Update unet/config.json
Browse files- unet/config.json +0 -4
unet/config.json
CHANGED
|
@@ -1,7 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"_class_name": "UNetSpatioTemporalConditionModel",
|
| 3 |
"_diffusers_version": "0.34.0",
|
| 4 |
-
"_name_or_path": "/shared_disk/users/wenkang.qin/checkpoints/robotransfer-high-resolution/unet",
|
| 5 |
"addition_time_embed_dim": 256,
|
| 6 |
"block_out_channels": [
|
| 7 |
320,
|
|
@@ -9,7 +8,6 @@
|
|
| 9 |
1280,
|
| 10 |
1280
|
| 11 |
],
|
| 12 |
-
"caption_channels": null,
|
| 13 |
"cross_attention_dim": 1024,
|
| 14 |
"down_block_types": [
|
| 15 |
"CrossAttnDownBlockSpatioTemporal",
|
|
@@ -17,7 +15,6 @@
|
|
| 17 |
"CrossAttnDownBlockSpatioTemporal",
|
| 18 |
"DownBlockSpatioTemporal"
|
| 19 |
],
|
| 20 |
-
"exvideo_cfg": null,
|
| 21 |
"in_channels": 16,
|
| 22 |
"layers_per_block": 2,
|
| 23 |
"num_attention_heads": [
|
|
@@ -28,7 +25,6 @@
|
|
| 28 |
],
|
| 29 |
"num_frames": 25,
|
| 30 |
"out_channels": 4,
|
| 31 |
-
"patch_size": 1,
|
| 32 |
"projection_class_embeddings_input_dim": 768,
|
| 33 |
"sample_size": 96,
|
| 34 |
"transformer_layers_per_block": 1,
|
|
|
|
| 1 |
{
|
| 2 |
"_class_name": "UNetSpatioTemporalConditionModel",
|
| 3 |
"_diffusers_version": "0.34.0",
|
|
|
|
| 4 |
"addition_time_embed_dim": 256,
|
| 5 |
"block_out_channels": [
|
| 6 |
320,
|
|
|
|
| 8 |
1280,
|
| 9 |
1280
|
| 10 |
],
|
|
|
|
| 11 |
"cross_attention_dim": 1024,
|
| 12 |
"down_block_types": [
|
| 13 |
"CrossAttnDownBlockSpatioTemporal",
|
|
|
|
| 15 |
"CrossAttnDownBlockSpatioTemporal",
|
| 16 |
"DownBlockSpatioTemporal"
|
| 17 |
],
|
|
|
|
| 18 |
"in_channels": 16,
|
| 19 |
"layers_per_block": 2,
|
| 20 |
"num_attention_heads": [
|
|
|
|
| 25 |
],
|
| 26 |
"num_frames": 25,
|
| 27 |
"out_channels": 4,
|
|
|
|
| 28 |
"projection_class_embeddings_input_dim": 768,
|
| 29 |
"sample_size": 96,
|
| 30 |
"transformer_layers_per_block": 1,
|