adding model files

global_sampler_clip/checkpoints/last.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a645439032025805fe831f28c92e45381a40ad9c2760a5a9bd518f05c1f559cb
+size 1508432949

global_sampler_clip/configs/2022-04-23T05-49-02-lightning.yaml

@@ -0,0 +1,4 @@
+lightning:
+  trainer:
+    distributed_backend: ddp
+    gpus: 0,1,2,3,4,5,6,7

global_sampler_clip/configs/2022-04-23T05-49-02-project.yaml

@@ -0,0 +1,62 @@
+model:
+  base_learning_rate: 4.5e-06
+  target: taming.models.global_sampler.GlobalSamplerWithCLIP
+  params:
+    cond_stage_key: knn
+    transformer_config:
+      target: taming.modules.transformer.mingpt.GPT
+      params:
+        vocab_size: 1024
+        block_size: 133
+        n_layer: 12
+        n_head: 8
+        n_embd: 1024
+    first_stage_config:
+      target: taming.models.vqgan.VQCirModel
+      params:
+        ckpt_path: 
+        embed_dim: 256
+        n_embed: 1024
+        ddconfig:
+          double_z: false
+          z_channels: 256
+          resolution: 256
+          in_channels: 3
+          out_ch: 3
+          ch: 128
+          ch_mult:
+          - 1
+          - 1
+          - 2
+          - 2
+          - 4
+          num_res_blocks: 2
+          attn_resolutions:
+          - 16
+          dropout: 0.0
+        lossconfig:
+          target: taming.modules.losses.DummyLoss
+    cond_stage_config:
+      target: taming.modules.misc.coord.CoordStage
+      params:
+        n_embed: 1024
+        down_factor: 16
+data:
+  target: main.DataModuleFromConfig
+  params:
+    batch_size: 16
+    num_workers: 8
+    train:
+      target: taming.data.custom.CustomTrainHolistic
+      params:
+        training_images_list_file: /home/chenzhaoxi/imgs/train_norot.txt
+        size: 256
+        holistic: 128
+        clip_emb: clip_emb.npy
+    validation:
+      target: taming.data.custom.CustomTestHolistic
+      params:
+        test_images_list_file: /home/chenzhaoxi/imgs/test.txt
+        size: 256
+        holistic: 128
+        clip_emb: clip_emb.npy

local_sampler/checkpoints/last.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5c1e693f346959b491d84f632c594c15ca88b9f4ec1210ca08f7a9f22b8ade8
+size 1812376146

local_sampler_indoor/checkpoints/last.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4440bb6c35d1d7f8b76cf615e334c7494e9eb3c9e06a9b049ff707c312f50e1c
+size 1812376146

local_sampler_indoor/configs/2022-04-26T13-11-31-lightning.yaml

@@ -0,0 +1,4 @@
+lightning:
+  trainer:
+    distributed_backend: ddp
+    gpus: 0,1,2,3,4,5,6,7

local_sampler_indoor/configs/2022-04-26T13-11-31-project.yaml

@@ -0,0 +1,89 @@
+model:
+  base_learning_rate: 4.5e-06
+  target: taming.models.local_sampler.LocalSamplerDualCodebookPE
+  params:
+    cond_stage_key: coord
+    holistic_config:
+      target: taming.models.vqgan.VQCirModel
+      params:
+        ckpt_path: 
+        embed_dim: 256
+        n_embed: 1024
+        ddconfig:
+          double_z: false
+          z_channels: 256
+          resolution: 256
+          in_channels: 3
+          out_ch: 3
+          ch: 128
+          ch_mult:
+          - 1
+          - 1
+          - 2
+          - 2
+          - 4
+          num_res_blocks: 2
+          attn_resolutions:
+          - 16
+          dropout: 0.0
+        lossconfig:
+          target: taming.modules.losses.DummyLoss
+    transformer_config:
+      target: taming.modules.transformer.mingpt.GPT
+      params:
+        vocab_size: 1024
+        block_size: 402
+        n_layer: 24
+        n_head: 16
+        n_embd: 1024
+    first_stage_config:
+      target: taming.models.vqgan.VQModel
+      params:
+        ckpt_path: 
+        embed_dim: 256
+        n_embed: 1024
+        ddconfig:
+          double_z: false
+          z_channels: 256
+          resolution: 256
+          in_channels: 3
+          out_ch: 3
+          ch: 128
+          ch_mult:
+          - 1
+          - 1
+          - 2
+          - 2
+          - 4
+          num_res_blocks: 2
+          attn_resolutions:
+          - 16
+          dropout: 0.0
+        lossconfig:
+          target: taming.modules.losses.DummyLoss
+    cond_stage_config:
+      target: taming.modules.misc.coord.CoordStage
+      params:
+        n_embed: 1024
+        down_factor: 16
+data:
+  target: main.DataModuleFromConfig
+  params:
+    batch_size: 24
+    num_workers: 16
+    train:
+      target: taming.data.custom.CustomTrain
+      params:
+        training_images_list_file: /home/chenzhaoxi/imgs/train_indoor.txt
+        size: 256
+        coord: true
+        random_crop: true
+        holistic: 128
+    validation:
+      target: taming.data.custom.CustomTest
+      params:
+        test_images_list_file: /home/chenzhaoxi/imgs/test.txt
+        size: 256
+        coord: true
+        random_crop: true
+        holistic: 128

