Hugging Face's logo

CompVis
/
ldm-text2im-large-256

Text-to-Image
Diffusers
PyTorch
LDMTextToImagePipeline
Model card Files
xet
Community
10
patrickvonplaten commited on
Commit
8310ff5
·
1 Parent(s): a7bdda3

update config

Browse files
Files changed (4) hide show
  1. model_index.json +1 -1
  2. unet/config.json +25 -20
  3. unet/{diffusion_model.pt → diffusion_pytorch_model.bin} +0 -0
  4. vqvae/{diffusion_model.pt → diffusion_pytorch_model.bin} +0 -0
model_index.json CHANGED
@@ -15,7 +15,7 @@
15
  ],
16
  "unet": [
17
  "diffusers",
18
- "UNetConditionalModel"
19
  ],
20
  "vqvae": [
21
  "diffusers",
 
15
  ],
16
  "unet": [
17
  "diffusers",
18
+ "UNet2DConditionModel"
19
  ],
20
  "vqvae": [
21
  "diffusers",
unet/config.json CHANGED
@@ -1,44 +1,49 @@
1
  {
2
- "_class_name": "UNetConditionalModel",
3
  "_diffusers_version": "0.0.4",
4
- "attention_resolutions": [
5
- 4,
6
- 2,
7
- 1
8
- ],
9
- "attn_resolutions": null,
10
  "block_channels": [
11
  320,
12
  640,
13
  1280,
14
  1280
15
  ],
 
 
 
 
 
 
16
  "center_input_sample": false,
17
  "conv_resample": true,
18
  "down_blocks": [
19
- "UNetResCrossAttnDownBlock2D",
20
- "UNetResCrossAttnDownBlock2D",
21
- "UNetResCrossAttnDownBlock2D",
22
- "UNetResDownBlock2D"
23
  ],
24
  "downsample_padding": 1,
25
  "downscale_freq_shift": 0,
26
- "dropout": 0,
27
  "flip_sin_to_cos": true,
28
- "image_size": 32,
 
29
  "in_channels": 4,
 
30
  "mid_block_scale_factor": 1,
31
- "name_or_path": "fusing/latent-diffusion-text2im-large",
 
32
  "num_head_channels": 8,
33
- "num_attention_heads": 8,
34
  "num_res_blocks": 2,
35
  "out_channels": 4,
 
36
  "resnet_eps": 1e-05,
37
- "resnet_num_groups": 32,
 
38
  "up_blocks": [
39
- "UNetResUpBlock2D",
40
- "UNetResCrossAttnUpBlock2D",
41
- "UNetResCrossAttnUpBlock2D",
42
- "UNetResCrossAttnUpBlock2D"
43
  ]
44
  }
 
1
  {
2
+ "_class_name": "UNet2DConditionModel",
3
  "_diffusers_version": "0.0.4",
4
+ "act_fn": "silu",
5
+ "attention_head_dim": 8,
 
 
 
 
6
  "block_channels": [
7
  320,
8
  640,
9
  1280,
10
  1280
11
  ],
12
+ "block_out_channels": [
13
+ 320,
14
+ 640,
15
+ 1280,
16
+ 1280
17
+ ],
18
  "center_input_sample": false,
19
  "conv_resample": true,
20
  "down_blocks": [
21
+ "CrossAttnDownBlock2D",
22
+ "CrossAttnDownBlock2D",
23
+ "CrossAttnDownBlock2D",
24
+ "DownBlock2D"
25
  ],
26
  "downsample_padding": 1,
27
  "downscale_freq_shift": 0,
 
28
  "flip_sin_to_cos": true,
29
+ "freq_shift": 0,
30
+ "image_size": null,
31
  "in_channels": 4,
32
+ "layers_per_block": 2,
33
  "mid_block_scale_factor": 1,
34
+ "norm_eps": 1e-05,
35
+ "norm_num_groups": 32,
36
  "num_head_channels": 8,
 
37
  "num_res_blocks": 2,
38
  "out_channels": 4,
39
+ "resnet_act_fn": "silu",
40
  "resnet_eps": 1e-05,
41
+ "resnet_num_groups": 30,
42
+ "sample_size": 32,
43
  "up_blocks": [
44
+ "UpBlock2D",
45
+ "CrossAttnUpBlock2D",
46
+ "CrossAttnUpBlock2D",
47
+ "CrossAttnUpBlock2D"
48
  ]
49
  }
unet/{diffusion_model.pt → diffusion_pytorch_model.bin} RENAMED
File without changes
vqvae/{diffusion_model.pt → diffusion_pytorch_model.bin} RENAMED
File without changes