Buckets:
| # SequentialPipelineBlocks | |
| [SequentialPipelineBlocks](/docs/diffusers/v0.43.0/id/api/modular_diffusers/pipeline_blocks#diffusers.SequentialPipelineBlocks) adalah tipe multi-blok yang menyusun beberapa [ModularPipelineBlocks](/docs/diffusers/v0.43.0/id/api/modular_diffusers/pipeline_blocks#diffusers.ModularPipelineBlocks) lainnya secara berurutan. Data mengalir secara linear dari satu blok ke blok berikutnya menggunakan `inputs` dan `intermediate_outputs`. Setiap blok dalam [SequentialPipelineBlocks](/docs/diffusers/v0.43.0/id/api/modular_diffusers/pipeline_blocks#diffusers.SequentialPipelineBlocks) biasanya mewakili satu langkah dalam pipeline, dan dengan menggabungkannya, Anda secara bertahap membangun sebuah pipeline. | |
| Panduan ini menunjukkan cara menghubungkan dua blok menjadi [SequentialPipelineBlocks](/docs/diffusers/v0.43.0/id/api/modular_diffusers/pipeline_blocks#diffusers.SequentialPipelineBlocks). | |
| Buat dua [ModularPipelineBlocks](/docs/diffusers/v0.43.0/id/api/modular_diffusers/pipeline_blocks#diffusers.ModularPipelineBlocks). Blok pertama, `InputBlock`, menghasilkan nilai `batch_size` dan blok kedua, `ImageEncoderBlock` menggunakan `batch_size` sebagai `inputs`. | |
| ```py | |
| from diffusers.modular_pipelines import ModularPipelineBlocks, InputParam, OutputParam | |
| class InputBlock(ModularPipelineBlocks): | |
| @property | |
| def inputs(self): | |
| return [ | |
| InputParam(name="prompt", type_hint=list, description="list of text prompts"), | |
| InputParam(name="num_images_per_prompt", type_hint=int, description="number of images per prompt"), | |
| ] | |
| @property | |
| def intermediate_outputs(self): | |
| return [ | |
| OutputParam(name="batch_size", description="calculated batch size"), | |
| ] | |
| @property | |
| def description(self): | |
| return "A block that determines batch_size based on the number of prompts and num_images_per_prompt argument." | |
| def __call__(self, components, state): | |
| block_state = self.get_block_state(state) | |
| batch_size = len(block_state.prompt) | |
| block_state.batch_size = batch_size * block_state.num_images_per_prompt | |
| self.set_block_state(state, block_state) | |
| return components, state | |
| ``` | |
| ```py | |
| import torch | |
| from diffusers.modular_pipelines import ModularPipelineBlocks, InputParam, OutputParam | |
| class ImageEncoderBlock(ModularPipelineBlocks): | |
| @property | |
| def inputs(self): | |
| return [ | |
| InputParam(name="image", type_hint="PIL.Image", description="raw input image to process"), | |
| InputParam(name="batch_size", type_hint=int), | |
| ] | |
| @property | |
| def intermediate_outputs(self): | |
| return [ | |
| OutputParam(name="image_latents", description="latents representing the image"), | |
| ] | |
| @property | |
| def description(self): | |
| return "Encode raw image into its latent presentation" | |
| def __call__(self, components, state): | |
| block_state = self.get_block_state(state) | |
| # Simulate processing the image | |
| # This will change the state of the image from a PIL image to a tensor for all blocks | |
| block_state.image = torch.randn(1, 3, 512, 512) | |
| block_state.batch_size = block_state.batch_size * 2 | |
| block_state.image_latents = torch.randn(1, 4, 64, 64) | |
| self.set_block_state(state, block_state) | |
| return components, state | |
| ``` | |
| Hubungkan kedua blok tersebut dengan mendefinisikan [SequentialPipelineBlocks](/docs/diffusers/v0.43.0/id/api/modular_diffusers/pipeline_blocks#diffusers.SequentialPipelineBlocks). Daftarkan instance blok di `block_classes` dan nama yang sesuai di `block_names`. Blok-blok tersebut dieksekusi sesuai urutan kemunculannya di `block_classes`, dan data mengalir dari satu blok ke blok berikutnya melalui [PipelineState](/docs/diffusers/v0.43.0/id/api/modular_diffusers/pipeline_states#diffusers.modular_pipelines.PipelineState). | |
| ```py | |
| class ImageProcessingStep(SequentialPipelineBlocks): | |
| """ | |
| # auto_docstring | |
| """ | |
| model_name = "my_model" | |
| block_classes = [InputBlock(), ImageEncoderBlock()] | |
| block_names = ["input", "image_encoder"] | |
| @property | |
| def description(self): | |
| return ( | |
| "Process text prompts and images for the pipeline. It:\n" | |
| " - Determines the batch size from the prompts.\n" | |
| " - Encodes the image into latent space." | |
| ) | |
| ``` | |
| Saat Anda membuat [SequentialPipelineBlocks](/docs/diffusers/v0.43.0/id/api/modular_diffusers/pipeline_blocks#diffusers.SequentialPipelineBlocks), properti seperti `inputs`, `intermediate_outputs`, dan `expected_components` secara otomatis diagregasi dari sub-blok, sehingga tidak perlu mendefinisikannya lagi. | |
| Ada beberapa properti yang perlu Anda atur: | |
| - `description`: Kami merekomendasikan untuk menambahkan deskripsi pada blok yang dirakit untuk menjelaskan apa yang dilakukan oleh langkah gabungan tersebut. | |
| - `model_name`: Ini diturunkan secara otomatis dari sub-blok, tetapi tidak selalu benar, jadi Anda mungkin perlu menimpanya. | |
| - `outputs`: Secara default, ini sama dengan `intermediate_outputs`, tetapi Anda dapat menyetelnya secara manual untuk mengontrol nilai mana yang muncul di dokumen. Ini berguna untuk menampilkan hanya output akhir, bukan semua nilai perantara. | |
| Properti-properti ini, bersama dengan `inputs`, `intermediate_outputs`, dan `expected_components` yang telah diagregasi, digunakan untuk secara otomatis menghasilkan properti `doc`. | |
| Cetak blok `ImageProcessingStep` untuk memeriksa sub-bloknya, dan gunakan `doc` untuk ringkasan lengkap mengenai input, output, dan komponen blok tersebut. | |
| ```py | |
| blocks = ImageProcessingStep() | |
| print(blocks) | |
| print(blocks.doc) | |
| ``` | |
Xet Storage Details
- Size:
- 5.68 kB
- Xet hash:
- 7fb3c3b6894dfb69749c6b0c3714b54fa12dfa55fe4566808b5dbf93cbd931a0
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.