Upload checkpoint at step 40000

stage1_vae_diffusion_test3/step_040000/condition_encoder.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddefc26c56dd07939e7409a05021522279555f0246f3c3d3c4af8da4ee304d61
+size 1738104

stage1_vae_diffusion_test3/step_040000/config.json

@@ -0,0 +1,39 @@
+{
+  "parquet_path": "datasets/diffusion_stage1_train_10pct_merged.parquet",
+  "val_parquet_path": "datasets/diffusion_stage1_validation_1pct.parquet",
+  "output_dir": "checkpoints/diffusion_stage1",
+  "image_size": 512,
+  "max_transitions": -1,
+  "val_max_transitions": -1,
+  "vae_path": "Qwen/Qwen-Image",
+  "vae_subfolder": "vae",
+  "vae_variant": "qwen",
+  "condition_dim": 512,
+  "attention_head_dim": 64,
+  "scheduler_num_steps": 1000,
+  "num_workers": 8,
+  "batch_size": 2,
+  "gradient_accumulation_steps": 1,
+  "num_epochs": 10,
+  "learning_rate": 0.0001,
+  "weight_decay": 0.01,
+  "mixed_precision": "bf16",
+  "max_grad_norm": 1.0,
+  "vae_recon_weight": 10.0,
+  "latent_diffusion_weight": 1.0,
+  "log_interval": 10,
+  "val_interval": 100,
+  "save_interval": 500,
+  "use_wandb": true,
+  "wandb_project": "latent_agent_diffusion",
+  "wandb_run_name": "stage1_vae_diffusion_test3",
+  "wandb_entity": "",
+  "use_lora": false,
+  "lora_rank": 8,
+  "lora_alpha": 16.0,
+  "lora_weights_path": "",
+  "lora_subfolder": "",
+  "hf_upload": true,
+  "hf_repo_id": "KoeYe/latent-agent-diffusion-stage1",
+  "hf_upload_every_n_steps": 1000
+}

stage1_vae_diffusion_test3/step_040000/unet/config.json

@@ -0,0 +1,67 @@
+{
+  "_class_name": "UNet2DConditionModel",
+  "_diffusers_version": "0.35.2",
+  "act_fn": "silu",
+  "addition_embed_type": null,
+  "addition_embed_type_num_heads": 64,
+  "addition_time_embed_dim": null,
+  "attention_head_dim": 64,
+  "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": 512,
+  "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": 16,
+  "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": 16,
+  "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": 64,
+  "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": false
+}

stage1_vae_diffusion_test3/step_040000/unet/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36b444589b96b5e02aa43555fca7b7f6db054eb26491178f60c936576e1bff8b
+size 1706484728