ai-edge-torch-nightly 0.2.0.dev20240714__py3-none-any.whl

ai_edge_torch/generative/examples/stable_diffusion/diffusion.py

@@ -0,0 +1,573 @@
+# Copyright 2024 The AI Edge Torch Authors.
+#
+# 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.
+# ==============================================================================
+
+import torch
+from torch import nn
+
+import ai_edge_torch.generative.layers.builder as layers_builder
+import ai_edge_torch.generative.layers.model_config as layers_cfg
+import ai_edge_torch.generative.layers.unet.blocks_2d as blocks_2d
+import ai_edge_torch.generative.layers.unet.model_config as unet_cfg
+import ai_edge_torch.generative.utilities.stable_diffusion_loader as stable_diffusion_loader
+
+_down_encoder_blocks_tensor_names = [
+    stable_diffusion_loader.DownEncoderBlockTensorNames(
+        residual_block_tensor_names=[
+            stable_diffusion_loader.ResidualBlockTensorNames(
+                norm_1=f"model.diffusion_model.input_blocks.{i*3+j+1}.0.in_layers.0",
+                conv_1=f"model.diffusion_model.input_blocks.{i*3+j+1}.0.in_layers.2",
+                norm_2=f"model.diffusion_model.input_blocks.{i*3+j+1}.0.out_layers.0",
+                conv_2=f"model.diffusion_model.input_blocks.{i*3+j+1}.0.out_layers.3",
+                time_embedding=f"model.diffusion_model.input_blocks.{i*3+j+1}.0.emb_layers.1",
+                residual_layer=f"model.diffusion_model.input_blocks.{i*3+j+1}.0.skip_connection"
+                if (i * 3 + j + 1) in [4, 7]
+                else None,
+            )
+            for j in range(2)
+        ],
+        transformer_block_tensor_names=[
+            stable_diffusion_loader.TransformerBlockTensorNames(
+                pre_conv_norm=f"model.diffusion_model.input_blocks.{i*3+j+1}.1.norm",
+                conv_in=f"model.diffusion_model.input_blocks.{i*3+j+1}.1.proj_in",
+                conv_out=f"model.diffusion_model.input_blocks.{i*3+j+1}.1.proj_out",
+                self_attention=stable_diffusion_loader.AttentionBlockTensorNames(
+                    norm=f"model.diffusion_model.input_blocks.{i*3+j+1}.1.transformer_blocks.0.norm1",
+                    q_proj=f"model.diffusion_model.input_blocks.{i*3+j+1}.1.transformer_blocks.0.attn1.to_q",
+                    k_proj=f"model.diffusion_model.input_blocks.{i*3+j+1}.1.transformer_blocks.0.attn1.to_k",
+                    v_proj=f"model.diffusion_model.input_blocks.{i*3+j+1}.1.transformer_blocks.0.attn1.to_v",
+                    output_proj=f"model.diffusion_model.input_blocks.{i*3+j+1}.1.transformer_blocks.0.attn1.to_out.0",
+                ),
+                cross_attention=stable_diffusion_loader.CrossAttentionBlockTensorNames(
+                    norm=f"model.diffusion_model.input_blocks.{i*3+j+1}.1.transformer_blocks.0.norm2",
+                    q_proj=f"model.diffusion_model.input_blocks.{i*3+j+1}.1.transformer_blocks.0.attn2.to_q",
+                    k_proj=f"model.diffusion_model.input_blocks.{i*3+j+1}.1.transformer_blocks.0.attn2.to_k",
+                    v_proj=f"model.diffusion_model.input_blocks.{i*3+j+1}.1.transformer_blocks.0.attn2.to_v",
+                    output_proj=f"model.diffusion_model.input_blocks.{i*3+j+1}.1.transformer_blocks.0.attn2.to_out.0",
+                ),
+                feed_forward=stable_diffusion_loader.FeedForwardBlockTensorNames(
+                    norm=f"model.diffusion_model.input_blocks.{i*3+j+1}.1.transformer_blocks.0.norm3",
+                    ge_glu=f"model.diffusion_model.input_blocks.{i*3+j+1}.1.transformer_blocks.0.ff.net.0.proj",
