56 lines
1.8 KiB
Markdown
56 lines
1.8 KiB
Markdown
|
|
# Documentation
|
|
- Class name: SaltScheduleSmoothing
|
|
- Category: SALT/Scheduling/Filter
|
|
- Output node: False
|
|
|
|
SaltScheduleSmoothing节点旨在对调度列表应用平滑过滤器，使用指定的平滑因子来混合连续的值。这个过程的目标是在调度中的点之间创建更加渐进的过渡，从而提高整个序列的平滑度。
|
|
|
|
# Input types
|
|
## Required
|
|
- schedule_list
|
|
- schedule_list是要进行平滑处理的值序列。它作为平滑操作的主要数据来源，直接影响最终调度的平滑度。
|
|
- Comfy dtype: LIST
|
|
- Python dtype: List[float]
|
|
- smoothing_factor
|
|
- smoothing_factor决定了应用于调度列表的平滑程度。较高的因子会导致点之间的过渡更加平滑，直接影响输出的平滑度。
|
|
- Comfy dtype: FLOAT
|
|
- Python dtype: float
|
|
|
|
# Output types
|
|
- schedule_list
|
|
- 输出是输入调度列表的平滑版本，通过平滑处理使点之间的过渡更加渐进。
|
|
- Comfy dtype: LIST
|
|
- Python dtype: List[float]
|
|
|
|
|
|
## Usage tips
|
|
- Infra type: `CPU`
|
|
- Common nodes: unknown
|
|
|
|
|
|
## Source code
|
|
```python
|
|
class SaltScheduleSmoothing:
|
|
@classmethod
|
|
def INPUT_TYPES(cls):
|
|
return {
|
|
"required": {
|
|
"schedule_list": ("LIST", ),
|
|
"smoothing_factor": ("FLOAT", {"min": 0.0, "max": 1.0, "default": 0.5}),
|
|
},
|
|
}
|
|
|
|
RETURN_TYPES = ("LIST",)
|
|
RETURN_NAMES = ("schedule_list", )
|
|
FUNCTION = "smooth"
|
|
CATEGORY = "SALT/Scheduling/Filter"
|
|
|
|
def smooth(self, schedule_list, smoothing_factor):
|
|
smoothed_schedule = schedule_list[:]
|
|
for i in range(1, len(schedule_list)):
|
|
smoothed_schedule[i] = smoothed_schedule[i-1] * (1 - smoothing_factor) + schedule_list[i] * smoothing_factor
|
|
return (smoothed_schedule, )
|
|
|
|
```
|