optimum-rbln 0.1.12__py3-none-any.whl → 0.1.15__py3-none-any.whl

optimum/rbln/diffusers/models/controlnet.py

@@ -21,17 +21,17 @@
 # copied, modified, or distributed without prior written permission
 # from Rebellions Inc.
 
+import importlib
 import logging
-from pathlib import Path
 from typing import TYPE_CHECKING, Any, Dict, Optional, Union
 
 import torch
 from diffusers import ControlNetModel
-from optimum.exporters import TasksManager
-from transformers import AutoConfig, AutoModel, PretrainedConfig
+from transformers import PretrainedConfig
 
-from ...modeling_base import RBLNModel
+from ...modeling import RBLNModel
 from ...modeling_config import RBLNCompileConfig, RBLNConfig
+from ...modeling_diffusers import RBLNDiffusionMixin
 
 
 if TYPE_CHECKING:
@@ -105,33 +105,15 @@ class _ControlNetModel_Cross_Attention(torch.nn.Module):
 
 
 class RBLNControlNetModel(RBLNModel):
+    hf_library_name = "diffusers"
+    auto_model_class = ControlNetModel
+
     def __post_init__(self, **kwargs):
         super().__post_init__(**kwargs)
         self.use_encoder_hidden_states = any(
             item[0] == "encoder_hidden_states" for item in self.rbln_config.compile_cfgs[0].input_info
         )
 
-    @classmethod
-    def from_pretrained(cls, *args, **kwargs):
-        def get_model_from_task(
-            task: str,
-            model_name_or_path: Union[str, Path],
-            **kwargs,
-        ):
-            return ControlNetModel.from_pretrained(pretrained_model_name_or_path=model_name_or_path, **kwargs)
-
-        tasktmp = TasksManager.get_model_from_task
-        configtmp = AutoConfig.from_pretrained
-        modeltmp = AutoModel.from_pretrained
-        TasksManager.get_model_from_task = get_model_from_task
-        AutoConfig.from_pretrained = ControlNetModel.load_config
-        AutoModel.from_pretrained = ControlNetModel.from_pretrained
-        rt = super().from_pretrained(*args, **kwargs)
-        AutoConfig.from_pretrained = configtmp
-        AutoModel.from_pretrained = modeltmp
-        TasksManager.get_model_from_task = tasktmp
-        return rt
-
     @classmethod
     def wrap_model_if_needed(cls, model: torch.nn.Module, rbln_config: RBLNConfig) -> torch.nn.Module:
         use_encoder_hidden_states = False
@@ -144,6 +126,38 @@ class RBLNControlNetModel(RBLNModel):
         else:
             return _ControlNetModel(model).eval()
 
+    @classmethod
+    def update_rbln_config_using_pipe(cls, pipe: RBLNDiffusionMixin, rbln_config: Dict[str, Any]) -> Dict[str, Any]:
+        rbln_vae_cls = getattr(importlib.import_module("optimum.rbln"), f"RBLN{pipe.vae.__class__.__name__}")
+        rbln_unet_cls = getattr(importlib.import_module("optimum.rbln"), f"RBLN{pipe.unet.__class__.__name__}")
+        text_model_hidden_size = pipe.text_encoder_2.config.hidden_size if hasattr(pipe, "text_encoder_2") else None
+
+        batch_size = rbln_config.get("batch_size")
+        if not batch_size:
+            do_classifier_free_guidance = (
+                rbln_config.get("guidance_scale", 5.0) > 1.0 and pipe.unet.config.time_cond_proj_dim is None
+            )
+            batch_size = 2 if do_classifier_free_guidance else 1
+        else:
+            if rbln_config.get("guidance_scale"):
+                logger.warning(
+                    "guidance_scale is ignored because batch size is explicitly specified. "
+                    "To ensure consistent behavior, consider removing the guidance scale or "
+                    "adjusting the batch size configuration as needed."
+                )
+
+        rbln_config.update(
+            {
+                "max_seq_len": pipe.text_encoder.config.max_position_embeddings,
+                "text_model_hidden_size": text_model_hidden_size,
+                "vae_sample_size": rbln_vae_cls.get_vae_sample_size(pipe, rbln_config),
+                "unet_sample_size": rbln_unet_cls.get_unet_sample_size(pipe, rbln_config),
+                "batch_size": batch_size,
+            }
+        )
+
+        return rbln_config
+
     @classmethod
     def _get_rbln_config(
         cls,
@@ -151,33 +165,35 @@ class RBLNControlNetModel(RBLNModel):
         model_config: "PretrainedConfig",
         rbln_kwargs: Dict[str, Any] = {},
     ) -> RBLNConfig:
-        rbln_max_seq_len = rbln_kwargs.get("max_seq_len", None)
-        rbln_text_model_hidden_size = rbln_kwargs.get("text_model_hidden_size", None)
-        rbln_batch_size = rbln_kwargs.get("batch_size", None)
-        rbln_img_width = rbln_kwargs.get("img_width", None)
-        rbln_img_height = rbln_kwargs.get("img_height", None)
-        rbln_vae_scale_factor = rbln_kwargs.get("vae_scale_factor", None)
+        batch_size = rbln_kwargs.get("batch_size")
+        max_seq_len = rbln_kwargs.get("max_seq_len")
+        unet_sample_size = rbln_kwargs.get("unet_sample_size")
+        vae_sample_size = rbln_kwargs.get("vae_sample_size")
 