+                    w2=f"model.diffusion_model.input_blocks.{i*3+j+1}.1.transformer_blocks.0.ff.net.2",
+                ),
+            )
+            for j in range(2)
+        ]
+        if i < 3
+        else None,
+        downsample_conv=f"model.diffusion_model.input_blocks.{i*3+3}.0.op"
+        if i < 3
+        else None,
+    )
+    for i in range(4)
+]
+
+_mid_block_tensor_names = stable_diffusion_loader.MidBlockTensorNames(
+    residual_block_tensor_names=[
+        stable_diffusion_loader.ResidualBlockTensorNames(
+            norm_1=f"model.diffusion_model.middle_block.{i}.in_layers.0",
+            conv_1=f"model.diffusion_model.middle_block.{i}.in_layers.2",
+            norm_2=f"model.diffusion_model.middle_block.{i}.out_layers.0",
+            conv_2=f"model.diffusion_model.middle_block.{i}.out_layers.3",
+            time_embedding=f"model.diffusion_model.middle_block.{i}.emb_layers.1",
+        )
+        for i in [0, 2]
+    ],
+    transformer_block_tensor_names=[
+        stable_diffusion_loader.TransformerBlockTensorNames(
+            pre_conv_norm=f"model.diffusion_model.middle_block.{i}.norm",
+            conv_in=f"model.diffusion_model.middle_block.{i}.proj_in",
+            conv_out=f"model.diffusion_model.middle_block.{i}.proj_out",
+            self_attention=stable_diffusion_loader.AttentionBlockTensorNames(
+                norm=f"model.diffusion_model.middle_block.{i}.transformer_blocks.0.norm1",
+                q_proj=f"model.diffusion_model.middle_block.{i}.transformer_blocks.0.attn1.to_q",
+                k_proj=f"model.diffusion_model.middle_block.{i}.transformer_blocks.0.attn1.to_k",
+                v_proj=f"model.diffusion_model.middle_block.{i}.transformer_blocks.0.attn1.to_v",
+                output_proj=f"model.diffusion_model.middle_block.{i}.transformer_blocks.0.attn1.to_out.0",
+            ),
+            cross_attention=stable_diffusion_loader.CrossAttentionBlockTensorNames(
+                norm=f"model.diffusion_model.middle_block.{i}.transformer_blocks.0.norm2",
+                q_proj=f"model.diffusion_model.middle_block.{i}.transformer_blocks.0.attn2.to_q",
+                k_proj=f"model.diffusion_model.middle_block.{i}.transformer_blocks.0.attn2.to_k",
+                v_proj=f"model.diffusion_model.middle_block.{i}.transformer_blocks.0.attn2.to_v",
+                output_proj=f"model.diffusion_model.middle_block.{i}.transformer_blocks.0.attn2.to_out.0",
+            ),
+            feed_forward=stable_diffusion_loader.FeedForwardBlockTensorNames(
+                norm=f"model.diffusion_model.middle_block.{i}.transformer_blocks.0.norm3",
+                ge_glu=f"model.diffusion_model.middle_block.{i}.transformer_blocks.0.ff.net.0.proj",
+                w2=f"model.diffusion_model.middle_block.{i}.transformer_blocks.0.ff.net.2",
+            ),
+        )
+        for i in [1]
+    ],
+)
+
+_up_decoder_blocks_tensor_names = [
+    stable_diffusion_loader.SkipUpDecoderBlockTensorNames(
+        residual_block_tensor_names=[
+            stable_diffusion_loader.ResidualBlockTensorNames(
+                norm_1=f"model.diffusion_model.output_blocks.{i*3+j}.0.in_layers.0",
+                conv_1=f"model.diffusion_model.output_blocks.{i*3+j}.0.in_layers.2",
+                norm_2=f"model.diffusion_model.output_blocks.{i*3+j}.0.out_layers.0",
+                conv_2=f"model.diffusion_model.output_blocks.{i*3+j}.0.out_layers.3",
+                time_embedding=f"model.diffusion_model.output_blocks.{i*3+j}.0.emb_layers.1",
