loading adapter model ontop

server/text_generation_server/models/causal_lm.py

@@ -1,10 +1,12 @@
 import torch
 import inspect
-
+import json
+  
 from dataclasses import dataclass
 from opentelemetry import trace
 from transformers import AutoConfig, AutoTokenizer, AutoModelForCausalLM, PreTrainedTokenizerBase, BitsAndBytesConfig
 from typing import Optional, Tuple, List, Type, Dict
+from peft import PeftModelForCausalLM, get_peft_config, PeftConfig
 
 from text_generation_server.models import Model
 from text_generation_server.models.types import (
@@ -476,11 +478,15 @@ class CausalLM(Model):
         )
 
         should_quantize = quantize == "bitsandbytes" 
-        if(should_quantize):
+        if should_quantize:
             quantization_config = BitsAndBytesConfig(
                 load_in_4bit=True,
                 bnb_4bit_compute_dtype=torch.float16
             )
+
+        has_peft_model = True
+        peft_model_id_or_path = "/mnt/TOFU/HF_MODELS/Llama-2-7b-chat-hf-instruct-pl-lora_adapter_model"
+
         model = AutoModelForCausalLM.from_pretrained(
             model_id,
             revision=revision,
@@ -490,6 +496,15 @@ class CausalLM(Model):
             quantization_config = quantization_config if should_quantize else None,
             trust_remote_code=trust_remote_code,
         )
+        if has_peft_model:
+            with open(f'{peft_model_id_or_path}/adapter_config.json') as config_file:
+                config  = json.load(config_file)
+                peft_config = get_peft_config(config)
+                model = PeftModelForCausalLM(model, peft_config)
+                ## Llama does not have a load_adapter method - we need to think about hot swapping here and implement this for Llama
+                # model.load_adapter(peft_model_id_or_path)
+                # model.enable_adapters()
+
         ## ValueError: Calling `cuda()` is not supported for `4-bit` or `8-bit` quantized models. Please use the model as it is, since the model has already been set to the correct devices and casted to the correct `dtype`.
         # if torch.cuda.is_available() and torch.cuda.device_count() == 1:
             # model = model.cuda()