Upload CIFAR-10 checkpoint and config (2025-09-13T16:23:39.476367Z)

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+ckpt-150000.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text

ckpt-150000.data-00000-of-00001

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca88d7fd9f9e32d5d52e171c2b57aeef05262866ed29bebed957969f0daf19ff
+size 495423956

config.json

@@ -0,0 +1,124 @@
+{
+  "dataset": {
+    "batch_duplicates": 1,
+    "cache_dataset": true,
+    "cropping": "none",
+    "eval_split": "test",
+    "flipping": "left_right",
+    "image_size": 32,
+    "name": "object_recognition",
+    "num_classes": 10,
+    "tfds_name": "cifar10",
+    "train_split": "train"
+  },
+  "datasets": [
+    {
+      "batch_duplicates": 1,
+      "cache_dataset": true,
+      "cropping": "none",
+      "eval_split": "test",
+      "flipping": "left_right",
+      "image_size": 32,
+      "name": "object_recognition",
+      "num_classes": 10,
+      "tfds_name": "cifar10",
+      "train_split": "train"
+    }
+  ],
+  "eval": {
+    "batch_size": 64,
+    "checkpoint_dir": "",
+    "steps": 2,
+    "tag": "eval",
+    "write_images_to_file": true
+  },
+  "model": {
+    "arch_name": "tape",
+    "b_scale": 1.0,
+    "b_type": "uint8",
+    "cond_decoupled_read": false,
+    "cond_dim": 0,
+    "cond_dropout": 0.0,
+    "cond_on_latent": true,
+    "cond_proj": true,
+    "cond_tape_writable": false,
+    "conditional": "class",
+    "conv_drop_units": 0.0,
+    "conv_kernel_size": 0,
+    "drop_att": 0.0,
+    "drop_path": 0.1,
+    "drop_sc": 0.0,
+    "drop_units": 0.1,
+    "flip_rate": 0.0,
+    "guidance": 0.0,
+    "infer_iterations": 100,
+    "infer_schedule": "cosine",
+    "latent_dim": 512,
+    "latent_mlp_ratio": 4,
+    "latent_num_heads": 16,
+    "latent_pos_encoding": "learned",
+    "latent_slots": 128,
+    "loss_type": "eps",
+    "name": "image_diffusion_model",
+    "normalize_noisy_input": false,
+    "num_layers": "2,2,2",
+    "patch_size": 2,
+    "pos_encoding": "sin_cos",
+    "pred_type": "eps",
+    "pretrained_ckpt": "",
+    "rw_num_heads": 8,
+    "sampler_name": "ddpm",
+    "self_cond": "latent",
+    "self_cond_by_masking": true,
+    "self_cond_rate": 0.9,
+    "tape_dim": 256,
+    "tape_mlp_ratio": 2,
+    "tape_pos_encoding": "learned",
+    "td": 0.0,
+    "time_on_latent": true,
+    "time_scaling": 1000,
+    "train_schedule": "sigmoid@-3,3,0.9",
+    "use_cls_token": false,
+    "x0_clip": "auto",
+    "xattn_enc_ln": false
+  },
+  "model_dir": "results/cifar10",
+  "optimization": {
+    "beta1": 0.9,
+    "beta2": 0.999,
+    "ema_decay": 0.9999,
+    "ema_name_exact_match": true,
+    "end_lr_factor": 0.0,
+    "eps": 1e-08,
+    "exclude_from_weight_decay": "bias,beta,gamma",
+    "global_clipnorm": 1.0,
+    "learning_rate": 0.003,
+    "learning_rate_scaling": "none",
+    "learning_rate_schedule": "cosine@0.8",
+    "momentum": 0.9,
+    "optimizer": "lamb",
+    "tail_steps": 0,
+    "warmup_epochs": 0,
+    "warmup_steps": 10000,
+    "weight_decay": 0.01
+  },
+  "task": {
+    "name": "image_generation",
+    "weight": 1.0
+  },
+  "tasks": [
+    {
+      "name": "image_generation",
+      "weight": 1.0
+    }
+  ],
+  "train": {
+    "batch_size": 256,
+    "checkpoint_epochs": 40,
+    "checkpoint_steps": 1000,
+    "epochs": 100,
+    "keep_checkpoint_max": 10,
+    "label_smoothing": 0.0,
+    "steps": 150000
+  }
+}

events.out.tfevents.1757712604.viscam-hgx-2.stanford.edu.2124946.0.v2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7836654b71b5840b1dcf6d1a68de9331ae80313d9d6f3f39c1723138f62465dd
+size 84116