+                residual_layer=f"model.diffusion_model.output_blocks.{i*3+j}.0.skip_connection",
+            )
+            for j in range(3)
+        ],
+        transformer_block_tensor_names=[
+            stable_diffusion_loader.TransformerBlockTensorNames(
+                pre_conv_norm=f"model.diffusion_model.output_blocks.{i*3+j}.1.norm",
+                conv_in=f"model.diffusion_model.output_blocks.{i*3+j}.1.proj_in",
+                conv_out=f"model.diffusion_model.output_blocks.{i*3+j}.1.proj_out",
+                self_attention=stable_diffusion_loader.AttentionBlockTensorNames(
+                    norm=f"model.diffusion_model.output_blocks.{i*3+j}.1.transformer_blocks.0.norm1",
+                    q_proj=f"model.diffusion_model.output_blocks.{i*3+j}.1.transformer_blocks.0.attn1.to_q",
+                    k_proj=f"model.diffusion_model.output_blocks.{i*3+j}.1.transformer_blocks.0.attn1.to_k",
+                    v_proj=f"model.diffusion_model.output_blocks.{i*3+j}.1.transformer_blocks.0.attn1.to_v",
+                    output_proj=f"model.diffusion_model.output_blocks.{i*3+j}.1.transformer_blocks.0.attn1.to_out.0",
+                ),
+                cross_attention=stable_diffusion_loader.CrossAttentionBlockTensorNames(
+                    norm=f"model.diffusion_model.output_blocks.{i*3+j}.1.transformer_blocks.0.norm2",
+                    q_proj=f"model.diffusion_model.output_blocks.{i*3+j}.1.transformer_blocks.0.attn2.to_q",
+                    k_proj=f"model.diffusion_model.output_blocks.{i*3+j}.1.transformer_blocks.0.attn2.to_k",
+                    v_proj=f"model.diffusion_model.output_blocks.{i*3+j}.1.transformer_blocks.0.attn2.to_v",
+                    output_proj=f"model.diffusion_model.output_blocks.{i*3+j}.1.transformer_blocks.0.attn2.to_out.0",
+                ),
+                feed_forward=stable_diffusion_loader.FeedForwardBlockTensorNames(
+                    norm=f"model.diffusion_model.output_blocks.{i*3+j}.1.transformer_blocks.0.norm3",
+                    ge_glu=f"model.diffusion_model.output_blocks.{i*3+j}.1.transformer_blocks.0.ff.net.0.proj",
+                    w2=f"model.diffusion_model.output_blocks.{i*3+j}.1.transformer_blocks.0.ff.net.2",
+                ),
+            )
+            for j in range(3)
+        ]
+        if i > 0
+        else None,
+        upsample_conv=f"model.diffusion_model.output_blocks.{i*3+2}.2.conv"
+        if 0 < i < 3
+        else (f"model.diffusion_model.output_blocks.2.1.conv" if i == 0 else None),
+    )
+    for i in range(4)
+]
+
+TENSOR_NAMES = stable_diffusion_loader.DiffusionModelLoader.TensorNames(
+    time_embedding=stable_diffusion_loader.TimeEmbeddingTensorNames(
+        w1="model.diffusion_model.time_embed.0",
+        w2="model.diffusion_model.time_embed.2",
+    ),
+    conv_in="model.diffusion_model.input_blocks.0.0",
+    conv_out="model.diffusion_model.out.2",
+    final_norm="model.diffusion_model.out.0",
+    down_encoder_blocks_tensor_names=_down_encoder_blocks_tensor_names,
+    mid_block_tensor_names=_mid_block_tensor_names,
+    up_decoder_blocks_tensor_names=_up_decoder_blocks_tensor_names,
+)
+
+
+class TimeEmbedding(nn.Module):
+
+  def __init__(self, in_dim, out_dim):
+    super().__init__()
+    self.w1 = nn.Linear(in_dim, out_dim)
+    self.w2 = nn.Linear(out_dim, out_dim)
+    self.act = layers_builder.get_activation(
+        layers_cfg.ActivationConfig(layers_cfg.ActivationType.SILU)
+    )
+
+  def forward(self, x: torch.Tensor):
+    return self.w2(self.act(self.w1(x)))
+
+
+class Diffusion(nn.Module):
+  """The Diffusion model used in Stable Diffusion.
