Merge branch 'lora_inplace'

extensions-builtin/Lora/lora.py

@@ -8,14 +8,27 @@ from modules import shared, devices, sd_models, errors
 metadata_tags_order = {"ss_sd_model_name": 1, "ss_resolution": 2, "ss_clip_skip": 3, "ss_num_train_images": 10, "ss_tag_frequency": 20}
 
 re_digits = re.compile(r"\d+")
-re_unet_down_blocks = re.compile(r"lora_unet_down_blocks_(\d+)_attentions_(\d+)_(.+)")
+re_x_proj = re.compile(r"(.*)_([qkv]_proj)$")
-re_unet_mid_blocks = re.compile(r"lora_unet_mid_block_attentions_(\d+)_(.+)")
+re_compiled = {}
-re_unet_up_blocks = re.compile(r"lora_unet_up_blocks_(\d+)_attentions_(\d+)_(.+)")
+
-re_text_block = re.compile(r"lora_te_text_model_encoder_layers_(\d+)_(.+)")
+suffix_conversion = {
+    "attentions": {},
+    "resnets": {
+        "conv1": "in_layers_2",
+        "conv2": "out_layers_3",
+        "time_emb_proj": "emb_layers_1",
+        "conv_shortcut": "skip_connection",
+    }
+}
 
 
-def convert_diffusers_name_to_compvis(key):
+def convert_diffusers_name_to_compvis(key, is_sd2):
-    def match(match_list, regex):
+    def match(match_list, regex_text):
+        regex = re_compiled.get(regex_text)
+        if regex is None:
+            regex = re.compile(regex_text)
+            re_compiled[regex_text] = regex
+
         r = re.match(regex, key)
         if not r:
             return False
@@ -26,16 +39,33 @@ def convert_diffusers_name_to_compvis(key):
 
     m = []
 
-    if match(m, re_unet_down_blocks):
+    if match(m, r"lora_unet_down_blocks_(\d+)_(attentions|resnets)_(\d+)_(.+)"):
-        return f"diffusion_model_input_blocks_{1 + m[0] * 3 + m[1]}_1_{m[2]}"
+        suffix = suffix_conversion.get(m[1], {}).get(m[3], m[3])
+        return f"diffusion_model_input_blocks_{1 + m[0] * 3 + m[2]}_{1 if m[1] == 'attentions' else 0}_{suffix}"
 
-    if match(m, re_unet_mid_blocks):
+    if match(m, r"lora_unet_mid_block_(attentions|resnets)_(\d+)_(.+)"):
-        return f"diffusion_model_middle_block_1_{m[1]}"
+        suffix = suffix_conversion.get(m[0], {}).get(m[2], m[2])
+        return f"diffusion_model_middle_block_{1 if m[0] == 'attentions' else m[1] * 2}_{suffix}"
 
-    if match(m, re_unet_up_blocks):
+    if match(m, r"lora_unet_up_blocks_(\d+)_(attentions|resnets)_(\d+)_(.+)"):
-        return f"diffusion_model_output_blocks_{m[0] * 3 + m[1]}_1_{m[2]}"
+        suffix = suffix_conversion.get(m[1], {}).get(m[3], m[3])
+        return f"diffusion_model_output_blocks_{m[0] * 3 + m[2]}_{1 if m[1] == 'attentions' else 0}_{suffix}"
+
+    if match(m, r"lora_unet_down_blocks_(\d+)_downsamplers_0_conv"):
+        return f"diffusion_model_input_blocks_{3 + m[0] * 3}_0_op"
+
+    if match(m, r"lora_unet_up_blocks_(\d+)_upsamplers_0_conv"):
+        return f"diffusion_model_output_blocks_{2 + m[0] * 3}_{2 if m[0]>0 else 1}_conv"
+
+    if match(m, r"lora_te_text_model_encoder_layers_(\d+)_(.+)"):
+        if is_sd2:
+            if 'mlp_fc1' in m[1]:
+                return f"model_transformer_resblocks_{m[0]}_{m[1].replace('mlp_fc1', 'mlp_c_fc')}"
+            elif 'mlp_fc2' in m[1]:
+                return f"model_transformer_resblocks_{m[0]}_{m[1].replace('mlp_fc2', 'mlp_c_proj')}"
+            else:
+                return f"model_transformer_resblocks_{m[0]}_{m[1].replace('self_attn', 'attn')}"
 