local_sampler_outdoor/checkpoints/last.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3172f82fe835531bdecf643bc0b538aae85165ff974b0342a972f3c3a81e920e
+size 1812376146

local_sampler_outdoor/configs/2022-04-29T20-25-45-lightning.yaml

@@ -0,0 +1,4 @@
+lightning:
+  trainer:
+    distributed_backend: ddp
+    gpus: 0,1,2,3,4,5,6,7

local_sampler_outdoor/configs/2022-04-29T20-25-45-project.yaml

@@ -0,0 +1,89 @@
+model:
+  base_learning_rate: 4.5e-06
+  target: taming.models.local_sampler.LocalSamplerDualCodebookPE
+  params:
+    cond_stage_key: coord
+    holistic_config:
+      target: taming.models.vqgan.VQCirModel
+      params:
+        ckpt_path: 
+        embed_dim: 256
+        n_embed: 1024
+        ddconfig:
+          double_z: false
+          z_channels: 256
+          resolution: 256
+          in_channels: 3
+          out_ch: 3
+          ch: 128
+          ch_mult:
+          - 1
+          - 1
+          - 2
+          - 2
+          - 4
+          num_res_blocks: 2
+          attn_resolutions:
+          - 16
+          dropout: 0.0
+        lossconfig:
+          target: taming.modules.losses.DummyLoss
+    transformer_config:
+      target: taming.modules.transformer.mingpt.GPT
+      params:
+        vocab_size: 1024
+        block_size: 402
+        n_layer: 24
+        n_head: 16
+        n_embd: 1024
+    first_stage_config:
+      target: taming.models.vqgan.VQModel
+      params:
+        ckpt_path: 
+        embed_dim: 256
+        n_embed: 1024
+        ddconfig:
+          double_z: false
+          z_channels: 256
+          resolution: 256
+          in_channels: 3
+          out_ch: 3
+          ch: 128
+          ch_mult:
+          - 1
+          - 1
+          - 2
+          - 2
+          - 4
+          num_res_blocks: 2
+          attn_resolutions:
+          - 16
+          dropout: 0.0
+        lossconfig:
+          target: taming.modules.losses.DummyLoss
+    cond_stage_config:
+      target: taming.modules.misc.coord.CoordStage
+      params:
+        n_embed: 1024
+        down_factor: 16
+data:
+  target: main.DataModuleFromConfig
+  params:
+    batch_size: 24
+    num_workers: 16
+    train:
+      target: taming.data.custom.CustomTrain
+      params:
+        training_images_list_file: /home/chenzhaoxi/imgs/train_outdoor.txt
+        size: 256
+        coord: true
+        random_crop: true
+        holistic: 128
+    validation:
+      target: taming.data.custom.CustomTest
+      params:
+        test_images_list_file: /home/chenzhaoxi/imgs/test.txt
+        size: 256
+        coord: true
+        random_crop: true
+        holistic: 128

sritmo.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab14c0e0a3f810eddcc04f51457fa2e9ca5443670b1620f7b743234c70b2858c
+size 20491375