mirror of
https://github.com/invoke-ai/InvokeAI
synced 2024-08-30 20:32:17 +00:00
feat(api): chore: pydantic & fastapi upgrade
Upgrade pydantic and fastapi to latest.
- pydantic~=2.4.2
- fastapi~=103.2
- fastapi-events~=0.9.1
**Big Changes**
There are a number of logic changes needed to support pydantic v2. Most changes are very simple, like using the new methods to serialized and deserialize models, but there are a few more complex changes.
**Invocations**
The biggest change relates to invocation creation, instantiation and validation.
Because pydantic v2 moves all validation logic into the rust pydantic-core, we may no longer directly stick our fingers into the validation pie.
Previously, we (ab)used models and fields to allow invocation fields to be optional at instantiation, but required when `invoke()` is called. We directly manipulated the fields and invocation models when calling `invoke()`.
With pydantic v2, this is much more involved. Changes to the python wrapper do not propagate down to the rust validation logic - you have to rebuild the model. This causes problem with concurrent access to the invocation classes and is not a free operation.
This logic has been totally refactored and we do not need to change the model any more. The details are in `baseinvocation.py`, in the `InputField` function and `BaseInvocation.invoke_internal()` method.
In the end, this implementation is cleaner.
**Invocation Fields**
In pydantic v2, you can no longer directly add or remove fields from a model.
Previously, we did this to add the `type` field to invocations.
**Invocation Decorators**
With pydantic v2, we instead use the imperative `create_model()` API to create a new model with the additional field. This is done in `baseinvocation.py` in the `invocation()` wrapper.
A similar technique is used for `invocation_output()`.
**Minor Changes**
There are a number of minor changes around the pydantic v2 models API.
**Protected `model_` Namespace**
All models' pydantic-provided methods and attributes are prefixed with `model_` and this is considered a protected namespace. This causes some conflict, because "model" means something to us, and we have a ton of pydantic models with attributes starting with "model_".
Forunately, there are no direct conflicts. However, in any pydantic model where we define an attribute or method that starts with "model_", we must tell set the protected namespaces to an empty tuple.
```py
class IPAdapterModelField(BaseModel):
model_name: str = Field(description="Name of the IP-Adapter model")
base_model: BaseModelType = Field(description="Base model")
model_config = ConfigDict(protected_namespaces=())
```
**Model Serialization**
Pydantic models no longer have `Model.dict()` or `Model.json()`.
Instead, we use `Model.model_dump()` or `Model.model_dump_json()`.
**Model Deserialization**
Pydantic models no longer have `Model.parse_obj()` or `Model.parse_raw()`, and there are no `parse_raw_as()` or `parse_obj_as()` functions.
Instead, you need to create a `TypeAdapter` object to parse python objects or JSON into a model.
```py
adapter_graph = TypeAdapter(Graph)
deserialized_graph_from_json = adapter_graph.validate_json(graph_json)
deserialized_graph_from_dict = adapter_graph.validate_python(graph_dict)
```
**Field Customisation**
Pydantic `Field`s no longer accept arbitrary args.
Now, you must put all additional arbitrary args in a `json_schema_extra` arg on the field.
**Schema Customisation**
FastAPI and pydantic schema generation now follows the OpenAPI version 3.1 spec.
This necessitates two changes:
- Our schema customization logic has been revised
- Schema parsing to build node templates has been revised
The specific aren't important, but this does present additional surface area for bugs.
**Performance Improvements**
Pydantic v2 is a full rewrite with a rust backend. This offers a substantial performance improvement (pydantic claims 5x to 50x depending on the task). We'll notice this the most during serialization and deserialization of sessions/graphs, which happens very very often - a couple times per node.
I haven't done any benchmarks, but anecdotally, graph execution is much faster. Also, very larges graphs - like with massive iterators - are much, much faster.