-    if match(m, re_text_block):
         return f"transformer_text_model_encoder_layers_{m[0]}_{m[1]}"
 
     return key
@@ -101,15 +131,22 @@ def load_lora(name, filename):
 
     sd = sd_models.read_state_dict(filename)
 
-    keys_failed_to_match = []
+    keys_failed_to_match = {}
+    is_sd2 = 'model_transformer_resblocks' in shared.sd_model.lora_layer_mapping
 
     for key_diffusers, weight in sd.items():
-        fullkey = convert_diffusers_name_to_compvis(key_diffusers)
+        key_diffusers_without_lora_parts, lora_key = key_diffusers.split(".", 1)
-        key, lora_key = fullkey.split(".", 1)
+        key = convert_diffusers_name_to_compvis(key_diffusers_without_lora_parts, is_sd2)
 
         sd_module = shared.sd_model.lora_layer_mapping.get(key, None)
+
         if sd_module is None:
-            keys_failed_to_match.append(key_diffusers)
+            m = re_x_proj.match(key)
+            if m:
+                sd_module = shared.sd_model.lora_layer_mapping.get(m.group(1), None)
+
+        if sd_module is None:
+            keys_failed_to_match[key_diffusers] = key
             continue
 
         lora_module = lora.modules.get(key, None)
@@ -123,15 +160,21 @@ def load_lora(name, filename):
 
         if type(sd_module) == torch.nn.Linear:
             module = torch.nn.Linear(weight.shape[1], weight.shape[0], bias=False)
+        elif type(sd_module) == torch.nn.modules.linear.NonDynamicallyQuantizableLinear:
+            module = torch.nn.Linear(weight.shape[1], weight.shape[0], bias=False)
+        elif type(sd_module) == torch.nn.MultiheadAttention:
+            module = torch.nn.Linear(weight.shape[1], weight.shape[0], bias=False)
         elif type(sd_module) == torch.nn.Conv2d:
             module = torch.nn.Conv2d(weight.shape[1], weight.shape[0], (1, 1), bias=False)
         else:
+            print(f'Lora layer {key_diffusers} matched a layer with unsupported type: {type(sd_module).__name__}')
+            continue
             assert False, f'Lora layer {key_diffusers} matched a layer with unsupported type: {type(sd_module).__name__}'
 
         with torch.no_grad():
             module.weight.copy_(weight)
 
-        module.to(device=devices.device, dtype=devices.dtype)
+        module.to(device=devices.cpu, dtype=devices.dtype)
 
         if lora_key == "lora_up.weight":
             lora_module.up = module
@@ -177,29 +220,120 @@ def load_loras(names, multipliers=None):
         loaded_loras.append(lora)
 