+
+    For details, see https://arxiv.org/abs/2103.00020
+
+    Sturcture of the Diffusion model:
+
+                         latents        text context   time embed
+                            │                 │            │
+                            │                 │            │
+                  ┌─────────▼─────────┐       │  ┌─────────▼─────────┐
+                  │      ConvIn       │       │  │   Time Embedding  │
+                  └─────────┬─────────┘       │  └─────────┬─────────┘
+                            │                 │            │
+                  ┌─────────▼─────────┐       │            │
+           ┌──────┤   DownEncoder2D   │ ◄─────┼────────────┤
+           │      └─────────┬─────────┘ x 4   │            │
+           │                │                 │            │
+           │      ┌─────────▼─────────┐       │            │
+  skip connection │     MidBlock2D    │ ◄─────┼────────────┤
+           │      └─────────┬─────────┘       │            │
+           │                │                 │            │
+           │      ┌─────────▼─────────┐       │            │
+           └──────►  SkipUpDecoder2D  │ ◄─────┴────────────┘
+                  └─────────┬─────────┘ x 4
+                            │
+                  ┌─────────▼─────────┐
+                  │     FinalNorm     │
+                  └─────────┬─────────┘
+                            │
+                  ┌─────────▼─────────┐
+                  │    Activation     │
+                  └─────────┬─────────┘
+                            │
+                  ┌─────────▼─────────┐
+                  │      ConvOut      │
+                  └─────────┬─────────┘
+                            │
+                            ▼
+                      output image
+  """
+
+  def __init__(self, config: unet_cfg.DiffusionModelConfig):
+    super().__init__()
+
+    self.config = config
+    block_out_channels = config.block_out_channels
+    reversed_block_out_channels = list(reversed(block_out_channels))
+
+    time_embedding_blocks_dim = config.time_embedding_blocks_dim
+    self.time_embedding = TimeEmbedding(
+        config.time_embedding_dim, config.time_embedding_blocks_dim
+    )
+
+    self.conv_in = nn.Conv2d(
+        config.in_channels, block_out_channels[0], kernel_size=3, padding=1
+    )
+
+    attention_config = layers_cfg.AttentionConfig(
+        num_heads=config.transformer_num_attention_heads,
+        num_query_groups=config.transformer_num_attention_heads,
+        rotary_percentage=0.0,
+        qkv_transpose_before_split=True,
+        qkv_use_bias=False,
+        output_proj_use_bias=True,
+        enable_kv_cache=False,
+        qkv_fused_interleaved=False,
+    )
+
+    # Down encoders.
