Upload policy weights, train config and readme

README.md

@@ -2,20 +2,20 @@
 datasets: Deason11/pick_the_blue_cuboid_and_put_it_into_the_box
 library_name: lerobot
 license: apache-2.0
-model_name: pi0
+model_name: act
 pipeline_tag: robotics
 tags:
-- pi0
-- robotics
 - lerobot
+- robotics
+- act
 ---
 
-# Model Card for pi0
+# Model Card for act
 
 <!-- Provide a quick summary of what the model is/does. -->
 
 
-[Pi0](https://huggingface.co/papers/2410.24164) is a generalist vision-language-action transformer that converts multimodal observations and text instructions into robot actions for zero-shot task transfer.
+[Action Chunking with Transformers (ACT)](https://huggingface.co/papers/2304.13705) is an imitation-learning method that predicts short action chunks instead of single steps. It learns from teleoperated data and often achieves high success rates.
 
 
 This policy has been trained and pushed to the Hub using [LeRobot](https://github.com/huggingface/lerobot).

config.json

@@ -1,8 +1,8 @@
 {
-    "type": "pi0",
+    "type": "act",
     "n_obs_steps": 1,
     "normalization_mapping": {
-        "VISUAL": "IDENTITY",
+        "VISUAL": "MEAN_STD",
         "STATE": "MEAN_STD",
         "ACTION": "MEAN_STD"
     },
@@ -45,33 +45,25 @@
     "private": null,
     "tags": null,
     "license": null,
-    "chunk_size": 50,
-    "n_action_steps": 50,
-    "max_state_dim": 32,
-    "max_action_dim": 32,
-    "resize_imgs_with_padding": [
-        224,
-        224
-    ],
-    "empty_cameras": 0,
-    "adapt_to_pi_aloha": false,
-    "use_delta_joint_actions_aloha": false,
-    "tokenizer_max_length": 48,
-    "proj_width": 1024,
-    "num_steps": 10,
-    "use_cache": true,
-    "attention_implementation": "eager",
-    "freeze_vision_encoder": true,
-    "train_expert_only": true,
-    "train_state_proj": true,
-    "optimizer_lr": 2.5e-05,
-    "optimizer_betas": [
-        0.9,
-        0.95
-    ],
-    "optimizer_eps": 1e-08,
-    "optimizer_weight_decay": 1e-10,
-    "scheduler_warmup_steps": 1000,
-    "scheduler_decay_steps": 30000,
-    "scheduler_decay_lr": 2.5e-06
+    "chunk_size": 100,
+    "n_action_steps": 100,
+    "vision_backbone": "resnet18",
+    "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 sha256:0cea14bb651f84eb15719b3f48006ec3c3a8cc18ea1c560ccf611174d971faac
-size 7536025176
+oid sha256:2e28263eaf07f46e83e66418c5e1627e6e2773588aff099d3d559f6870556e7d
+size 206709364

train_config.json

@@ -66,10 +66,10 @@
     },
     "env": null,
     "policy": {
-        "type": "pi0",
+        "type": "act",
         "n_obs_steps": 1,
         "normalization_mapping": {
-            "VISUAL": "IDENTITY",
+            "VISUAL": "MEAN_STD",
             "STATE": "MEAN_STD",
             "ACTION": "MEAN_STD"
         },
@@ -112,38 +112,30 @@
         "private": null,
         "tags": null,
         "license": null,
-        "chunk_size": 50,
-        "n_action_steps": 50,
-        "max_state_dim": 32,
-        "max_action_dim": 32,
-        "resize_imgs_with_padding": [
-            224,
-            224
-        ],
-        "empty_cameras": 0,
-        "adapt_to_pi_aloha": false,
-        "use_delta_joint_actions_aloha": false,
-        "tokenizer_max_length": 48,
-        "proj_width": 1024,
-        "num_steps": 10,
-        "use_cache": true,
-        "attention_implementation": "eager",
-        "freeze_vision_encoder": true,
-        "train_expert_only": true,
-        "train_state_proj": true,
-        "optimizer_lr": 2.5e-05,
-        "optimizer_betas": [
-            0.9,
-            0.95
-        ],
-        "optimizer_eps": 1e-08,
-        "optimizer_weight_decay": 1e-10,
-        "scheduler_warmup_steps": 1000,
-        "scheduler_decay_steps": 30000,
-        "scheduler_decay_lr": 2.5e-06
+        "chunk_size": 100,
+        "n_action_steps": 100,
+        "vision_backbone": "resnet18",
+        "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/2025-08-08/21-24-14_pi0",
-    "job_name": "pi0",
+    "output_dir": "outputs/train/2025-08-10/20-24-41_act",
+    "job_name": "act",
     "resume": false,
     "seed": 1000,
     "num_workers": 4,
@@ -156,22 +148,16 @@
     "use_policy_training_preset": true,
     "optimizer": {
         "type": "adamw",
-        "lr": 2.5e-05,
-        "weight_decay": 1e-10,
+        "lr": 1e-05,
+        "weight_decay": 0.0001,
         "grad_clip_norm": 10.0,
         "betas": [
             0.9,
-            0.95
+            0.999
         ],
         "eps": 1e-08
     },
-    "scheduler": {
-        "type": "cosine_decay_with_warmup",
-        "num_warmup_steps": 1000,
-        "num_decay_steps": 30000,
-        "peak_lr": 2.5e-05,
-        "decay_lr": 2.5e-06
-    },
+    "scheduler": null,
     "eval": {
         "n_episodes": 50,
         "batch_size": 50,
@@ -183,7 +169,7 @@
         "project": "lerobot",
         "entity": null,
         "notes": null,
-        "run_id": "1u4th40g",
+        "run_id": "j4ecaamf",
         "mode": null
     }
 }