-        if rbln_batch_size is None:
-            rbln_batch_size = 1
+        if batch_size is None:
+            batch_size = 1
 
-        if rbln_max_seq_len is None:
-            rbln_max_seq_len = 77
+        if unet_sample_size is None:
+            raise ValueError(
+                "`rbln_unet_sample_size` (latent height, widht) must be specified (ex. unet's sample_size)"
+            )
 
-        if rbln_img_width is None or rbln_img_height is None or rbln_vae_scale_factor is None:
-            raise ValueError("rbln_img_width, rbln_img_height, and rbln_vae_scale_factor must be provided")
+        if vae_sample_size is None:
+            raise ValueError(
+                "`rbln_vae_sample_size` (input image height, width) must be specified (ex. vae's sample_size)"
+            )
 
-        input_width = rbln_img_width // rbln_vae_scale_factor
-        input_height = rbln_img_height // rbln_vae_scale_factor
+        if max_seq_len is None:
+            raise ValueError("`rbln_max_seq_len` (ex. text_encoder's max_position_embeddings )must be specified")
 
         input_info = [
             (
                 "sample",
                 [
-                    rbln_batch_size,
+                    batch_size,
                     model_config.in_channels,
-                    input_height,
-                    input_width,
+                    unet_sample_size[0],
+                    unet_sample_size[1],
                 ],
                 "float32",
             ),
@@ -189,23 +205,24 @@ class RBLNControlNetModel(RBLNModel):
             input_info.append(
                 (
                     "encoder_hidden_states",
-                    [
-                        rbln_batch_size,
-                        rbln_max_seq_len,
-                        model_config.cross_attention_dim,
-                    ],
+                    [batch_size, max_seq_len, model_config.cross_attention_dim],
                     "float32",
                 )
             )
 
-        input_info.append(("controlnet_cond", [rbln_batch_size, 3, rbln_img_height, rbln_img_width], "float32"))
+        input_info.append(
+            (
+                "controlnet_cond",
+                [batch_size, 3, vae_sample_size[0], vae_sample_size[1]],
+                "float32",
+            )
+        )
         input_info.append(("conditioning_scale", [], "float32"))
 
         if hasattr(model_config, "addition_embed_type") and model_config.addition_embed_type == "text_time":
-            if rbln_text_model_hidden_size is None:
-                rbln_text_model_hidden_size = 768
-            input_info.append(("text_embeds", [rbln_batch_size, rbln_text_model_hidden_size], "float32"))
-            input_info.append(("time_ids", [rbln_batch_size, 6], "float32"))
+            rbln_text_model_hidden_size = rbln_kwargs["text_model_hidden_size"]
+            input_info.append(("text_embeds", [batch_size, rbln_text_model_hidden_size], "float32"))
+            input_info.append(("time_ids", [batch_size, 6], "float32"))
 
         rbln_compile_config = RBLNCompileConfig(input_info=input_info)
 
@@ -215,18 +232,12 @@ class RBLNControlNetModel(RBLNModel):
             rbln_kwargs=rbln_kwargs,
         )
 
-        rbln_config.model_cfg.update(
-            {
-                "max_seq_len": rbln_max_seq_len,
-                "batch_size": rbln_batch_size,
-                "img_width": rbln_img_width,
-                "img_height": rbln_img_height,
-                "vae_scale_factor": rbln_vae_scale_factor,
-            }
-        )
-
         return rbln_config
 