+    down_encoders = []
+    output_channel = block_out_channels[0]
+    for i, block_out_channel in enumerate(block_out_channels):
+      input_channel = output_channel
+      output_channel = block_out_channel
+      not_final_block = i < len(block_out_channels) - 1
+      if not_final_block:
+        down_encoders.append(
+            blocks_2d.DownEncoderBlock2D(
+                unet_cfg.DownEncoderBlock2DConfig(
+                    in_channels=input_channel,
+                    out_channels=output_channel,
+                    normalization_config=config.residual_norm_config,
+                    activation_config=layers_cfg.ActivationConfig(
+                        config.residual_activation_type
+                    ),
+                    num_layers=config.layers_per_block,
+                    padding=config.downsample_padding,
+                    time_embedding_channels=time_embedding_blocks_dim,
+                    add_downsample=True,
+                    sampling_config=unet_cfg.DownSamplingConfig(
+                        mode=unet_cfg.SamplingType.CONVOLUTION,
+                        in_channels=output_channel,
+                        out_channels=output_channel,
+                        kernel_size=3,
+                        stride=2,
+                        padding=config.downsample_padding,
+                    ),
+                    transformer_block_config=unet_cfg.TransformerBlock2DConfig(
+                        attention_block_config=unet_cfg.AttentionBlock2DConfig(
+                            dim=output_channel,
+                            attention_batch_size=config.transformer_batch_size,
+                            normalization_config=config.transformer_norm_config,
+                            attention_config=attention_config,
+                        ),
+                        cross_attention_block_config=unet_cfg.CrossAttentionBlock2DConfig(
+                            query_dim=output_channel,
+                            cross_dim=config.transformer_cross_attention_dim,
+                            attention_batch_size=config.transformer_batch_size,
+                            normalization_config=config.transformer_norm_config,
+                            attention_config=attention_config,
+                        ),
+                        pre_conv_normalization_config=config.transformer_pre_conv_norm_config,
+                        feed_forward_block_config=unet_cfg.FeedForwardBlock2DConfig(
+                            dim=output_channel,
+                            hidden_dim=output_channel * 4,
+                            normalization_config=config.transformer_norm_config,
+                            activation_config=layers_cfg.ActivationConfig(
+                                type=config.transformer_ff_activation_type,
+                                dim_in=output_channel,
+                                dim_out=output_channel * 4,
+                            ),
+                            use_bias=True,
+                        ),
+                    ),
+                )
+            )
+        )
+      else:
+        down_encoders.append(
+            blocks_2d.DownEncoderBlock2D(
+                unet_cfg.DownEncoderBlock2DConfig(
+                    in_channels=input_channel,
+                    out_channels=output_channel,
+                    normalization_config=config.residual_norm_config,
+                    activation_config=layers_cfg.ActivationConfig(
+                        config.residual_activation_type
+                    ),
+                    num_layers=config.layers_per_block,
+                    padding=config.downsample_padding,
+                    time_embedding_channels=time_embedding_blocks_dim,
+                    add_downsample=False,
+                )
+            )
+        )
+    self.down_encoders = nn.ModuleList(down_encoders)
+
+    # Mid block.
+    mid_block_channels = block_out_channels[-1]
+    self.mid_block = blocks_2d.MidBlock2D(
+        unet_cfg.MidBlock2DConfig(
+            in_channels=block_out_channels[-1],
+            normalization_config=config.residual_norm_config,
+            activation_config=layers_cfg.ActivationConfig(
+                config.residual_activation_type
+            ),
+            num_layers=config.mid_block_layers,
+            time_embedding_channels=config.time_embedding_blocks_dim,
+            transformer_block_config=unet_cfg.TransformerBlock2DConfig(
+                attention_block_config=unet_cfg.AttentionBlock2DConfig(
+                    dim=mid_block_channels,
+                    attention_batch_size=config.transformer_batch_size,
+                    normalization_config=config.transformer_norm_config,
+                    attention_config=attention_config,
+                ),
+                cross_attention_block_config=unet_cfg.CrossAttentionBlock2DConfig(
+                    query_dim=mid_block_channels,
+                    cross_dim=config.transformer_cross_attention_dim,
+                    attention_batch_size=config.transformer_batch_size,
+                    normalization_config=config.transformer_norm_config,
+                    attention_config=attention_config,
+                ),
+                pre_conv_normalization_config=config.transformer_pre_conv_norm_config,
+                feed_forward_block_config=unet_cfg.FeedForwardBlock2DConfig(
+                    dim=mid_block_channels,
+                    hidden_dim=mid_block_channels * 4,
+                    normalization_config=config.transformer_norm_config,
+                    activation_config=layers_cfg.ActivationConfig(
+                        type=config.transformer_ff_activation_type,
+                        dim_in=mid_block_channels,
+                        dim_out=mid_block_channels * 4,
+                    ),
+                    use_bias=True,
+                ),
+            ),
+        )
+    )
+
+    # Up decoders.
