adding support for int8 lora training

setup.py

@@ -43,6 +43,7 @@
         "torch>=1.13.0",
         "transformers",
         "accelerate",
+        "bitsandbytes",
     ],
     extras_require=extras,
     classifiers=[

src/peft/tuners/lora.py

@@ -25,6 +25,8 @@
 import torch.nn.functional as F
 from transformers.pytorch_utils import Conv1D
 
+import bitsandbytes as bnb
+
 from ..utils import PeftConfig, PeftType, transpose
 
 
@@ -118,7 +120,9 @@ def _find_and_replace(self):
             if any(key.endswith(target_key) for target_key in self.peft_config.target_modules):
                 parent, target, target_name = self._get_submodules(key)
                 bias = target.bias is not None
-                if isinstance(target, torch.nn.Linear) and self.peft_config.enable_lora is None:
+                if isinstance(target, bnb.nn.Linear8bitLt) and self.peft_config.enable_lora is None:
+                    new_module = Linear8bitLt(target.in_features, target.out_features, **kwargs)
+                elif isinstance(target, torch.nn.Linear) and self.peft_config.enable_lora is None:
                     new_module = Linear(target.in_features, target.out_features, bias=bias, **kwargs)
                 elif self.peft_config.enable_lora is not None:
                     kwargs.update({"enable_lora": self.peft_config.enable_lora})
@@ -358,3 +362,38 @@ def forward(self, x: torch.Tensor):
                 after_B = self.lora_B(after_A.transpose(-2, -1)).transpose(-2, -1)
                 result += self.zero_pad(after_B) * self.scaling
             return result
+
+
+class Linear8bitLt(bnb.nn.Linear8bitLt, LoraLayer):
+    # Lora implemented in a dense layer
+    def __init__(
+        self,
+        in_features,
+        out_features,
+        r: int = 0,
+        lora_alpha: int = 1,
+        lora_dropout: float = 0.0,
+        **kwargs,
+    ):
+        bnb.nn.Linear8bitLt.__init__(self, in_features, out_features, kwargs.get("bias", None))
+        LoraLayer.__init__(self, r=r, lora_alpha=lora_alpha, lora_dropout=lora_dropout, merge_weights=False)
+        # Actual trainable parameters
+        if r > 0:
+            self.lora_A = nn.Linear(in_features, r, bias=False)
+            self.lora_B = nn.Linear(r, out_features, bias=False)
+            self.scaling = self.lora_alpha / self.r
+            # Freezing the pre-trained weight matrix
+            self.weight.requires_grad = False
+        self.reset_parameters()
+
+    def reset_parameters(self):
+        if hasattr(self, "lora_A"):
+            # initialize A the same way as the default for nn.Linear and B to zero
+            nn.init.kaiming_uniform_(self.lora_A.weight, a=math.sqrt(5))
+            nn.init.zeros_(self.lora_B.weight)
+
+    def forward(self, x: torch.Tensor):
+        result = super().forward(x)
+        if self.r > 0:
+            result += self.lora_B(self.lora_A(self.lora_dropout(x))) * self.scaling
+        return result