Upload folder using huggingface_hub

config.json

@@ -1,10 +1,10 @@
 {
-    "type": "diffusion",
-    "n_obs_steps": 2,
+    "type": "act",
+    "n_obs_steps": 1,
     "normalization_mapping": {
         "VISUAL": "MEAN_STD",
-        "STATE": "MIN_MAX",
-        "ACTION": "MIN_MAX"
+        "STATE": "MEAN_STD",
+        "ACTION": "MEAN_STD"
     },
     "input_features": {
         "observation.state": {
@@ -13,7 +13,7 @@
                 6
             ]
         },
-        "observation.images.side": {
+        "observation.images.handeye": {
             "type": "VISUAL",
             "shape": [
                 3,
@@ -21,7 +21,7 @@
                 640
             ]
         },
-        "observation.images.wrist": {
+        "observation.images.side": {
             "type": "VISUAL",
             "shape": [
                 3,
@@ -48,45 +48,25 @@
     },
     "device": "cuda",
     "use_amp": false,
-    "horizon": 16,
-    "n_action_steps": 8,
-    "drop_n_last_frames": 7,
+    "chunk_size": 100,
+    "n_action_steps": 100,
     "vision_backbone": "resnet18",
-    "crop_shape": [
-        84,
-        84
-    ],
-    "crop_is_random": true,
-    "pretrained_backbone_weights": null,
-    "use_group_norm": true,
-    "spatial_softmax_num_keypoints": 32,
-    "use_separate_rgb_encoder_per_camera": false,
-    "down_dims": [
-        512,
-        1024,
-        2048
-    ],
-    "kernel_size": 5,
-    "n_groups": 8,
-    "diffusion_step_embed_dim": 128,
-    "use_film_scale_modulation": true,
-    "noise_scheduler_type": "DDPM",
-    "num_train_timesteps": 100,
-    "beta_schedule": "squaredcos_cap_v2",
-    "beta_start": 0.0001,
-    "beta_end": 0.02,
-    "prediction_type": "epsilon",
-    "clip_sample": true,
-    "clip_sample_range": 1.0,
-    "num_inference_steps": null,
-    "do_mask_loss_for_padding": false,
-    "optimizer_lr": 0.0001,
-    "optimizer_betas": [
-        0.95,
-        0.999
-    ],
-    "optimizer_eps": 1e-08,
-    "optimizer_weight_decay": 1e-06,
-    "scheduler_name": "cosine",
-    "scheduler_warmup_steps": 500
+    "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
+    "replace_final_stride_with_dilation": false,
+    "pre_norm": false,
+    "dim_model": 512,
+    "n_heads": 8,
+    "dim_feedforward": 3200,
+    "feedforward_activation": "relu",
+    "n_encoder_layers": 4,
+    "n_decoder_layers": 1,
+    "use_vae": true,
+    "latent_dim": 32,
+    "n_vae_encoder_layers": 4,
+    "temporal_ensemble_coeff": null,
+    "dropout": 0.1,
+    "kl_weight": 10.0,
+    "optimizer_lr": 1e-05,
+    "optimizer_weight_decay": 0.0001,
+    "optimizer_lr_backbone": 1e-05
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5bf47709f69c24494d92aa825696a1c533cd6f7b907ac4a5c4c3b265a86a710f
-size 1081198080
+oid sha256:7a58b510b038c0f927f75b2c10427ecd2b3cff6a155fcc9e62a991df2564ea37
+size 206701344