 
-def lora_forward(module, input, res):
+def lora_calc_updown(lora, module, target):
-    input = devices.cond_cast_unet(input)
+    with torch.no_grad():
-    if len(loaded_loras) == 0:
+        up = module.up.weight.to(target.device, dtype=target.dtype)
-        return res
+        down = module.down.weight.to(target.device, dtype=target.dtype)
+
+        if up.shape[2:] == (1, 1) and down.shape[2:] == (1, 1):
+            updown = (up.squeeze(2).squeeze(2) @ down.squeeze(2).squeeze(2)).unsqueeze(2).unsqueeze(3)
+        else:
+            updown = up @ down
+
+        updown = updown * lora.multiplier * (module.alpha / module.up.weight.shape[1] if module.alpha else 1.0)
+
+        return updown
+
+
+def lora_apply_weights(self: torch.nn.Conv2d | torch.nn.Linear | torch.nn.MultiheadAttention):
+    """
+    Applies the currently selected set of Loras to the weights of torch layer self.
+    If weights already have this particular set of loras applied, does nothing.
+    If not, restores orginal weights from backup and alters weights according to loras.
+    """
+
+    lora_layer_name = getattr(self, 'lora_layer_name', None)
+    if lora_layer_name is None:
+        return
+
+    current_names = getattr(self, "lora_current_names", ())
+    wanted_names = tuple((x.name, x.multiplier) for x in loaded_loras)
+
+    weights_backup = getattr(self, "lora_weights_backup", None)
+    if weights_backup is None:
+        if isinstance(self, torch.nn.MultiheadAttention):
+            weights_backup = (self.in_proj_weight.to(devices.cpu, copy=True), self.out_proj.weight.to(devices.cpu, copy=True))
+        else:
+            weights_backup = self.weight.to(devices.cpu, copy=True)
+
+        self.lora_weights_backup = weights_backup
+
+    if current_names != wanted_names:
+        if weights_backup is not None:
+            if isinstance(self, torch.nn.MultiheadAttention):
+                self.in_proj_weight.copy_(weights_backup[0])
+                self.out_proj.weight.copy_(weights_backup[1])
+            else:
+                self.weight.copy_(weights_backup)
 
-    lora_layer_name = getattr(module, 'lora_layer_name', None)
         for lora in loaded_loras:
             module = lora.modules.get(lora_layer_name, None)
-        if module is not None:
+            if module is not None and hasattr(self, 'weight'):
-            if shared.opts.lora_apply_to_outputs and res.shape == input.shape:
+                self.weight += lora_calc_updown(lora, module, self.weight)
-                res = res + module.up(module.down(res)) * lora.multiplier * (module.alpha / module.up.weight.shape[1] if module.alpha else 1.0)
+                continue
-            else:
-                res = res + module.up(module.down(input)) * lora.multiplier * (module.alpha / module.up.weight.shape[1] if module.alpha else 1.0)
 
-    return res
+            module_q = lora.modules.get(lora_layer_name + "_q_proj", None)
+            module_k = lora.modules.get(lora_layer_name + "_k_proj", None)
+            module_v = lora.modules.get(lora_layer_name + "_v_proj", None)
+            module_out = lora.modules.get(lora_layer_name + "_out_proj", None)
+
+            if isinstance(self, torch.nn.MultiheadAttention) and module_q and module_k and module_v and module_out:
+                updown_q = lora_calc_updown(lora, module_q, self.in_proj_weight)
+                updown_k = lora_calc_updown(lora, module_k, self.in_proj_weight)
+                updown_v = lora_calc_updown(lora, module_v, self.in_proj_weight)
+                updown_qkv = torch.vstack([updown_q, updown_k, updown_v])
+
+                self.in_proj_weight += updown_qkv
+                self.out_proj.weight += lora_calc_updown(lora, module_out, self.out_proj.weight)
+                continue
+
+            if module is None:
+                continue
+
+            print(f'failed to calculate lora weights for layer {lora_layer_name}')
+
+        setattr(self, "lora_current_names", wanted_names)
+
+
+def lora_reset_cached_weight(self: torch.nn.Conv2d | torch.nn.Linear):
+    setattr(self, "lora_current_names", ())
+    setattr(self, "lora_weights_backup", None)
 
 
 def lora_Linear_forward(self, input):
-    return lora_forward(self, input, torch.nn.Linear_forward_before_lora(self, input))
+    lora_apply_weights(self)
+
+    return torch.nn.Linear_forward_before_lora(self, input)
+
+
+def lora_Linear_load_state_dict(self, *args, **kwargs):
+    lora_reset_cached_weight(self)
+
+    return torch.nn.Linear_load_state_dict_before_lora(self, *args, **kwargs)
 