+    up_decoders = []
+    up_decoder_layers_per_block = config.layers_per_block + 1
+    output_channel = reversed_block_out_channels[0]
+    for i, block_out_channel in enumerate(reversed_block_out_channels):
+      prev_out_channel = output_channel
+      output_channel = block_out_channel
+      input_channel = reversed_block_out_channels[
+          min(i + 1, len(reversed_block_out_channels) - 1)
+      ]
+      not_final_block = i < len(reversed_block_out_channels) - 1
+      not_first_block = i != 0
+      if not_first_block:
+        up_decoders.append(
+            blocks_2d.SkipUpDecoderBlock2D(
+                unet_cfg.SkipUpDecoderBlock2DConfig(
+                    in_channels=input_channel,
+                    out_channels=output_channel,
+                    prev_out_channels=prev_out_channel,
+                    normalization_config=config.residual_norm_config,
+                    activation_config=layers_cfg.ActivationConfig(
+                        config.residual_activation_type
+                    ),
+                    num_layers=up_decoder_layers_per_block,
+                    time_embedding_channels=time_embedding_blocks_dim,
+                    add_upsample=not_final_block,
+                    upsample_conv=True,
+                    sampling_config=unet_cfg.UpSamplingConfig(
+                        mode=unet_cfg.SamplingType.NEAREST,
+                        scale_factor=2,
+                    ),
+                    transformer_block_config=unet_cfg.TransformerBlock2DConfig(
+                        attention_block_config=unet_cfg.AttentionBlock2DConfig(
+                            dim=output_channel,
+                            attention_batch_size=config.transformer_batch_size,
+                            normalization_config=config.transformer_norm_config,
+                            attention_config=attention_config,
+                        ),
+                        cross_attention_block_config=unet_cfg.CrossAttentionBlock2DConfig(
+                            query_dim=output_channel,
+                            cross_dim=config.transformer_cross_attention_dim,
+                            attention_batch_size=config.transformer_batch_size,
+                            normalization_config=config.transformer_norm_config,
+                            attention_config=attention_config,
+                        ),
+                        pre_conv_normalization_config=config.transformer_pre_conv_norm_config,
+                        feed_forward_block_config=unet_cfg.FeedForwardBlock2DConfig(
+                            dim=output_channel,
+                            hidden_dim=output_channel * 4,
+                            normalization_config=config.transformer_norm_config,
+                            activation_config=layers_cfg.ActivationConfig(
+                                type=config.transformer_ff_activation_type,
+                                dim_in=output_channel,
+                                dim_out=output_channel * 4,
+                            ),
+                            use_bias=True,
+                        ),
+                    ),
+                )
+            )
+        )
+      else:
+        up_decoders.append(
+            blocks_2d.SkipUpDecoderBlock2D(
+                unet_cfg.SkipUpDecoderBlock2DConfig(
+                    in_channels=input_channel,
+                    out_channels=output_channel,
+                    prev_out_channels=prev_out_channel,
+                    normalization_config=config.residual_norm_config,
+                    activation_config=layers_cfg.ActivationConfig(
+                        config.residual_activation_type
+                    ),
+                    num_layers=up_decoder_layers_per_block,
+                    time_embedding_channels=time_embedding_blocks_dim,
+                    add_upsample=not_final_block,
+                    upsample_conv=True,
+                    sampling_config=unet_cfg.UpSamplingConfig(
+                        mode=unet_cfg.SamplingType.NEAREST, scale_factor=2
+                    ),
+                )
+            )
+        )
+    self.up_decoders = nn.ModuleList(up_decoders)
+
+    self.final_norm = layers_builder.build_norm(
+        reversed_block_out_channels[-1], config.final_norm_config
+    )
+    self.final_act = layers_builder.get_activation(
+        layers_cfg.ActivationConfig(config.final_activation_type)
+    )
+    self.conv_out = nn.Conv2d(
+        reversed_block_out_channels[-1], config.out_channels, kernel_size=3, padding=1
+    )
+
+  @torch.inference_mode
+  def forward(
+      self, latents: torch.Tensor, context: torch.Tensor, time_emb: torch.Tensor
+  ) -> torch.Tensor:
+    """Forward function of diffusion model.
