Hugging Face's logo

CompVis
/
ldm-text2im-large-256

Text-to-Image
Diffusers
PyTorch
LDMTextToImagePipeline
Model card Files Community
9
patrickvonplaten commited on
Commit
44daea0
•
1 Parent(s): 526f89b

upload

Browse files
Files changed (11) hide show
  1. bert/config.json +27 -0
  2. bert/pytorch_model.bin +3 -0
  3. model_index.json +24 -0
  4. scheduler/scheduler_config.json +11 -0
  5. tokenizer/special_tokens_map.json +7 -0
  6. tokenizer/tokenizer_config.json +16 -0
  7. tokenizer/vocab.txt +0 -0
  8. unet/config.json +44 -0
  9. unet/diffusion_model.pt +3 -0
  10. vqvae/config.json +25 -0
  11. vqvae/diffusion_model.pt +3 -0
bert/config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "fusing/latent-diffusion-text2im-large",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "gelu",
5
+ "architectures": [
6
+ "LDMBertModel"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "classifier_dropout": 0.0,
10
+ "d_model": 1280,
11
+ "dropout": 0.1,
12
+ "encoder_attention_heads": 8,
13
+ "encoder_ffn_dim": 5120,
14
+ "encoder_layerdrop": 0.0,
15
+ "encoder_layers": 32,
16
+ "head_dim": 64,
17
+ "init_std": 0.02,
18
+ "max_position_embeddings": 77,
19
+ "model_type": "ldmbert",
20
+ "num_hidden_layers": 32,
21
+ "pad_token_id": 0,
22
+ "scale_embedding": false,
23
+ "torch_dtype": "float32",
24
+ "transformers_version": "4.21.0.dev0",
25
+ "use_cache": true,
26
+ "vocab_size": 30522
27
+ }
bert/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0f946e73c5df896feef89add8becc37f92ddf4359d816d0b3c4825f28d008bd1
3
+ size 2328106869
model_index.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "LatentDiffusionPipeline",
3
+ "_diffusers_version": "0.0.4",
4
+ "bert": [
5
+ "latent_diffusion",
6
+ "LDMBertModel"
7
+ ],
8
+ "scheduler": [
9
+ "diffusers",
10
+ "DDIMScheduler"
11
+ ],
12
+ "tokenizer": [
13
+ "transformers",
14
+ "BertTokenizer"
15
+ ],
16
+ "unet": [
17
+ "diffusers",
18
+ "UNetConditionalModel"
19
+ ],
20
+ "vqvae": [
21
+ "diffusers",
22
+ "AutoencoderKL"
23
+ ]
24
+ }
scheduler/scheduler_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "DDIMScheduler",
3
+ "_diffusers_version": "0.0.4",
4
+ "beta_end": 0.012,
5
+ "beta_schedule": "linear",
6
+ "beta_start": 0.00085,
7
+ "clip_sample": false,
8
+ "num_train_timesteps": 1000,
9
+ "timestep_values": null,
10
+ "trained_betas": null
11
+ }
tokenizer/special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
tokenizer/tokenizer_config.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "do_basic_tokenize": true,
4
+ "do_lower_case": true,
5
+ "mask_token": "[MASK]",
6
+ "model_max_length": 512,
7
+ "name_or_path": "fusing/latent-diffusion-text2im-large",
8
+ "never_split": null,
9
+ "pad_token": "[PAD]",
10
+ "sep_token": "[SEP]",
11
+ "special_tokens_map_file": null,
12
+ "strip_accents": null,
13
+ "tokenize_chinese_chars": true,
14
+ "tokenizer_class": "BertTokenizer",
15
+ "unk_token": "[UNK]"
16
+ }
tokenizer/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
unet/config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ }
unet/diffusion_model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7fc9295026da8bcbbf4eedfe41714d4691e954ebcfbed224010c8bcd5aaea5e2
3
+ size 3489480869
vqvae/config.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "AutoencoderKL",
3
+ "_diffusers_version": "0.0.4",
4
+ "attn_resolutions": [],
5
+ "ch": 128,
6
+ "ch_mult": [
7
+ 1,
8
+ 2,
9
+ 4,
10
+ 4
11
+ ],
12
+ "double_z": true,
13
+ "dropout": 0.0,
14
+ "embed_dim": 4,
15
+ "give_pre_end": false,
16
+ "in_channels": 3,
17
+ "name_or_path": "fusing/latent-diffusion-text2im-large",
18
+ "num_res_blocks": 2,
19
+ "out_ch": 3,
20
+ "remap": null,
21
+ "resamp_with_conv": true,
22
+ "resolution": 256,
23
+ "sane_index_shape": false,
24
+ "z_channels": 4
25
+ }
vqvae/diffusion_model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c2953d1dbaa0b45217adada7507d214ef5b97f0d7f3b85ec920016f3c300e8b
3
+ size 343109057