updating the inference section

README.md

@@ -57,20 +57,24 @@ Users should:
 ## How to Get Started with the Model
 
 ```python
-from transformers import AutoTokenizer, AutoModelForCausalLM
-from peft import PeftModel
+import torch
+from peft import PeftModel, PeftConfig
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
 peft_model_id = "Jay24-AI/bloom-3b-lora-tagger"
-tokenizer = AutoTokenizer.from_pretrained("bigscience/bloom-3b")
-base_model = AutoModelForCausalLM.from_pretrained("bigscience/bloom-3b", load_in_8bit=True, device_map="auto")
+config = PeftConfig.from_pretrained(peft_model_id)
+model = AutoModelForCausalLM.from_pretrained(config.base_model_name_or_path, return_dict=True, load_in_8bit=True, device_map='auto')
+tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path)
 
-# Load LoRA adapters
-model = PeftModel.from_pretrained(base_model, peft_model_id)
+# Load the Lora model
+model = PeftModel.from_pretrained(model, peft_model_id)
 
-prompt = "“Training models with PEFT and LoRa is cool” ->: "
-inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
-outputs = model.generate(**inputs, max_new_tokens=50)
-print(tokenizer.decode(outputs[0], skip_special_tokens=True))
+batch = tokenizer("“The only way to do great work is to love what you do.” ->:", return_tensors='pt')
+
+with torch.cuda.amp.autocast():
+  output_tokens = model.generate(**batch, max_new_tokens=50)
+
+print('\n\n', tokenizer.decode(output_tokens[0], skip_special_tokens=True))
 ```
 
 ## Training Details