+
+    Args:
+        latents (torch.Tensor): latents space tensor.
+        context (torch.Tensor): context tensor from CLIP text encoder.
+        time_emb (torch.Tensor): the time embedding tensor.
+
+    Returns:
+        output latents from diffusion model.
+    """
+    time_emb = self.time_embedding(time_emb)
+    x = self.conv_in(latents)
+    skip_connection_tensors = [x]
+    for encoder in self.down_encoders:
+      x, hidden_states = encoder(x, time_emb, context, output_hidden_states=True)
+      skip_connection_tensors.extend(hidden_states)
+    x = self.mid_block(x, time_emb, context)
+    for decoder in self.up_decoders:
+      encoder_tensors = [
+          skip_connection_tensors.pop() for i in range(self.config.layers_per_block + 1)
+      ]
+      x = decoder(x, encoder_tensors, time_emb, context)
+    x = self.final_norm(x)
+    x = self.final_act(x)
+    x = self.conv_out(x)
+    return x
+
+
+def get_model_config(batch_size: int) -> unet_cfg.DiffusionModelConfig:
+  """Get configs for the Diffusion model of Stable Diffusion v1.5
+
+  Args:
+    batch_size (int): the batch size of input.
+
+  Retruns:
+    The configuration of diffusion model of Stable Diffusion v1.5.
+
+  """
+  in_channels = 4
+  out_channels = 4
+  block_out_channels = [320, 640, 1280, 1280]
+  layers_per_block = 2
+  downsample_padding = 1
+
+  # Residual configs.
+  residual_norm_config = layers_cfg.NormalizationConfig(
+      layers_cfg.NormalizationType.GROUP_NORM, group_num=32
+  )
+  residual_activation_type = layers_cfg.ActivationType.SILU
+
+  # Transformer configs.
+  transformer_num_attention_heads = 8
+  transformer_batch_size = batch_size
+  transformer_cross_attention_dim = 768  # Embedding fomr CLIP model
+  transformer_pre_conv_norm_config = layers_cfg.NormalizationConfig(
+      layers_cfg.NormalizationType.GROUP_NORM, epsilon=1e-6, group_num=32
+  )
+  transformer_norm_config = layers_cfg.NormalizationConfig(
+      layers_cfg.NormalizationType.LAYER_NORM
+  )
+  transformer_ff_activation_type = layers_cfg.ActivationType.GE_GLU
+
+  # Time embedding configs.
+  time_embedding_dim = 320
+  time_embedding_blocks_dim = 1280
+
+  # Mid block configs.
+  mid_block_layers = 1
+
+  # Finaly layer configs.
+  final_norm_config = layers_cfg.NormalizationConfig(
+      layers_cfg.NormalizationType.GROUP_NORM, group_num=32
+  )
+  final_activation_type = layers_cfg.ActivationType.SILU
+
+  return unet_cfg.DiffusionModelConfig(
+      in_channels=in_channels,
+      out_channels=out_channels,
+      block_out_channels=block_out_channels,
+      layers_per_block=layers_per_block,
+      downsample_padding=downsample_padding,
+      residual_norm_config=residual_norm_config,
+      residual_activation_type=residual_activation_type,
+      transformer_batch_size=transformer_batch_size,
+      transformer_num_attention_heads=transformer_num_attention_heads,
+      transformer_cross_attention_dim=transformer_cross_attention_dim,
+      transformer_pre_conv_norm_config=transformer_pre_conv_norm_config,
+      transformer_norm_config=transformer_norm_config,
+      transformer_ff_activation_type=transformer_ff_activation_type,
+      mid_block_layers=mid_block_layers,
+      time_embedding_dim=time_embedding_dim,
+      time_embedding_blocks_dim=time_embedding_blocks_dim,
+      final_norm_config=final_norm_config,
+      final_activation_type=final_activation_type,
+  )

ai_edge_torch/generative/examples/stable_diffusion/encoder.py

@@ -0,0 +1,118 @@
+# Copyright 2024 The AI Edge Torch Authors.
