chendl02
commited on
Commit
•
c816139
1
Parent(s):
579b746
add Goal-depth LDM
Browse files- README.md +9 -0
- unet/config.json +73 -0
- unet/diffusion_pytorch_model.safetensors +3 -0
README.md
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
---
|
| 4 |
+
|
| 5 |
+
# 3D-VLA: A 3D Vision-Language-Action Generative World Model
|
| 6 |
+
|
| 7 |
+
Model card for 3D-VLA
|
| 8 |
+
|
| 9 |
+
<img src="https://vis-www.cs.umass.edu/3dvla/static/video/3dvla_arch.mp4" alt="Logo" width="90%">
|
unet/config.json
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_class_name": "UNet2DConditionModel",
|
| 3 |
+
"_diffusers_version": "0.29.2",
|
| 4 |
+
"_name_or_path": "./lavis/output/LDM/goal_depth/runs/checkpoint-98000",
|
| 5 |
+
"act_fn": "silu",
|
| 6 |
+
"addition_embed_type": null,
|
| 7 |
+
"addition_embed_type_num_heads": 64,
|
| 8 |
+
"addition_time_embed_dim": null,
|
| 9 |
+
"attention_head_dim": [
|
| 10 |
+
5,
|
| 11 |
+
10,
|
| 12 |
+
20,
|
| 13 |
+
20
|
| 14 |
+
],
|
| 15 |
+
"attention_type": "default",
|
| 16 |
+
"block_out_channels": [
|
| 17 |
+
320,
|
| 18 |
+
640,
|
| 19 |
+
1280,
|
| 20 |
+
1280
|
| 21 |
+
],
|
| 22 |
+
"center_input_sample": false,
|
| 23 |
+
"class_embed_type": null,
|
| 24 |
+
"class_embeddings_concat": false,
|
| 25 |
+
"conv_in_kernel": 3,
|
| 26 |
+
"conv_out_kernel": 3,
|
| 27 |
+
"cross_attention_dim": 1024,
|
| 28 |
+
"cross_attention_norm": null,
|
| 29 |
+
"down_block_types": [
|
| 30 |
+
"CrossAttnDownBlock2D",
|
| 31 |
+
"CrossAttnDownBlock2D",
|
| 32 |
+
"CrossAttnDownBlock2D",
|
| 33 |
+
"DownBlock2D"
|
| 34 |
+
],
|
| 35 |
+
"downsample_padding": 1,
|
| 36 |
+
"dropout": 0.0,
|
| 37 |
+
"dual_cross_attention": false,
|
| 38 |
+
"encoder_hid_dim": null,
|
| 39 |
+
"encoder_hid_dim_type": null,
|
| 40 |
+
"flip_sin_to_cos": true,
|
| 41 |
+
"freq_shift": 0,
|
| 42 |
+
"in_channels": 16,
|
| 43 |
+
"layers_per_block": 2,
|
| 44 |
+
"mid_block_only_cross_attention": null,
|
| 45 |
+
"mid_block_scale_factor": 1,
|
| 46 |
+
"mid_block_type": "UNetMidBlock2DCrossAttn",
|
| 47 |
+
"norm_eps": 1e-05,
|
| 48 |
+
"norm_num_groups": 32,
|
| 49 |
+
"num_attention_heads": null,
|
| 50 |
+
"num_class_embeds": null,
|
| 51 |
+
"only_cross_attention": false,
|
| 52 |
+
"out_channels": 8,
|
| 53 |
+
"projection_class_embeddings_input_dim": null,
|
| 54 |
+
"resnet_out_scale_factor": 1.0,
|
| 55 |
+
"resnet_skip_time_act": false,
|
| 56 |
+
"resnet_time_scale_shift": "default",
|
| 57 |
+
"reverse_transformer_layers_per_block": null,
|
| 58 |
+
"sample_size": 96,
|
| 59 |
+
"time_cond_proj_dim": null,
|
| 60 |
+
"time_embedding_act_fn": null,
|
| 61 |
+
"time_embedding_dim": null,
|
| 62 |
+
"time_embedding_type": "positional",
|
| 63 |
+
"timestep_post_act": null,
|
| 64 |
+
"transformer_layers_per_block": 1,
|
| 65 |
+
"up_block_types": [
|
| 66 |
+
"UpBlock2D",
|
| 67 |
+
"CrossAttnUpBlock2D",
|
| 68 |
+
"CrossAttnUpBlock2D",
|
| 69 |
+
"CrossAttnUpBlock2D"
|
| 70 |
+
],
|
| 71 |
+
"upcast_attention": false,
|
| 72 |
+
"use_linear_projection": true
|
| 73 |
+
}
|
unet/diffusion_pytorch_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7944e8a9b946acd4f91de7f40c49c48bae05302a488eafcbbbb06782fbe4a99d
|
| 3 |
+
size 3463910864
|