+    @property
+    def compiled_batch_size(self):
+        return self.rbln_config.compile_cfgs[0].input_info[0][1][0]
+
     def forward(
         self,
         sample: torch.FloatTensor,
@@ -237,9 +248,18 @@ class RBLNControlNetModel(RBLNModel):
         added_cond_kwargs: Dict[str, torch.Tensor] = {},
         **kwargs,
     ):
-        """
-        The [`ControlNetModel`] forward method.
-        """
+        sample_batch_size = sample.size()[0]
+        compiled_batch_size = self.compiled_batch_size
+        if sample_batch_size != compiled_batch_size and (
+            sample_batch_size * 2 == compiled_batch_size or sample_batch_size == compiled_batch_size * 2
+        ):
+            raise ValueError(
+                f"Mismatch between ControlNet's runtime batch size ({sample_batch_size}) and compiled batch size ({compiled_batch_size}). "
+                "This may be caused by the 'guidance scale' parameter, which doubles the runtime batch size in Stable Diffusion. "
+                "Adjust the batch size during compilation or modify the 'guidance scale' to match the compiled batch size.\n\n"
+                "For details, see: https://docs.rbln.ai/software/optimum/model_api.html#stable-diffusion"
+            )
+
         added_cond_kwargs = {} if added_cond_kwargs is None else added_cond_kwargs
         if self.use_encoder_hidden_states:
             output = super().forward(

optimum/rbln/diffusers/models/transformers/__init__.py

@@ -0,0 +1,24 @@
+# Copyright 2024 Rebellions Inc.
+
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+
+#     http://www.apache.org/licenses/LICENSE-2.0
+
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+# Portions of this software are licensed under the Apache License,
+# Version 2.0. See the NOTICE file distributed with this work for
+# additional information regarding copyright ownership.
+
+# All other portions of this software, including proprietary code,
+# are the intellectual property of Rebellions Inc. and may not be
+# copied, modified, or distributed without prior written permission
+# from Rebellions Inc.
+
+from .transformer_sd3 import RBLNSD3Transformer2DModel

