diff --git a/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/config.json b/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..51a27627806ad961fce2c5204c9d1fa286e4c5ff --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/config.json @@ -0,0 +1,92 @@ +{ + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/model.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..22463c02ca030e23491ded8d309867c555df10f1 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bd803a592fc7659e0ccefa374fb48a9cfb9c5e3802e22316ac8606b6048dc64e +size 1067235520 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/policy_postprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..7f3d1455ad86832c79e49187567da5d47aa068e7 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/policy_postprocessor.json @@ -0,0 +1,32 @@ +{ + "name": "policy_postprocessor", + "steps": [ + { + "registry_name": "unnormalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors" + }, + { + "registry_name": "device_processor", + "config": { + "device": "cpu", + "float_dtype": null + } + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/policy_preprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..c7ddebe581f2032f63d45ea5161a28bc1de8194a --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/policy_preprocessor.json @@ -0,0 +1,64 @@ +{ + "name": "policy_preprocessor", + "steps": [ + { + "registry_name": "rename_observations_processor", + "config": { + "rename_map": {} + } + }, + { + "registry_name": "to_batch_processor", + "config": {} + }, + { + "registry_name": "device_processor", + "config": { + "device": "cuda", + "float_dtype": null + } + }, + { + "registry_name": "normalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/train_config.json b/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..6dbcbb45cc6ab3592379df4548bee3518cae49c4 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/020000/pretrained_model/train_config.json @@ -0,0 +1,226 @@ +{ + "dataset": { + "repo_id": "AI-Final/AI-aiCapstoneData-lerobot-advanced-v2-2", + "root": null, + "episodes": null, + "image_transforms": { + "enable": false, + "max_num_transforms": 3, + "random_order": false, + "tfs": { + "brightness": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "brightness": [ + 0.8, + 1.2 + ] + } + }, + "contrast": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "contrast": [ + 0.8, + 1.2 + ] + } + }, + "saturation": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "saturation": [ + 0.5, + 1.5 + ] + } + }, + "hue": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "hue": [ + -0.05, + 0.05 + ] + } + }, + "sharpness": { + "weight": 1.0, + "type": "SharpnessJitter", + "kwargs": { + "sharpness": [ + 0.5, + 1.5 + ] + } + }, + "affine": { + "weight": 1.0, + "type": "RandomAffine", + "kwargs": { + "degrees": [ + -5.0, + 5.0 + ], + "translate": [ + 0.05, + 0.05 + ] + } + } + } + }, + "revision": "main", + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 + }, + "output_dir": "outputs/train/act_dining_cleanup_v2_2_20260609_033312", + "job_name": "act_dining_cleanup_v2_2_20260609_033312", + "resume": false, + "seed": 1000, + "num_workers": 4, + "batch_size": 16, + "steps": 200000, + "eval_freq": 0, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 20000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adam", + "lr": 0.0001, + "weight_decay": 1e-06, + "grad_clip_norm": 10.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": { + "type": "diffuser", + "num_warmup_steps": 500, + "name": "cosine" + }, + "eval": { + "n_episodes": 50, + "batch_size": 50, + "use_async_envs": false + }, + "wandb": { + "enable": false, + "disable_artifact": false, + "project": "lerobot", + "entity": null, + "notes": null, + "run_id": null, + "mode": null + }, + "peft": null, + "use_rabc": false, + "rabc_progress_path": null, + "rabc_kappa": 0.01, + "rabc_epsilon": 1e-06, + "rabc_head_mode": "sparse", + "rename_map": {}, + "checkpoint_path": null +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/020000/training_state/optimizer_param_groups.json b/training_runs/advanced-diffusion-v3/checkpoints/020000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..7336261fe7b0e542992ec11b1b75c627acf0d327 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/020000/training_state/optimizer_param_groups.json @@ -0,0 +1,233 @@ +[ + { + "lr": 9.766112029998846e-05, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 1e-06, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": false, + "initial_lr": 