[0], skip_special_tokens=True).split('Answer:')[-1]
+    clean_answer = re.sub(pattern, '', answer)
+
+    return clean_answer
+
+
+demo = gr.ChatInterface(echo, 
+                        examples = [["what is the negative effect of alcohol?"],
+                                    ["i have lack of sleep, what happend if continously do this?"]],
+                        title="dokter Bot",
+                        retry_btn=None,
+                        undo_btn="Delete Previous",
+                        clear_btn="Clear",
+                        additional_inputs=[
+                            gr.Slider(64, 256, value=124)
+                        ], 
+    
+                       )
+demo.launch()

requirements.txt

@@ -0,0 +1,4 @@
+gradio
+peft
+transformers
+torch