 
 def lora_Conv2d_forward(self, input):
-    return lora_forward(self, input, torch.nn.Conv2d_forward_before_lora(self, input))
+    lora_apply_weights(self)
+
+    return torch.nn.Conv2d_forward_before_lora(self, input)
+
+
+def lora_Conv2d_load_state_dict(self, *args, **kwargs):
+    lora_reset_cached_weight(self)
+
+    return torch.nn.Conv2d_load_state_dict_before_lora(self, *args, **kwargs)
+
+
+def lora_MultiheadAttention_forward(self, *args, **kwargs):
+    lora_apply_weights(self)
+
+    return torch.nn.MultiheadAttention_forward_before_lora(self, *args, **kwargs)
+
+
+def lora_MultiheadAttention_load_state_dict(self, *args, **kwargs):
+    lora_reset_cached_weight(self)
+
+    return torch.nn.MultiheadAttention_load_state_dict_before_lora(self, *args, **kwargs)
 
 
 def list_available_loras():

extensions-builtin/Lora/scripts/lora_script.py

@@ -9,7 +9,11 @@ from modules import script_callbacks, ui_extra_networks, extra_networks, shared
 
 def unload():
     torch.nn.Linear.forward = torch.nn.Linear_forward_before_lora
+    torch.nn.Linear._load_from_state_dict = torch.nn.Linear_load_state_dict_before_lora
     torch.nn.Conv2d.forward = torch.nn.Conv2d_forward_before_lora
+    torch.nn.Conv2d._load_from_state_dict = torch.nn.Conv2d_load_state_dict_before_lora
+    torch.nn.MultiheadAttention.forward = torch.nn.MultiheadAttention_forward_before_lora
+    torch.nn.MultiheadAttention._load_from_state_dict = torch.nn.MultiheadAttention_load_state_dict_before_lora
 
 
 def before_ui():
@@ -20,11 +24,27 @@ def before_ui():
 if not hasattr(torch.nn, 'Linear_forward_before_lora'):
     torch.nn.Linear_forward_before_lora = torch.nn.Linear.forward
 
+if not hasattr(torch.nn, 'Linear_load_state_dict_before_lora'):
+    torch.nn.Linear_load_state_dict_before_lora = torch.nn.Linear._load_from_state_dict
+
 if not hasattr(torch.nn, 'Conv2d_forward_before_lora'):
     torch.nn.Conv2d_forward_before_lora = torch.nn.Conv2d.forward
 
+if not hasattr(torch.nn, 'Conv2d_load_state_dict_before_lora'):
+    torch.nn.Conv2d_load_state_dict_before_lora = torch.nn.Conv2d._load_from_state_dict
+
+if not hasattr(torch.nn, 'MultiheadAttention_forward_before_lora'):
+    torch.nn.MultiheadAttention_forward_before_lora = torch.nn.MultiheadAttention.forward
+
+if not hasattr(torch.nn, 'MultiheadAttention_load_state_dict_before_lora'):
+    torch.nn.MultiheadAttention_load_state_dict_before_lora = torch.nn.MultiheadAttention._load_from_state_dict
+
 torch.nn.Linear.forward = lora.lora_Linear_forward
+torch.nn.Linear._load_from_state_dict = lora.lora_Linear_load_state_dict
 torch.nn.Conv2d.forward = lora.lora_Conv2d_forward
+torch.nn.Conv2d._load_from_state_dict = lora.lora_Conv2d_load_state_dict
+torch.nn.MultiheadAttention.forward = lora.lora_MultiheadAttention_forward
+torch.nn.MultiheadAttention._load_from_state_dict = lora.lora_MultiheadAttention_load_state_dict
 
 script_callbacks.on_model_loaded(lora.assign_lora_names_to_compvis_modules)
 script_callbacks.on_script_unloaded(unload)
@@ -33,6 +53,4 @@ script_callbacks.on_before_ui(before_ui)
 
 shared.options_templates.update(shared.options_section(('extra_networks', "Extra Networks"), {
     "sd_lora": shared.OptionInfo("None", "Add Lora to prompt", gr.Dropdown, lambda: {"choices": [""] + [x for x in lora.available_loras]}, refresh=lora.list_available_loras),
-    "lora_apply_to_outputs": shared.OptionInfo(False, "Apply Lora to outputs rather than inputs when possible (experimental)"),
-
 }))