This commit is contained in:
psychedelicious
@ -1,7 +1,7 @@
|
||||
import copy
|
||||
from typing import List, Optional
|
||||
|
||||
from pydantic import BaseModel, Field
|
||||
from pydantic import BaseModel, ConfigDict, Field
|
||||
|
||||
from ...backend.model_management import BaseModelType, ModelType, SubModelType
|
||||
from .baseinvocation import (
|
||||
@ -24,6 +24,8 @@ class ModelInfo(BaseModel):
|
||||
model_type: ModelType = Field(description="Info to load submodel")
|
||||
submodel: Optional[SubModelType] = Field(default=None, description="Info to load submodel")
|
||||
|
||||
model_config = ConfigDict(protected_namespaces=())
|
||||
|
||||
|
||||
class LoraInfo(ModelInfo):
|
||||
weight: float = Field(description="Lora's weight which to use when apply to model")
|
||||
@ -65,6 +67,8 @@ class MainModelField(BaseModel):
|
||||
base_model: BaseModelType = Field(description="Base model")
|
||||
model_type: ModelType = Field(description="Model Type")
|
||||
|
||||
model_config = ConfigDict(protected_namespaces=())
|
||||
|
||||
|
||||
class LoRAModelField(BaseModel):
|
||||
"""LoRA model field"""
|
||||
@ -72,8 +76,16 @@ class LoRAModelField(BaseModel):
|
||||
model_name: str = Field(description="Name of the LoRA model")
|
||||
base_model: BaseModelType = Field(description="Base model")
|
||||
|
||||
model_config = ConfigDict(protected_namespaces=())
|
||||
|
||||
@invocation("main_model_loader", title="Main Model", tags=["model"], category="model", version="1.0.0")
|
||||
|
||||
@invocation(
|
||||
"main_model_loader",
|
||||
title="Main Model",
|
||||
tags=["model"],
|
||||
category="model",
|
||||
version="1.0.0",
|
||||
)
|
||||
class MainModelLoaderInvocation(BaseInvocation):
|
||||
"""Loads a main model, outputting its submodels."""
|
||||
|
||||
@ -180,10 +192,16 @@ class LoraLoaderInvocation(BaseInvocation):
|
||||
lora: LoRAModelField = InputField(description=FieldDescriptions.lora_model, input=Input.Direct, title="LoRA")
|
||||
weight: float = InputField(default=0.75, description=FieldDescriptions.lora_weight)
|
||||
unet: Optional[UNetField] = InputField(
|
||||
default=None, description=FieldDescriptions.unet, input=Input.Connection, title="UNet"
|
||||
default=None,
|
||||
description=FieldDescriptions.unet,
|
||||
input=Input.Connection,
|
||||
title="UNet",
|
||||
)
|
||||
clip: Optional[ClipField] = InputField(
|
||||
default=None, description=FieldDescriptions.clip, input=Input.Connection, title="CLIP"
|
||||
default=None,
|
||||
description=FieldDescriptions.clip,
|
||||
input=Input.Connection,
|
||||
title="CLIP",
|
||||
)
|
||||
|
||||
def invoke(self, context: InvocationContext) -> LoraLoaderOutput:
|
||||
@ -244,20 +262,35 @@ class SDXLLoraLoaderOutput(BaseInvocationOutput):
|
||||
clip2: Optional[ClipField] = OutputField(default=None, description=FieldDescriptions.clip, title="CLIP 2")
|
||||
|
||||
|
||||
@invocation("sdxl_lora_loader", title="SDXL LoRA", tags=["lora", "model"], category="model", version="1.0.0")
|
||||
@invocation(
|
||||
"sdxl_lora_loader",
|
||||
title="SDXL LoRA",
|
||||
tags=["lora", "model"],
|
||||
category="model",
|
||||
version="1.0.0",
|
||||
)
|
||||
class SDXLLoraLoaderInvocation(BaseInvocation):
|
||||
"""Apply selected lora to unet and text_encoder."""
