updated license

README.md

@@ -1,3 +1,14 @@
 ---
-license: apache-2.0
+license: mit
 ---
+
+# VQVAE
+
+This repository is a clone of the [VideoGPT](https://github.com/wilson1yan/VideoGPT/tree/master) in order to convert the VQ-VAE model to the Hugging Face format for easier model loading.
+
+Paper: [VideoGPT: Video Generation using VQ-VAE and Transformers](https://arxiv.org/abs/2104.10157)
+
+## License
+
+We follow the MIT license distributed by the [VideoGPT](https://github.com/wilson1yan/VideoGPT/tree/master) project.
+

_utils.py

@@ -1,3 +1,32 @@
+"""
+MIT License
+
+Copyright (c) 2021 Wilson Yan
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+
+
+This file is copied from https://github.com/wilson1yan/VideoGPT/blob/master/videogpt/utils.py
+We adapted it to Hugging Face AutoModel for easier model loading.
+"""
+
+
 # Shifts src_tf dim to dest dim
 # i.e. shift_dim(x, 1, -1) would be (b, c, t, h, w) -> (b, t, h, w, c)
 def shift_dim(x, src_dim=-1, dest_dim=-1, make_contiguous=True):

attention.py

@@ -1,3 +1,32 @@
+"""
+MIT License
+
+Copyright (c) 2021 Wilson Yan
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+
+
+This file is copied from https://github.com/wilson1yan/VideoGPT/blob/master/videogpt/attention.py
+We adapted it to Hugging Face AutoModel for easier model loading.
+"""
+
+
 import numpy as np
 
 import torch

modeling_vqvae.py

@@ -1,3 +1,32 @@
+"""
+MIT License
+
+Copyright (c) 2021 Wilson Yan
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+
+
+This file is copied from https://github.com/wilson1yan/VideoGPT/blob/master/videogpt/vqvae.py
+We adapted it to Hugging Face AutoModel for easier model loading.
+"""
+
+
 import os
 import math
 import numpy as np
@@ -6,43 +35,13 @@ import torch
 import torch.nn as nn
 import torch.nn.functional as F
 import torch.distributed as dist
-import gdown
 
 from .attention import MultiHeadAttention
 from ._utils import shift_dim
 from transformers import PreTrainedModel
-from typing import Tuple
 from .configuration_vqvae import VQVAEConfig
 
 
-_VQVAE = {
-    'bair_stride4x2x2': '1iIAYJ2Qqrx5Q94s5eIXQYJgAydzvT_8L', # trained on 16 frames of 64 x 64 images
-    'ucf101_stride4x4x4': '1uuB_8WzHP_bbBmfuaIV7PK_Itl3DyHY5', # trained on 16 frames of 128 x 128 images
-    'kinetics_stride4x4x4': '1DOvOZnFAIQmux6hG7pN_HkyJZy3lXbCB', # trained on 16 frames of 128 x 128 images
-    'kinetics_stride2x4x4': '1jvtjjtrtE4cy6pl7DK_zWFEPY3RZt2pB' # trained on 16 frames of 128 x 128 images
-}
-
-def download(id, fname, root=None):
-    """
-    Download the VQVAE weights from Google Drive.
-    
-    Args:
-        id (str): the ID of the file to download
-        fname (str): the name of the file to save
-        root (str): the directory to save the file to
-    """
-    if root is None:
-        root = os.path.expanduser('~/.cache/sora')
-    os.makedirs(root, exist_ok=True)
-    destination = os.path.join(root, fname)
-
-    if os.path.exists(destination):
-        return destination
-
-    gdown.download(id=id, output=destination, quiet=False)
-    return destination
-
-
 class VQVAE(PreTrainedModel):
     config_class = VQVAEConfig