Upload 16 files

models/stage1/StableDiffusion-FlowGen/controlnet/controlnet.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b79229819acb0439452c168d13600d003702714d14766276f6dec53784da9b91
+size 2269066191

models/stage1/StableDiffusion-FlowGen/controlnet/convert.py

@@ -0,0 +1,6 @@
+import torch
+
+weight = torch.load("checkpoint14000.ckpt")["state_dict"]
+weight = {k.replace("module.", ""):v for k, v in weight.items() if "controlnet" not in k}
+torch.save(weight,"diffusion_pytorch_model.bin")
+print(weight.keys())

models/stage1/StableDiffusion-FlowGen/text_encoder/config.json

@@ -0,0 +1,25 @@
+{
+  "_name_or_path": "openai/clip-vit-large-patch14",
+  "architectures": [
+    "CLIPTextModel"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "dropout": 0.0,
+  "eos_token_id": 2,
+  "hidden_act": "quick_gelu",
+  "hidden_size": 768,
+  "initializer_factor": 1.0,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 77,
+  "model_type": "clip_text_model",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "projection_dim": 768,
+  "torch_dtype": "float32",
+  "transformers_version": "4.22.0.dev0",
+  "vocab_size": 49408
+}

models/stage1/StableDiffusion-FlowGen/text_encoder/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:770a47a9ffdcfda0b05506a7888ed714d06131d60267e6cf52765d61cf59fd67
+size 492305335

models/stage1/StableDiffusion-FlowGen/tokenizer/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<|startoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|endoftext|>",
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

models/stage1/StableDiffusion-FlowGen/tokenizer/tokenizer_config.json

@@ -0,0 +1,34 @@
+{
+  "add_prefix_space": false,
+  "bos_token": {
+    "__type": "AddedToken",
+    "content": "<|startoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "do_lower_case": true,
+  "eos_token": {
+    "__type": "AddedToken",
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "errors": "replace",
+  "model_max_length": 77,
+  "name_or_path": "openai/clip-vit-large-patch14",
+  "pad_token": "<|endoftext|>",
+  "special_tokens_map_file": "./special_tokens_map.json",
+  "tokenizer_class": "CLIPTokenizer",
+  "unk_token": {
+    "__type": "AddedToken",
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

models/stage1/StableDiffusion-FlowGen/unet/add_channel.py

@@ -0,0 +1,17 @@
+import torch
+
+state_dict = torch.load("diffusion_pytorch_model.origin_bk.bin", map_location="cpu")
+
+zero_data=torch.zeros(320,4,3,3)
+new_weight=torch.cat((state_dict['conv_in.weight'], zero_data),dim=1)
+state_dict['conv_in.weight'] = new_weight
+
+#zero_data=torch.zeros(4,320,3,3)
+#new_weight=torch.cat((state_dict['conv_out.weight'], zero_data),dim=0)
+#state_dict['conv_out.weight'] = new_weight
+
+#zero_data=torch.zeros(4)
+#new_weight=torch.cat((state_dict['conv_out.bias'], zero_data),dim=0)
+#state_dict['conv_out.bias'] = new_weight
+
+torch.save(state_dict, "diffusion_pytorch_model_in8out4.bin")

models/stage1/StableDiffusion-FlowGen/unet/config.json

@@ -0,0 +1,36 @@
+{
+  "_class_name": "UNet2DConditionModel",
+  "_diffusers_version": "0.6.0",
+  "act_fn": "silu",
+  "attention_head_dim": 8,
+  "block_out_channels": [
+    320,
+    640,
+    1280,
+    1280
+  ],
+  "center_input_sample": false,
+  "cross_attention_dim": 768,
+  "down_block_types": [
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D",
+    "DownBlock2D"
+  ],
+  "downsample_padding": 1,
+  "flip_sin_to_cos": true,
+  "freq_shift": 0,
+  "in_channels": 8,
+  "layers_per_block": 2,
+  "mid_block_scale_factor": 1,
+  "norm_eps": 1e-05,
+  "norm_num_groups": 32,
+  "out_channels": 4,
+  "sample_size": 64,
+  "up_block_types": [
+    "UpBlock2D",
+    "CrossAttnUpBlock2D",
+    "CrossAttnUpBlock2D",
+    "CrossAttnUpBlock2D"
+  ]
+}

models/stage1/StableDiffusion-FlowGen/unet/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48b149c17dc685a32748371ac221009001dcbadad868950655c113ca52886487
+size 5264486111

models/stage1/StableDiffusion-FlowGen/vae/add_channel.py

@@ -0,0 +1,10 @@
+import torch
+
+state_dict = torch.load("diffusion_pytorch_model.bin", map_location="cpu")
+
+
+state_dict['encoder.conv_in.weight'] = state_dict['encoder.conv_in.weight'][:, :2, ...]
+state_dict['decoder.conv_out.weight'] = state_dict['decoder.conv_out.weight'][:2, :, ...]
+state_dict['decoder.conv_out.bias'] = state_dict['decoder.conv_out.bias'][:2]
+
+torch.save(state_dict, "diffusion_pytorch_model_in2out2.bin")

models/stage1/StableDiffusion-FlowGen/vae/config.json

@@ -0,0 +1,29 @@
+{
+  "_class_name": "AutoencoderKL",
+  "_diffusers_version": "0.6.0",
+  "act_fn": "silu",
+  "block_out_channels": [
+    128,
+    256,
+    512,
+    512
+  ],
+  "down_block_types": [
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D"
+  ],
+  "in_channels": 2,
+  "latent_channels": 4,
+  "layers_per_block": 2,
+  "norm_num_groups": 32,
+  "out_channels": 2,
+  "sample_size": 512,
+  "up_block_types": [
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D"
+  ]
+}

models/stage1/StableDiffusion-FlowGen/vae/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b134cded8eb78b184aefb8805b6b572f36fa77b255c483665dda931fa0130c5
+size 334707217

models/stage1/StableDiffusion-FlowGen/vae_flow/config.json

@@ -0,0 +1,29 @@
+{
+  "_class_name": "AutoencoderKL",
+  "_diffusers_version": "0.6.0",
+  "act_fn": "silu",
+  "block_out_channels": [
+    128,
+    256,
+    512,
+    512
+  ],
+  "down_block_types": [
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D"
+  ],
+  "in_channels": 2,
+  "latent_channels": 4,
+  "layers_per_block": 2,
+  "norm_num_groups": 32,
+  "out_channels": 2,
+  "sample_size": 512,
+  "up_block_types": [
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D"
+  ]
+}

models/stage1/StableDiffusion-FlowGen/vae_flow/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c95a62d729d5025c09022c7ed7bf14515ae8831e26195e1dd1abd0ab0969370d
+size 334692337