0.0001, + "params": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19, + 20, + 21, + 22, + 23, + 24, + 25, + 26, + 27, + 28, + 29, + 30, + 31, + 32, + 33, + 34, + 35, + 36, + 37, + 38, + 39, + 40, + 41, + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61, + 62, + 63, + 64, + 65, + 66, + 67, + 68, + 69, + 70, + 71, + 72, + 73, + 74, + 75, + 76, + 77, + 78, + 79, + 80, + 81, + 82, + 83, + 84, + 85, + 86, + 87, + 88, + 89, + 90, + 91, + 92, + 93, + 94, + 95, + 96, + 97, + 98, + 99, + 100, + 101, + 102, + 103, + 104, + 105, + 106, + 107, + 108, + 109, + 110, + 111, + 112, + 113, + 114, + 115, + 116, + 117, + 118, + 119, + 120, + 121, + 122, + 123, + 124, + 125, + 126, + 127, + 128, + 129, + 130, + 131, + 132, + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152, + 153, + 154, + 155, + 156, + 157, + 158, + 159, + 160, + 161, + 162, + 163, + 164, + 165, + 166, + 167, + 168, + 169, + 170, + 171, + 172, + 173, + 174, + 175, + 176, + 177, + 178, + 179, + 180, + 181, + 182, + 183, + 184, + 185, + 186, + 187, + 188, + 189, + 190, + 191, + 192, + 193, + 194, + 195, + 196, + 197, + 198, + 199, + 200, + 201, + 202, + 203, + 204, + 205, + 206, + 207, + 208, + 209, + 210, + 211 + ] + } +] \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/020000/training_state/optimizer_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/020000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..3f0d7e484b1ca277eef28096ff36d12910b57879 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/020000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bdb7d2c2983f37ff3a2d12902fcc6a9ef8ce531d949ec203cd2a7ffc83709fac +size 2134472696 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/020000/training_state/rng_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/020000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..6e37baed9038eaa92b3564235f838b449396fced --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/020000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:30b4bc2eb89c1acd2146d668da03bcea46ff027bf7ddcc8115bc613179bbf9ab +size 15708 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/020000/training_state/scheduler_state.json b/training_runs/advanced-diffusion-v3/checkpoints/020000/training_state/scheduler_state.json new file mode 100644 index 0000000000000000000000000000000000000000..c8c235583f974dd100b1b646e34ace76dfc896ed --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/020000/training_state/scheduler_state.json @@ -0,0 +1,15 @@ +{ + "base_lrs": [ + 0.0001 + ], + "last_epoch": 20000, + "_step_count": 20001, + "_is_initial": false, + "_get_lr_called_within_step": false, + "_last_lr": [ + 9.766112029998846e-05 + ], + "lr_lambdas": [ + null + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/020000/training_state/training_step.json b/training_runs/advanced-diffusion-v3/checkpoints/020000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..dc9bb47026c5d5237ca6fc5dbff6020dd122ea05 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/020000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 20000 +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/config.json b/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..51a27627806ad961fce2c5204c9d1fa286e4c5ff --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/config.json @@ -0,0 +1,92 @@ +{ + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/model.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e201aee37e5d7d45d370a5457695f131b4c2b393 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7c2714f8fb74baa779a3f0b0d1db254110f3ea0d8f322d0385a79bf6f656f992 +size 1067235520 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/policy_postprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..7f3d1455ad86832c79e49187567da5d47aa068e7 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/policy_postprocessor.json @@ -0,0 +1,32 @@ +{ + "name": "policy_postprocessor", + "steps": [ + { + "registry_name": "unnormalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors" + }, + { + "registry_name": "device_processor", + "config": { + "device": "cpu", + "float_dtype": null + } + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/policy_preprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..c7ddebe581f2032f63d45ea5161a28bc1de8194a --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/policy_preprocessor.json @@ -0,0 +1,64 @@ +{ + "name": "policy_preprocessor", + "steps": [ + { + "registry_name": "rename_observations_processor", + "config": { + "rename_map": {} + } + }, + { + "registry_name": "to_batch_processor", + "config": {} + }, + { + "registry_name": "device_processor", + "config": { + "device": "cuda", + "float_dtype": null + } + }, + { + "registry_name": "normalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/train_config.json b/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..6dbcbb45cc6ab3592379df4548bee3518cae49c4 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/040000/pretrained_model/train_config.json @@ -0,0 +1,226 @@ +{ + "dataset": { + "repo_id": "AI-Final/AI-aiCapstoneData-lerobot-advanced-v2-2", + "root": null, + "episodes": null, + "image_transforms": { + "enable": false, + "max_num_transforms": 3, + "random_order": false, + "tfs": { + "brightness": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "brightness": [ + 0.8, + 1.2 + ] + } + }, + "contrast": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "contrast": [ + 0.8, + 1.2 + ] + } + }, + "saturation": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "saturation": [ + 0.5, + 1.5 + ] + } + }, + "hue": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "hue": [ + -0.05, + 0.05 + ] + } + }, + "sharpness": { + "weight": 1.0, + "type": "SharpnessJitter", + "kwargs": { + "sharpness": [ + 0.5, + 1.5 + ] + } + }, + "affine": { + "weight": 1.0, + "type": "RandomAffine", + "kwargs": { + "degrees": [ + -5.0, + 5.0 + ], + "translate": [ + 0.05, + 0.05 + ] + } + } + } + }, + "revision": "main", + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 + }, + "output_dir": "outputs/train/act_dining_cleanup_v2_2_20260609_033312", + "job_name": "act_dining_cleanup_v2_2_20260609_033312", + "resume": false, + "seed": 1000, + "num_workers": 4, + "batch_size": 16, + "steps": 200000, + "eval_freq": 0, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 20000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adam", + "lr": 0.0001, + "weight_decay": 1e-06, + "grad_clip_norm": 10.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": { + "type": "diffuser", + "num_warmup_steps": 500, + "name": "cosine" + }, + "eval": { + "n_episodes": 50, + "batch_size": 50, + "use_async_envs": false + }, + "wandb": { + "enable": false, + "disable_artifact": false, + "project": "lerobot", + "entity": null, + "notes": null, + "run_id": null, + "mode": null + }, + "peft": null, + "use_rabc": false, + "rabc_progress_path": null, + "rabc_kappa": 0.01, + "rabc_epsilon": 1e-06, + "rabc_head_mode": "sparse", + "rename_map": {}, + "checkpoint_path": null +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/040000/training_state/optimizer_param_groups.json b/training_runs/advanced-diffusion-v3/checkpoints/040000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..741ff0d71da2f4b05680b9b701f2f921ff461f91 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/040000/training_state/optimizer_param_groups.json @@ -0,0 +1,233 @@ +[ + { + "lr": 9.063516700762651e-05, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 1e-06, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": false, + "initial_lr": 0.0001, + "params": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19, + 20, + 21, + 22, + 23, + 24, + 25, + 26, + 27, + 28, + 29, + 30, + 31, + 32, + 33, + 34, + 35, + 36, + 37, + 38, + 39, + 40, + 41, + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61, + 62, + 63, + 64, + 65, + 66, + 67, + 68, + 69, + 70, + 71, + 72, + 73, + 74, + 75, + 76, + 77, + 78, + 79, + 80, + 81, + 82, + 83, + 84, + 85, + 86, + 87, + 88, + 89, + 90, + 91, + 92, + 93, + 94, + 95, + 96, + 97, + 98, + 99, + 100, + 101, + 102, + 103, + 104, + 105, + 106, + 107, + 108, + 109, + 110, + 111, + 112, + 113, + 114, + 115, + 116, + 117, + 118, + 119, + 120, + 121, + 122, + 123, + 124, + 125, + 126, + 127, + 128, + 129, + 130, + 131, + 132, + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152, + 153, + 154, + 155, + 156, + 157, + 158, + 159, + 160, + 161, + 162, + 163, + 164, + 165, + 166, + 167, + 168, + 169, + 170, + 171, + 172, + 173, + 174, + 175, + 176, + 177, + 178, + 179, + 180, + 181, + 182, + 183, + 184, + 185, + 186, + 187, + 188, + 189, + 190, + 191, + 192, + 193, + 194, + 195, + 196, + 197, + 198, + 199, + 200, + 201, + 202, + 203, + 204, + 205, + 206, + 207, + 208, + 209, + 210, + 211 + ] + } +] \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/040000/training_state/optimizer_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/040000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..9e042c349ebf3e134187e4b17cbe64fb016fd825 