upload cv-vae weights
Browse files- vae3d/config.json +34 -0
- vae3d/cv_vae.ckpt +3 -0
- vae3d/diffusion_pytorch_model.safetensors +3 -0
vae3d/config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_class_name": "CVVAEModel",
|
| 3 |
+
"_diffusers_version": "0.26.3",
|
| 4 |
+
"attn_resolutions": [],
|
| 5 |
+
"causal_decoder": false,
|
| 6 |
+
"causal_encoder": true,
|
| 7 |
+
"ch": 128,
|
| 8 |
+
"ch_mult": [
|
| 9 |
+
1,
|
| 10 |
+
2,
|
| 11 |
+
4,
|
| 12 |
+
4
|
| 13 |
+
],
|
| 14 |
+
"decoder_attn_type": "spatial-temporal-xformer",
|
| 15 |
+
"double_z": true,
|
| 16 |
+
"dropout": 0.0,
|
| 17 |
+
"en_de_n_frames_a_time": 16,
|
| 18 |
+
"encoder_attn_type": "vanilla-xformers",
|
| 19 |
+
"half_3d": true,
|
| 20 |
+
"in_channels": 3,
|
| 21 |
+
"num_res_blocks": 2,
|
| 22 |
+
"num_video_frames": null,
|
| 23 |
+
"out_ch": 3,
|
| 24 |
+
"scaling_factor": 0.18215,
|
| 25 |
+
"force_upcast": true,
|
| 26 |
+
"reshape_x_dim_to_4": false,
|
| 27 |
+
"reshape_z_dim_to_4": false,
|
| 28 |
+
"spatial_n_compress": 8,
|
| 29 |
+
"tile_overlap_ratio": 0.2222,
|
| 30 |
+
"tile_spatial_size": 576,
|
| 31 |
+
"time_n_compress": 4,
|
| 32 |
+
"use_3d_conv": true,
|
| 33 |
+
"z_channels": 4
|
| 34 |
+
}
|
vae3d/cv_vae.ckpt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3a71060522dd74162ec3f734dd2a02c511179bca0960ff499d1c9fe544e6f712
|
| 3 |
+
size 904840525
|
vae3d/diffusion_pytorch_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:db6d128fc5ad7ba6e79c8c6f5870f785b4b48e4d7e9180ccff4f0141ddd34c24
|
| 3 |
+
size 729812244
|