Upload folder using huggingface_hub

checkpoint/cld/cld_clear.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1954f732568e9d6b5b5dea482680ccb725670558378fdf7ef8164bfca73eb1b0
+size 783888141

checkpoint/diffusion/unet1/config.json

@@ -0,0 +1,73 @@
+{
+  "_class_name": "UNet2DConditionModel",
+  "_diffusers_version": "0.30.3",
+  "_name_or_path": "/fs/nexus-projects/ViewDreamer/U_Data_Gen/models/stable-diffusion-2",
+  "act_fn": "silu",
+  "addition_embed_type": null,
+  "addition_embed_type_num_heads": 64,
+  "addition_time_embed_dim": null,
+  "attention_head_dim": [
+    5,
+    10,
+    20,
+    20
+  ],
+  "attention_type": "default",
+  "block_out_channels": [
+    320,
+    640,
+    1280,
+    1280
+  ],
+  "center_input_sample": false,
+  "class_embed_type": null,
+  "class_embeddings_concat": false,
+  "conv_in_kernel": 3,
+  "conv_out_kernel": 3,
+  "cross_attention_dim": 1024,
+  "cross_attention_norm": null,
+  "down_block_types": [
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D",
+    "DownBlock2D"
+  ],
+  "downsample_padding": 1,
+  "dropout": 0.0,
+  "dual_cross_attention": false,
+  "encoder_hid_dim": null,
+  "encoder_hid_dim_type": null,
+  "flip_sin_to_cos": true,
+  "freq_shift": 0,
+  "in_channels": 8,
+  "layers_per_block": 2,
+  "mid_block_only_cross_attention": null,
+  "mid_block_scale_factor": 1,
+  "mid_block_type": "UNetMidBlock2DCrossAttn",
+  "norm_eps": 1e-05,
+  "norm_num_groups": 32,
+  "num_attention_heads": null,
+  "num_class_embeds": null,
+  "only_cross_attention": false,
+  "out_channels": 4,
+  "projection_class_embeddings_input_dim": null,
+  "resnet_out_scale_factor": 1.0,
+  "resnet_skip_time_act": false,
+  "resnet_time_scale_shift": "default",
+  "reverse_transformer_layers_per_block": null,
+  "sample_size": 96,
+  "time_cond_proj_dim": null,
+  "time_embedding_act_fn": null,
+  "time_embedding_dim": null,
+  "time_embedding_type": "positional",
+  "timestep_post_act": null,
+  "transformer_layers_per_block": 1,
+  "up_block_types": [
+    "UpBlock2D",
+    "CrossAttnUpBlock2D",
+    "CrossAttnUpBlock2D",
+    "CrossAttnUpBlock2D"
+  ],
+  "upcast_attention": false,
+  "use_linear_projection": true
+}

checkpoint/diffusion/unet1/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62e0ccc6fd2a972e6e769b809675f44da047394e436416ae8ee9ddd67457ed1c
+size 3470487155

checkpoint/diffusion/unet2/config.json

@@ -0,0 +1,73 @@
+{
+  "_class_name": "UNet2DConditionModel",
+  "_diffusers_version": "0.30.3",
+  "_name_or_path": "/fs/nexus-projects/ViewDreamer/U_Data_Gen/models/marigold-depth-v1-0",
+  "act_fn": "silu",
+  "addition_embed_type": null,
+  "addition_embed_type_num_heads": 64,
+  "addition_time_embed_dim": null,
+  "attention_head_dim": [
+    5,
+    10,
+    20,
+    20
+  ],
+  "attention_type": "default",
+  "block_out_channels": [
+    320,
+    640,
+    1280,
+    1280
+  ],
+  "center_input_sample": false,
+  "class_embed_type": null,
+  "class_embeddings_concat": false,
+  "conv_in_kernel": 3,
+  "conv_out_kernel": 3,
+  "cross_attention_dim": 1024,
+  "cross_attention_norm": null,
+  "down_block_types": [
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D",
+    "DownBlock2D"
+  ],
+  "downsample_padding": 1,
+  "dropout": 0.0,
+  "dual_cross_attention": false,
+  "encoder_hid_dim": null,
+  "encoder_hid_dim_type": null,
+  "flip_sin_to_cos": true,
+  "freq_shift": 0,
+  "in_channels": 12,
+  "layers_per_block": 2,
+  "mid_block_only_cross_attention": null,
+  "mid_block_scale_factor": 1,
+  "mid_block_type": "UNetMidBlock2DCrossAttn",
+  "norm_eps": 1e-05,
+  "norm_num_groups": 32,
+  "num_attention_heads": null,
+  "num_class_embeds": null,
+  "only_cross_attention": false,
+  "out_channels": 8,
+  "projection_class_embeddings_input_dim": null,
+  "resnet_out_scale_factor": 1.0,
+  "resnet_skip_time_act": false,
+  "resnet_time_scale_shift": "default",
+  "reverse_transformer_layers_per_block": null,
+  "sample_size": 96,
+  "time_cond_proj_dim": null,
+  "time_embedding_act_fn": null,
+  "time_embedding_dim": null,
+  "time_embedding_type": "positional",
+  "timestep_post_act": null,
+  "transformer_layers_per_block": 1,
+  "up_block_types": [
+    "UpBlock2D",
+    "CrossAttnUpBlock2D",
+    "CrossAttnUpBlock2D",
+    "CrossAttnUpBlock2D"
+  ],
+  "upcast_attention": false,
+  "use_linear_projection": true
+}

checkpoint/diffusion/unet2/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:300c1fd66d4fffc8160fc84dda498bf13862c52db0b5d94e4e2bc83e962c94a7
+size 3470579315

config.yaml

@@ -0,0 +1,74 @@
+logging:
+  filename: logging.log
+  format: ' %(asctime)s - %(levelname)s -%(filename)s - %(funcName)s >> %(message)s'
+  console_level: 20
+  file_level: 10
+wandb:
+  project: slurpp
+  reinit: true
+model:
+  name: slurpp_pipeline
+  pretrained_path: stable-diffusion-2
+  latent_scale_factor: 0.18215
+  pretrained_path2: marigold-depth-v1-0
+base_config:
+- ../config/default.yaml
+pipeline:
+  name: SlurppPipeline
+job_name: slurpp_release
+dataloader:
+  num_workers: 2
+  effective_batch_size: 1
+  max_train_batch_size: 1
+  seed: 2024
+  val_size: 20
+  train_vis_size: 20
+  image_size: 512
+trainer:
+  name: SlurppTrainer
+  training_noise_scheduler:
+    pretrained_path: stable-diffusion-2
+  init_seed: 2024
+  val_vis_frequency: 1
+  num_hour_between_val: 2
+  color_space: srgb
+  inputs:
+  - u
+  inputs2:
+  - u
+  output:
+  - clear
+  output2:
+  - bc
+  - ill
+  upsample: 1
+  num_hour_between_real_vis: 2
+max_epoch: 10000
+max_iter: 100000
+optimizer:
+  name: Adam
+loss:
+  name: mse_loss
+  kwargs:
+    reduction: mean
+lr: 1.0e-05
+lr_scheduler:
+  name: IterExponential
+  kwargs:
+    total_iter: 50000
+    final_ratio: 0.01
+    warmup_steps: 100
+validation:
+  denoising_steps: 50
+  main_val_metric: psnr
+  main_val_metric_goal: minimize
+  init_seed: 2024
+eval:
+  eval_metrics:
+  - psnr
+  - ssim
+  - lpips_loss
+dual: true
+rgb_loss: true
+reconstruction_loss: true
+one_step: true