train_config.json

@@ -1,6 +1,6 @@
 {
     "dataset": {
-        "repo_id": "Rorschach4153/so101_30_1",
+        "repo_id": "pbvr/so101_test014",
         "root": null,
         "episodes": null,
         "image_transforms": {
@@ -66,12 +66,12 @@
     },
     "env": null,
     "policy": {
-        "type": "diffusion",
-        "n_obs_steps": 2,
+        "type": "act",
+        "n_obs_steps": 1,
         "normalization_mapping": {
             "VISUAL": "MEAN_STD",
-            "STATE": "MIN_MAX",
-            "ACTION": "MIN_MAX"
+            "STATE": "MEAN_STD",
+            "ACTION": "MEAN_STD"
         },
         "input_features": {
             "observation.state": {
@@ -80,7 +80,7 @@
                     6
                 ]
             },
-            "observation.images.side": {
+            "observation.images.handeye": {
                 "type": "VISUAL",
                 "shape": [
                     3,
@@ -88,7 +88,7 @@
                     640
                 ]
             },
-            "observation.images.wrist": {
+            "observation.images.side": {
                 "type": "VISUAL",
                 "shape": [
                     3,
@@ -115,50 +115,30 @@
         },
         "device": "cuda",
         "use_amp": false,
-        "horizon": 16,
-        "n_action_steps": 8,
-        "drop_n_last_frames": 7,
+        "chunk_size": 100,
+        "n_action_steps": 100,
         "vision_backbone": "resnet18",
-        "crop_shape": [
-            84,
-            84
-        ],
-        "crop_is_random": true,
-        "pretrained_backbone_weights": null,
-        "use_group_norm": true,
-        "spatial_softmax_num_keypoints": 32,
-        "use_separate_rgb_encoder_per_camera": false,
-        "down_dims": [
-            512,
-            1024,
-            2048
-        ],
-        "kernel_size": 5,
-        "n_groups": 8,
-        "diffusion_step_embed_dim": 128,
-        "use_film_scale_modulation": true,
-        "noise_scheduler_type": "DDPM",
-        "num_train_timesteps": 100,
-        "beta_schedule": "squaredcos_cap_v2",
-        "beta_start": 0.0001,
-        "beta_end": 0.02,
-        "prediction_type": "epsilon",
-        "clip_sample": true,
-        "clip_sample_range": 1.0,
-        "num_inference_steps": null,
-        "do_mask_loss_for_padding": false,
-        "optimizer_lr": 0.0001,
-        "optimizer_betas": [
-            0.95,
-            0.999
-        ],
-        "optimizer_eps": 1e-08,
-        "optimizer_weight_decay": 1e-06,
-        "scheduler_name": "cosine",
-        "scheduler_warmup_steps": 500
+        "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
+        "replace_final_stride_with_dilation": false,
+        "pre_norm": false,
+        "dim_model": 512,
+        "n_heads": 8,
+        "dim_feedforward": 3200,
+        "feedforward_activation": "relu",
+        "n_encoder_layers": 4,
+        "n_decoder_layers": 1,
+        "use_vae": true,
+        "latent_dim": 32,
+        "n_vae_encoder_layers": 4,
+        "temporal_ensemble_coeff": null,
+        "dropout": 0.1,
+        "kl_weight": 10.0,
+        "optimizer_lr": 1e-05,
+        "optimizer_weight_decay": 0.0001,
+        "optimizer_lr_backbone": 1e-05
     },
-    "output_dir": "outputs/train/act_so101_test_09",
-    "job_name": "act_so101_test",
+    "output_dir": "outputs/train/act_so101_test_50",
+    "job_name": "act_so101_test_01",
     "resume": false,
     "seed": 1000,
     "num_workers": 4,
@@ -170,21 +150,17 @@
     "save_freq": 20000,
     "use_policy_training_preset": true,
     "optimizer": {
-        "type": "adam",
-        "lr": 0.0001,
-        "weight_decay": 1e-06,
+        "type": "adamw",
+        "lr": 1e-05,
+        "weight_decay": 0.0001,
         "grad_clip_norm": 10.0,
         "betas": [
-            0.95,
+            0.9,
             0.999
         ],
         "eps": 1e-08
     },
-    "scheduler": {
-        "type": "diffuser",
-        "num_warmup_steps": 500,
-        "name": "cosine"
-    },
+    "scheduler": null,
     "eval": {
         "n_episodes": 50,
         "batch_size": 50,