Upload folder using huggingface_hub

configs/datasets.yaml

@@ -0,0 +1,29 @@
+Defaults:
+  area_ratio: 0.01171
+  obj_thr: 2
+
+Train:
+  LVIS:
+    construct_dataset_dir: 'data/train/LVIS'
+    shards: "data/train/LVIS/{00000..00005}.tar"
+
+  VITONHD:
+    construct_dataset_dir: '/data/train/VITONHD'
+    shards: "data/train/VITONHD/{00000..00002}.tar"
+
+  Objects365:
+    construct_dataset_dir: 'data/train/Objects365'
+    # shards: "data/train/Objects365/{00000..00188}.tar"
+    shards: "data/train/Objects365/{00000..00020}.tar"
+
+  Cityscapes:
+    construct_dataset_dir: 'data/train/Cityscapes'
+    shards: "data/train/Cityscapes/00000.tar"
+
+  MapillaryVistas:
+    construct_dataset_dir: 'data/train/MapillaryVistas'
+    shards: "data/train/MapillaryVistas/00000.tar"
+
+  BDD100K:
+    construct_dataset_dir: 'data/train/BDD100K'
+    shards: "data/train/BDD100K/00000.tar"

configs/inference.yaml

@@ -0,0 +1,3 @@
+pretrained_model: checkpoints/pics-step=006000_ep18.ckpt
+config_file: configs/pics.yaml
+save_memory: False

configs/pics.yaml

@@ -0,0 +1,86 @@
+model:
+  target: cldm.cldm.ControlLDM
+  params:
+    linear_start: 0.00085
+    linear_end: 0.0120
+    num_timesteps_cond: 1
+    log_every_t: 200
+    timesteps: 1000
+    first_stage_key: "jpg"
+    cond_stage_key: "ref"
+    control_key: "hint"
+    image_size: 64
+    channels: 4
+    cond_stage_trainable: false
+    conditioning_key: crossattn
+    monitor: val/loss_simple_ema
+    scale_factor: 0.18215
+    use_ema: False
+    only_mid_control: False
+    obj_thr: 2
+
+    control_stage_config:
+      target: cldm.cldm.ControlNet
+      params:
+        use_checkpoint: True
+        image_size: 32 # unused
+        in_channels: 4 
+        hint_channels: 4
+        model_channels: 320
+        attention_resolutions: [ 4, 2, 1 ]
+        num_res_blocks: 2
+        channel_mult: [ 1, 2, 4, 4 ]
+        num_head_channels: 64 # need to fix for flash-attn
+        use_spatial_transformer: True
+        use_linear_in_transformer: True
+        transformer_depth: 1
+        context_dim: 1024
+        legacy: False
+
+    unet_config:
+      target: cldm.cldm.ControlledUnetModel
+      params:
+        use_checkpoint: True
+        image_size: 32 # unused
+        in_channels: 4
+        out_channels: 4
+        model_channels: 320
+        attention_resolutions: [ 4, 2, 1 ]
+        num_res_blocks: 2
+        channel_mult: [ 1, 2, 4, 4 ]
+        num_head_channels: 64 # need to fix for flash-attn
+        use_spatial_transformer: True
+        use_linear_in_transformer: True
+        transformer_depth: 1
+        context_dim: 1024
+        legacy: False
+
+    first_stage_config:
+      target: ldm.models.autoencoder.AutoencoderKL
+      params:
+        embed_dim: 4
+        monitor: val/rec_loss
+        ddconfig:
+          #attn_type: "vanilla-xformers"
+          double_z: true
+          z_channels: 4
+          resolution: 256
+          in_channels: 3
+          out_ch: 3
+          ch: 128
+          ch_mult:
+          - 1
+          - 2
+          - 4
+          - 4
+          num_res_blocks: 2
+          attn_resolutions: []
+          dropout: 0.0
+        lossconfig:
+          target: torch.nn.Identity
+
+    cond_stage_config:
+      target: ldm.modules.encoders.modules.FrozenDinoV2Encoder
+      weight: checkpoints/dinov2_vitg14_pretrain.pth
+
+