|
||||
|
||||
lora: LoRAModelField = InputField(description=FieldDescriptions.lora_model, input=Input.Direct, title="LoRA")
|
||||
weight: float = InputField(default=0.75, description=FieldDescriptions.lora_weight)
|
||||
unet: Optional[UNetField] = InputField(
|
||||
default=None, description=FieldDescriptions.unet, input=Input.Connection, title="UNet"
|
||||
default=None,
|
||||
description=FieldDescriptions.unet,
|
||||
input=Input.Connection,
|
||||
title="UNet",
|
||||
)
|
||||
clip: Optional[ClipField] = InputField(
|
||||
default=None, description=FieldDescriptions.clip, input=Input.Connection, title="CLIP 1"
|
||||
default=None,
|
||||
description=FieldDescriptions.clip,
|
||||
input=Input.Connection,
|
||||
title="CLIP 1",
|
||||
)
|
||||
clip2: Optional[ClipField] = InputField(
|
||||
default=None, description=FieldDescriptions.clip, input=Input.Connection, title="CLIP 2"
|
||||
default=None,
|
||||
description=FieldDescriptions.clip,
|
||||
input=Input.Connection,
|
||||
title="CLIP 2",
|
||||
)
|
||||
|
||||
def invoke(self, context: InvocationContext) -> SDXLLoraLoaderOutput:
|
||||
@ -330,6 +363,8 @@ class VAEModelField(BaseModel):
|
||||
model_name: str = Field(description="Name of the model")
|
||||
base_model: BaseModelType = Field(description="Base model")
|
||||
|
||||
model_config = ConfigDict(protected_namespaces=())
|
||||
|
||||
|
||||
@invocation_output("vae_loader_output")
|
||||
class VaeLoaderOutput(BaseInvocationOutput):
|
||||
@ -343,7 +378,10 @@ class VaeLoaderInvocation(BaseInvocation):
|
||||
"""Loads a VAE model, outputting a VaeLoaderOutput"""
|
||||
|
||||
vae_model: VAEModelField = InputField(
|
||||
description=FieldDescriptions.vae_model, input=Input.Direct, ui_type=UIType.VaeModel, title="VAE"
|
||||
description=FieldDescriptions.vae_model,
|
||||
input=Input.Direct,
|
||||
ui_type=UIType.VaeModel,
|
||||
title="VAE",
|
||||
)
|
||||
|
||||
def invoke(self, context: InvocationContext) -> VaeLoaderOutput:
|
||||
@ -372,19 +410,31 @@ class VaeLoaderInvocation(BaseInvocation):
|
||||
class SeamlessModeOutput(BaseInvocationOutput):
|
||||
"""Modified Seamless Model output"""
|
||||
|
||||
unet: Optional[UNetField] = OutputField(description=FieldDescriptions.unet, title="UNet")
|
||||
vae: Optional[VaeField] = OutputField(description=FieldDescriptions.vae, title="VAE")
|
||||
unet: Optional[UNetField] = OutputField(default=None, description=FieldDescriptions.unet, title="UNet")
|
||||
vae: Optional[VaeField] = OutputField(default=None, description=FieldDescriptions.vae, title="VAE")
|
||||
|
||||
|
||||
@invocation("seamless", title="Seamless", tags=["seamless", "model"], category="model", version="1.0.0")
|
||||
@invocation(
|
||||
"seamless",
|
||||
title="Seamless",
|
||||
tags=["seamless", "model"],
|
||||
category="model",
|
||||
version="1.0.0",
|
||||
)
|
||||
class SeamlessModeInvocation(BaseInvocation):
|
||||
"""Applies the seamless transformation to the Model UNet and VAE."""
|
||||
|
||||
unet: Optional[UNetField] = InputField(
|
||||
default=None, description=FieldDescriptions.unet, input=Input.Connection, title="UNet"
|
||||
default=None,
|
||||
description=FieldDescriptions.unet,
|
||||
input=Input.Connection,
|
||||
title="UNet",
|
||||
)
|
||||
vae: Optional[VaeField] = InputField(
|
||||
default=None, description=FieldDescriptions.vae_model, input=Input.Connection, title="VAE"
|
||||
default=None,
|
||||
description=FieldDescriptions.vae_model,
|
||||
input=Input.Connection,
|
||||
title="VAE",
|
||||
)
|
||||
seamless_y: bool = InputField(default=True, input=Input.Any, description="Specify whether Y axis is seamless")
|
||||
seamless_x: bool = InputField(default=True, input=Input.Any, description="Specify whether X axis is seamless")
|
||||
|
||||
Reference in New Issue
Block a user