+#
+# 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.
+# ==============================================================================
+
+import torch
+from torch import nn
+from torch.nn import functional as F
+
+from ai_edge_torch.generative.examples.stable_diffusion.attention import SelfAttention  # NOQA
+
+
+class AttentionBlock(nn.Module):
+
+  def __init__(self, channels):
+    super().__init__()
+    self.groupnorm = nn.GroupNorm(32, channels)
+    self.attention = SelfAttention(1, channels)
+
+  def forward(self, x):
+    residue = x
+    x = self.groupnorm(x)
+
+    n, c, h, w = x.shape
+    x = x.view((n, c, h * w))
+    x = x.transpose(-1, -2)
+    x = self.attention(x)
+    x = x.transpose(-1, -2)
+    x = x.view((n, c, h, w))
+
+    x += residue
+    return x
+
+
+class ResidualBlock(nn.Module):
+
+  def __init__(self, in_channels, out_channels):
+    super().__init__()
+    self.groupnorm_1 = nn.GroupNorm(32, in_channels)
+    self.conv_1 = nn.Conv2d(in_channels, out_channels, kernel_size=3, padding=1)
+
+    self.groupnorm_2 = nn.GroupNorm(32, out_channels)
+    self.conv_2 = nn.Conv2d(out_channels, out_channels, kernel_size=3, padding=1)
+
+    if in_channels == out_channels:
+      self.residual_layer = nn.Identity()
+    else:
+      self.residual_layer = nn.Conv2d(
+          in_channels, out_channels, kernel_size=1, padding=0
+      )
+
+  def forward(self, x):
+    residue = x
+
+    x = self.groupnorm_1(x)
+    x = F.silu(x)
+    x = self.conv_1(x)
+
+    x = self.groupnorm_2(x)
+    x = F.silu(x)
+    x = self.conv_2(x)
+
+    return x + self.residual_layer(residue)
+
+
+class Encoder(nn.Sequential):
+
+  def __init__(self):
+    super().__init__(
+        nn.Conv2d(3, 128, kernel_size=3, padding=1),
+        ResidualBlock(128, 128),
+        ResidualBlock(128, 128),
+        nn.Conv2d(128, 128, kernel_size=3, stride=2, padding=0),
+        ResidualBlock(128, 256),
+        ResidualBlock(256, 256),
+        nn.Conv2d(256, 256, kernel_size=3, stride=2, padding=0),
+        ResidualBlock(256, 512),
+        ResidualBlock(512, 512),
+        nn.Conv2d(512, 512, kernel_size=3, stride=2, padding=0),
+        ResidualBlock(512, 512),
+        ResidualBlock(512, 512),
+        ResidualBlock(512, 512),
+        AttentionBlock(512),
+        ResidualBlock(512, 512),
+        nn.GroupNorm(32, 512),
+        nn.SiLU(),
+        nn.Conv2d(512, 8, kernel_size=3, padding=1),
+        nn.Conv2d(8, 8, kernel_size=1, padding=0),
+    )
+
+  @torch.inference_mode
+  def forward(self, x, noise):
+    for module in self:
+      if getattr(module, 'stride', None) == (
+          2,
+          2,
+      ):  # Padding at downsampling should be asymmetric (see #8)
+        x = F.pad(x, (0, 1, 0, 1))
+      x = module(x)
+
+    mean, log_variance = torch.chunk(x, 2, dim=1)
+    log_variance = torch.clamp(log_variance, -30, 20)
+    variance = log_variance.exp()
+    stdev = variance.sqrt()
+    x = mean + stdev * noise
+
+    x *= 0.18215
+    return x