scraed
7708715590
LanPaint_VideoMaskEditor now outputs a per-frame video mask and an audio mask [F] (hard 0/1 at video frame rate) painted from time intervals in the new hidden audio_mask widget. The frontend shows an audio waveform strip under the video preview with red masked intervals and a playhead, and the editor dialog supports drag-to-paint / click-to-clear intervals. LanPaint_MiniMaxAudioEncode becomes a pure encoder (mask_start/mask_end removed); the workflow attaches the editor's audio mask to the audio latent with SetLatentNoiseMask and Concat AV Latent nests both per-stream masks. Video masks get the temporal union in reshape_mask: a latent token covers ~4 video frames and is all-or-nothing, so each token takes the window-4 max of its frames (nearest-exact) instead of trilinear averaging, which could erase sparse temporal strokes. The stock per-stream mask prep is routed through our prepare_mask (video_inpainting for 5D targets) and handles the 1D [F] audio mask -> audio latent tokens case. Also handles the VideoFromFile.get_frame_rate rename in newer ComfyUI.
34 lines
1.4 KiB
Python
34 lines
1.4 KiB
Python
import torch
|
|
from unittest.mock import MagicMock
|
|
from src.LanPaint.lanpaint import LanPaint
|
|
|
|
|
|
def test_langevin_dynamics_uses_first_order_scheme() -> None:
|
|
"""The second-order damped-oscillator scheme is disabled; langevin_dynamics
|
|
always runs the first-order (overdamped) scheme, whose state carries no
|
|
velocity and whose output stays finite."""
|
|
torch.manual_seed(0)
|
|
# Setup minimal LanPaint instance
|
|
lp = LanPaint(Model=MagicMock(), NSteps=10, Friction=1.0, Lambda=1.0, Beta=1.0, StepSize=0.1)
|
|
# Dummy inputs
|
|
# Shape: (Batch, Channel, Height, Width)
|
|
x_t = torch.randn(1, 4, 8, 8)
|
|
lp.img_dim_size = 4
|
|
mask = torch.zeros_like(x_t)
|
|
# Simple score function
|
|
def score(x):
|
|
return torch.zeros_like(x)
|
|
step_size = torch.tensor([0.1])
|
|
# (sigma, abt, flow_t)
|
|
current_times = (torch.tensor([0.5]), torch.tensor([0.5]), torch.tensor([0.5]))
|
|
# Execute langevin_dynamics (first-order overdamped scheme directly)
|
|
x_out, args_out = lp.langevin_dynamics(x_t, score, mask, step_size, current_times, sigma_y=1.0)
|
|
assert hasattr(args_out, "v")
|
|
assert hasattr(args_out, "C")
|
|
assert hasattr(args_out, "x0")
|
|
assert args_out.v is None # first-order scheme: no velocity state
|
|
assert args_out[1] is args_out.C
|
|
assert args_out[2] is args_out.x0
|
|
# Verify result is finite (the overdamped update is numerically robust)
|
|
assert torch.isfinite(x_out).all(), "Output contains NaNs"
|