File size: 2,318 Bytes
b386992 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 |
# Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import torch
from diffusers import AutoencoderKL
from einops import rearrange
class AutoencoderKLVAE(torch.nn.Module):
"""
A class that wraps the AutoencoderKL model and provides a decode method.
Attributes:
vae (AutoencoderKL): The underlying AutoencoderKL model loaded from a pretrained path.
"""
def __init__(self, path):
"""
Initialize the AutoencoderKLVAE instance.
Args:
path (str): The path to the pretrained AutoencoderKL model.
"""
super().__init__()
self.vae = AutoencoderKL.from_pretrained(path, torch_dtype=torch.bfloat16)
@torch.no_grad()
def decode(self, x):
"""
Decode a latent representation using the underlying VAE model.
This method takes a latent tensor `x` and decodes it into an image.
If `x` has a temporal dimension `T` of 1, it
rearranges the tensor before and after decoding.
Args:
x (torch.Tensor): A tensor of shape (B, C, T, H, W), where:
B = batch size
C = number of channels
T = temporal dimension
H = height
W = width
Returns:
torch.Tensor: Decoded image tensor with the same shape as the input (B, C, T, H, W).
"""
B, C, T, H, W = x.shape
if T == 1:
x = rearrange(x, 'b c t h w -> (b t) c h w')
x = x / self.vae.config.scaling_factor
out = self.vae.decode(x, return_dict=False)[0]
if T == 1:
return rearrange(out, '(b t) c h w -> b c t h w', t=1)
return out
|