optimum/rbln/diffusers/models/transformers/transformer_sd3.py

@@ -0,0 +1,203 @@
+# Copyright 2024 Rebellions Inc.
+
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+
+#     http://www.apache.org/licenses/LICENSE-2.0
+
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+# Portions of this software are licensed under the Apache License,
+# Version 2.0. See the NOTICE file distributed with this work for
+# additional information regarding copyright ownership.
+
+# All other portions of this software, including proprietary code,
+# are the intellectual property of Rebellions Inc. and may not be
+# copied, modified, or distributed without prior written permission
+# from Rebellions Inc.
+
+import logging
+from typing import TYPE_CHECKING, Any, Dict, List, Optional, Union
+
+import torch
+from diffusers.models.modeling_outputs import Transformer2DModelOutput
+from diffusers.models.transformers.transformer_sd3 import SD3Transformer2DModel
+from transformers import PretrainedConfig
+
+from ....modeling import RBLNModel
+from ....modeling_config import RBLNCompileConfig, RBLNConfig
+from ....modeling_diffusers import RBLNDiffusionMixin
+
+
+if TYPE_CHECKING:
+    from transformers import AutoFeatureExtractor, AutoProcessor, AutoTokenizer
+
+logger = logging.getLogger(__name__)
+
+
+class SD3Transformer2DModelWrapper(torch.nn.Module):
+    def __init__(self, model: "SD3Transformer2DModel") -> None:
+        super().__init__()
+        self.model = model
+
+    def forward(
+        self,
+        hidden_states: torch.FloatTensor,
+        encoder_hidden_states: torch.FloatTensor = None,
+        pooled_projections: torch.FloatTensor = None,
+        timestep: torch.LongTensor = None,
+        # need controlnet support?
+        block_controlnet_hidden_states: List = None,
+        joint_attention_kwargs: Optional[Dict[str, Any]] = None,
+        return_dict: bool = True,
+    ):
+        return self.model(
+            hidden_states=hidden_states,
+            encoder_hidden_states=encoder_hidden_states,
+            pooled_projections=pooled_projections,
+            timestep=timestep,
+            return_dict=False,
+        )
+
+
+class RBLNSD3Transformer2DModel(RBLNModel):
+    hf_library_name = "diffusers"
+
+    def __post_init__(self, **kwargs):
+        super().__post_init__(**kwargs)
+
+    @classmethod
+    def wrap_model_if_needed(cls, model: torch.nn.Module, rbln_config: RBLNConfig) -> torch.nn.Module:
+        return SD3Transformer2DModelWrapper(model).eval()
+
+    @classmethod
+    def update_rbln_config_using_pipe(cls, pipe: RBLNDiffusionMixin, rbln_config: Dict[str, Any]) -> Dict[str, Any]:
+        sample_size = rbln_config.get("sample_size", pipe.default_sample_size)
+        img_width = rbln_config.get("img_width")
+        img_height = rbln_config.get("img_height")
+
+        if (img_width is None) ^ (img_height is None):
+            raise RuntimeError
+
+        elif img_width and img_height:
+            sample_size = img_height // pipe.vae_scale_factor, img_width // pipe.vae_scale_factor
+
+        prompt_max_length = rbln_config.get("max_sequence_length", 256)
+        prompt_embed_length = pipe.tokenizer_max_length + prompt_max_length
+
+        batch_size = rbln_config.get("batch_size")
+        if not batch_size:
+            do_classifier_free_guidance = rbln_config.get("guidance_scale", 5.0) > 1.0
+            batch_size = 2 if do_classifier_free_guidance else 1
+        else:
+            if rbln_config.get("guidance_scale"):
+                logger.warning(
+                    "guidance_scale is ignored because batch size is explicitly specified. "
+                    "To ensure consistent behavior, consider removing the guidance scale or "
+                    "adjusting the batch size configuration as needed."
+                )
+
+        rbln_config.update(
+            {
+                "batch_size": batch_size,
+                "prompt_embed_length": prompt_embed_length,
+                "sample_size": sample_size,
+            }
+        )
+
+        return rbln_config
+
+    @classmethod
+    def _get_rbln_config(
+        cls,
+        preprocessors: Union["AutoFeatureExtractor", "AutoProcessor", "AutoTokenizer"],
+        model_config: "PretrainedConfig",
+        rbln_kwargs: Dict[str, Any] = {},
+    ) -> RBLNConfig:
+        rbln_batch_size = rbln_kwargs.get("batch_size", None)
+
+        sample_size = rbln_kwargs.get("sample_size", model_config.sample_size)
+        if isinstance(sample_size, int):
+            sample_size = (sample_size, sample_size)
+
+        rbln_prompt_embed_length = rbln_kwargs.get("prompt_embed_length")
+        if rbln_prompt_embed_length is None:
+            raise ValueError("rbln_prompt_embed_length should be specified.")
+
+        input_info = [
+            (
+                "hidden_states",
+                [
+                    rbln_batch_size,
+                    model_config.in_channels,
+                    sample_size[0],
+                    sample_size[1],
+                ],
+                "float32",
+            ),
+            (
+                "encoder_hidden_states",
+                [
+                    rbln_batch_size,
+                    rbln_prompt_embed_length,
+                    model_config.joint_attention_dim,
+                ],
+                "float32",
+            ),
+            (
+                "pooled_projections",
+                [
+                    rbln_batch_size,
+                    model_config.pooled_projection_dim,
+                ],
+                "float32",
+            ),
+            ("timestep", [rbln_batch_size], "float32"),
+        ]
+
+        rbln_compile_config = RBLNCompileConfig(input_info=input_info)
+
+        rbln_config = RBLNConfig(
+            rbln_cls=cls.__name__,
+            compile_cfgs=[rbln_compile_config],
+            rbln_kwargs=rbln_kwargs,
+        )
+
+        rbln_config.model_cfg.update({"batch_size": rbln_batch_size})
+
+        return rbln_config
+
+    @property
+    def compiled_batch_size(self):
+        return self.rbln_config.compile_cfgs[0].input_info[0][1][0]
+
+    def forward(
+        self,
+        hidden_states: torch.FloatTensor,
+        encoder_hidden_states: torch.FloatTensor = None,
+        pooled_projections: torch.FloatTensor = None,
+        timestep: torch.LongTensor = None,
+        block_controlnet_hidden_states: List = None,
+        joint_attention_kwargs: Optional[Dict[str, Any]] = None,
+        return_dict: bool = True,
+        **kwargs,
+    ):
+        sample_batch_size = hidden_states.size()[0]
+        compiled_batch_size = self.compiled_batch_size
+        if sample_batch_size != compiled_batch_size and (
+            sample_batch_size * 2 == compiled_batch_size or sample_batch_size == compiled_batch_size * 2
+        ):
+            raise ValueError(
+                f"Mismatch between Transformers' runtime batch size ({sample_batch_size}) and compiled batch size ({compiled_batch_size}). "
+                "This may be caused by the 'guidance scale' parameter, which doubles the runtime batch size in Stable Diffusion. "
+                "Adjust the batch size during compilation or modify the 'guidance scale' to match the compiled batch size.\n\n"
+                "For details, see: https://docs.rbln.ai/software/optimum/model_api.html#stable-diffusion"
+            )
+
+        sample = super().forward(hidden_states, encoder_hidden_states, pooled_projections, timestep)
+        return Transformer2DModelOutput(sample=sample)

optimum/rbln/diffusers/models/unets/__init__.py

@@ -0,0 +1,24 @@
+# Copyright 2024 Rebellions Inc.
+
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at:
+
+#     http://www.apache.org/licenses/LICENSE-2.0
+
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+# Portions of this software are licensed under the Apache License,
+# Version 2.0. See the NOTICE file distributed with this work for
+# additional information regarding copyright ownership.
+
+# All other portions of this software, including proprietary code,
+# are the intellectual property of Rebellions Inc. and may not be
+# copied, modified, or distributed without prior written permission
+# from Rebellions Inc.
+
+from .unet_2d_condition import RBLNUNet2DConditionModel