LanguageBind commited on
Commit
c41aa03
1 Parent(s): 39641a8

Upload 2 files

Browse files
vae/config.json CHANGED
@@ -1 +1,71 @@
1
- {"_class_name": "CausalVAEModel", "hidden_size": 128, "in_channels": 3, "out_channels": 3, "z_channels": 4, "ch_mult": [1, 2, 4, 4], "num_res_block": 2, "attn_resolutions": [], "dropout": 0.0, "resolution": 64, "attn_type": "vanilla3D", "use_linear_attn": false, "embed_dim": 4, "time_compress": 2, "logvar_init": 0.0, "kl_weight": 1e-06, "pixelloss_weight": 1, "perceptual_weight": 1, "disc_loss": "hinge", "model": "CausalVAEModel"}
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "CausalVAEModel",
3
+ "_diffusers_version": "0.27.2",
4
+ "attn_resolutions": [],
5
+ "decoder_attention": "AttnBlock3D",
6
+ "decoder_conv_in": "CausalConv3d",
7
+ "decoder_conv_out": "CausalConv3d",
8
+ "decoder_mid_resnet": "ResnetBlock3D",
9
+ "decoder_resnet_blocks": [
10
+ "ResnetBlock3D",
11
+ "ResnetBlock3D",
12
+ "ResnetBlock3D",
13
+ "ResnetBlock3D"
14
+ ],
15
+ "decoder_spatial_upsample": [
16
+ "",
17
+ "SpatialUpsample2x",
18
+ "SpatialUpsample2x",
19
+ "SpatialUpsample2x"
20
+ ],
21
+ "decoder_temporal_upsample": [
22
+ "",
23
+ "",
24
+ "TimeUpsample2x",
25
+ "TimeUpsample2x"
26
+ ],
27
+ "double_z": true,
28
+ "dropout": 0.0,
29
+ "embed_dim": 4,
30
+ "encoder_attention": "AttnBlock3D",
31
+ "encoder_conv_in": "CausalConv3d",
32
+ "encoder_conv_out": "CausalConv3d",
33
+ "encoder_mid_resnet": "ResnetBlock3D",
34
+ "encoder_resnet_blocks": [
35
+ "ResnetBlock3D",
36
+ "ResnetBlock3D",
37
+ "ResnetBlock3D",
38
+ "ResnetBlock3D"
39
+ ],
40
+ "encoder_spatial_downsample": [
41
+ "SpatialDownsample2x",
42
+ "SpatialDownsample2x",
43
+ "SpatialDownsample2x",
44
+ ""
45
+ ],
46
+ "encoder_temporal_downsample": [
47
+ "",
48
+ "TimeDownsample2x",
49
+ "TimeDownsample2x",
50
+ ""
51
+ ],
52
+ "hidden_size": 128,
53
+ "hidden_size_mult": [
54
+ 1,
55
+ 2,
56
+ 4,
57
+ 4
58
+ ],
59
+ "loss_params": {
60
+ "disc_start": 2001,
61
+ "disc_weight": 0.5,
62
+ "kl_weight": 1e-06,
63
+ "logvar_init": 0.0
64
+ },
65
+ "loss_type": "opensora.models.ae.videobase.losses.LPIPSWithDiscriminator",
66
+ "lr": 1e-05,
67
+ "num_res_blocks": 2,
68
+ "q_conv": "CausalConv3d",
69
+ "resolution": 256,
70
+ "z_channels": 4
71
+ }
vae/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3c4a1d9435c27f907913396ab9e27769393ec23591cf6c85aeb65fcbedcd493a
3
+ size 986646668