You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
guill 5cfe38f41c
Execution Model Inversion (#2666)
* Execution Model Inversion

This PR inverts the execution model -- from recursively calling nodes to
using a topological sort of the nodes. This change allows for
modification of the node graph during execution. This allows for two
major advantages:

    1. The implementation of lazy evaluation in nodes. For example, if a
    "Mix Images" node has a mix factor of exactly 0.0, the second image
    input doesn't even need to be evaluated (and visa-versa if the mix
    factor is 1.0).

    2. Dynamic expansion of nodes. This allows for the creation of dynamic
    "node groups". Specifically, custom nodes can return subgraphs that
    replace the original node in the graph. This is an incredibly
    powerful concept. Using this functionality, it was easy to
    implement:
        a. Components (a.k.a. node groups)
        b. Flow control (i.e. while loops) via tail recursion
        c. All-in-one nodes that replicate the WebUI functionality
        d. and more
    All of those were able to be implemented entirely via custom nodes,
    so those features are *not* a part of this PR. (There are some
    front-end changes that should occur before that functionality is
    made widely available, particularly around variant sockets.)

The custom nodes associated with this PR can be found at:
https://github.com/BadCafeCode/execution-inversion-demo-comfyui

Note that some of them require that variant socket types ("*") be
enabled.

* Allow `input_info` to be of type `None`

* Handle errors (like OOM) more gracefully

* Add a command-line argument to enable variants

This allows the use of nodes that have sockets of type '*' without
applying a patch to the code.

* Fix an overly aggressive assertion.

This could happen when attempting to evaluate `IS_CHANGED` for a node
during the creation of the cache (in order to create the cache key).

* Fix Pyright warnings

* Add execution model unit tests

* Fix issue with unused literals

Behavior should now match the master branch with regard to undeclared
inputs. Undeclared inputs that are socket connections will be used while
undeclared inputs that are literals will be ignored.

* Make custom VALIDATE_INPUTS skip normal validation

Additionally, if `VALIDATE_INPUTS` takes an argument named `input_types`,
that variable will be a dictionary of the socket type of all incoming
connections. If that argument exists, normal socket type validation will
not occur. This removes the last hurdle for enabling variant types
entirely from custom nodes, so I've removed that command-line option.

I've added appropriate unit tests for these changes.

* Fix example in unit test

This wouldn't have caused any issues in the unit test, but it would have
bugged the UI if someone copy+pasted it into their own node pack.

* Use fstrings instead of '%' formatting syntax

* Use custom exception types.

* Display an error for dependency cycles

Previously, dependency cycles that were created during node expansion
would cause the application to quit (due to an uncaught exception). Now,
we'll throw a proper error to the UI. We also make an attempt to 'blame'
the most relevant node in the UI.

* Add docs on when ExecutionBlocker should be used

* Remove unused functionality

* Rename ExecutionResult.SLEEPING to PENDING

* Remove superfluous function parameter

* Pass None for uneval inputs instead of default

This applies to `VALIDATE_INPUTS`, `check_lazy_status`, and lazy values
in evaluation functions.

* Add a test for mixed node expansion

This test ensures that a node that returns a combination of expanded
subgraphs and literal values functions correctly.

* Raise exception for bad get_node calls.

* Minor refactor of IsChangedCache.get

* Refactor `map_node_over_list` function

* Fix ui output for duplicated nodes

* Add documentation on `check_lazy_status`

* Add file for execution model unit tests

* Clean up Javascript code as per review

* Improve documentation

Converted some comments to docstrings as per review

* Add a new unit test for mixed lazy results

This test validates that when an output list is fed to a lazy node, the
node will properly evaluate previous nodes that are needed by any inputs
to the lazy node.

No code in the execution model has been changed. The test already
passes.

* Allow kwargs in VALIDATE_INPUTS functions

When kwargs are used, validation is skipped for all inputs as if they
had been mentioned explicitly.

* List cached nodes in `execution_cached` message

This was previously just bugged in this PR.
6 months ago
..
cldm "auto" type is only relevant to the SetUnionControlNetType node. 7 months ago
extra_samplers Cleanup uni_pc inpainting. 1 year ago
k_diffusion Switch to the real cfg++ method in the samplers. 8 months ago
ldm Fix issue. 6 months ago
sd1_tokenizer Initial commit. 2 years ago
t2i_adapter Controlnet refactor. 8 months ago
taesd Improvements to the TAESD3 implementation. 8 months ago
text_encoders Correct spelling 'token_weight_pars_t5' to 'token_weight_pairs_t5' (#4200) 7 months ago
caching.py Execution Model Inversion (#2666) 6 months ago
checkpoint_pickle.py Remove pytorch_lightning dependency. 2 years ago
cli_args.py Execution Model Inversion (#2666) 6 months ago
clip_config_bigg.json Fix potential issue with non clip text embeddings. 7 months ago
clip_model.py Lower CLIP memory usage by a bit. 7 months ago
clip_vision.py support clip-vit-large-patch14-336 (#4042) 7 months ago
clip_vision_config_g.json Add support for clip g vision model to CLIPVisionLoader. 2 years ago
clip_vision_config_h.json Add support for unCLIP SD2.x models. 2 years ago
clip_vision_config_vitl.json Add support for unCLIP SD2.x models. 2 years ago
clip_vision_config_vitl_336.json support clip-vit-large-patch14-336 (#4042) 7 months ago
conds.py Fix bug with cosxl edit model. 9 months ago
controlnet.py xlabs Flux controlnet implementation. (#4260) 6 months ago
diffusers_convert.py Fix saving text encoder in fp8. 11 months ago
diffusers_load.py load_unet -> load_diffusion_model with a model_options argument. 6 months ago
gligen.py Fix gligen lowvram mode. 1 year ago
graph.py Execution Model Inversion (#2666) 6 months ago
graph_utils.py Execution Model Inversion (#2666) 6 months ago
latent_formats.py Fast preview support for Flux. 7 months ago
lora.py Implement text encoder part of HunyuanDiT loras. 7 months ago
model_base.py Add a way to set model dtype and ops from load_checkpoint_guess_config. 7 months ago
model_detection.py Fix issues with #4302 and support loading diffusers format flux. 7 months ago
model_management.py Fix bug when model cannot be partially unloaded. 6 months ago
model_patcher.py Fix some issues with TE being in lowvram mode. 6 months ago
model_sampling.py Basic Flux Schnell and Flux Dev model implementation. 7 months ago
ops.py Lower T5 memory usage by a few hundred MB. 7 months ago
options.py Only parse command line args when main.py is called. 1 year ago
sample.py Reshape the empty latent image to the right amount of channels if needed. 9 months ago
sampler_helpers.py Make ComfyUI split batches a higher priority than weight offload. 7 months ago
samplers.py More aggressive batch splitting. 7 months ago
sd.py Fix some issues with TE being in lowvram mode. 6 months ago
sd1_clip.py Fix bundled embed. 7 months ago
sd1_clip_config.json Fix potential issue with non clip text embeddings. 7 months ago
sdxl_clip.py Let tokenizers return weights to be stored in the saved checkpoint. 7 months ago
supported_models.py Try to fix SDXL OOM issue on some configurations. 6 months ago
supported_models_base.py Add a way to set model dtype and ops from load_checkpoint_guess_config. 7 months ago
types.py Fix UnetParams annotation typo (#3589) 9 months ago
utils.py Fix issues with #4302 and support loading diffusers format flux. 7 months ago