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/040000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4359855b7a4e66e97cb8fbd3822cc3f4e673f4705ff35b8fe648c9a05a406378 +size 2134472696 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/040000/training_state/rng_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/040000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b19b394c1216244abe6061f04a3e41a28c9dfdd3 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/040000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7bad751b7bced848c3ca3fe3195d6eac016a574e45f2ce0bff567cf2c0a49c47 +size 15708 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/040000/training_state/scheduler_state.json b/training_runs/advanced-diffusion-v3/checkpoints/040000/training_state/scheduler_state.json new file mode 100644 index 0000000000000000000000000000000000000000..0209f8b21c604205d3c5f41a0c50528f07a71755 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/040000/training_state/scheduler_state.json @@ -0,0 +1,15 @@ +{ + "base_lrs": [ + 0.0001 + ], + "last_epoch": 40000, + "_step_count": 40001, + "_is_initial": false, + "_get_lr_called_within_step": false, + "_last_lr": [ + 9.063516700762651e-05 + ], + "lr_lambdas": [ + null + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/040000/training_state/training_step.json b/training_runs/advanced-diffusion-v3/checkpoints/040000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..fe40d8ff4312c7e5a8fba9bcc932a43a1384ba77 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/040000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 40000 +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/config.json b/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..51a27627806ad961fce2c5204c9d1fa286e4c5ff --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/config.json @@ -0,0 +1,92 @@ +{ + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/model.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..39909bc76a7d6e60ac4808ee57af5296dfd2f09e --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8b902fe2492341a138b2da57e93cd3890763ddeca25f50f15fbfbb7bb287034f +size 1067235520 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/policy_postprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..7f3d1455ad86832c79e49187567da5d47aa068e7 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/policy_postprocessor.json @@ -0,0 +1,32 @@ +{ + "name": "policy_postprocessor", + "steps": [ + { + "registry_name": "unnormalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors" + }, + { + "registry_name": "device_processor", + "config": { + "device": "cpu", + "float_dtype": null + } + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/policy_preprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..c7ddebe581f2032f63d45ea5161a28bc1de8194a --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/policy_preprocessor.json @@ -0,0 +1,64 @@ +{ + "name": "policy_preprocessor", + "steps": [ + { + "registry_name": "rename_observations_processor", + "config": { + "rename_map": {} + } + }, + { + "registry_name": "to_batch_processor", + "config": {} + }, + { + "registry_name": "device_processor", + "config": { + "device": "cuda", + "float_dtype": null + } + }, + { + "registry_name": "normalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/train_config.json b/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..6dbcbb45cc6ab3592379df4548bee3518cae49c4 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/060000/pretrained_model/train_config.json @@ -0,0 +1,226 @@ +{ + "dataset": { + "repo_id": "AI-Final/AI-aiCapstoneData-lerobot-advanced-v2-2", + "root": null, + "episodes": null, + "image_transforms": { + "enable": false, + "max_num_transforms": 3, + "random_order": false, + "tfs": { + "brightness": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "brightness": [ + 0.8, + 1.2 + ] + } + }, + "contrast": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "contrast": [ + 0.8, + 1.2 + ] + } + }, + "saturation": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "saturation": [ + 0.5, + 1.5 + ] + } + }, + "hue": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "hue": [ + -0.05, + 0.05 + ] + } + }, + "sharpness": { + "weight": 1.0, + "type": "SharpnessJitter", + "kwargs": { + "sharpness": [ + 0.5, + 1.5 + ] + } + }, + "affine": { + "weight": 1.0, + "type": "RandomAffine", + "kwargs": { + "degrees": [ + -5.0, + 5.0 + ], + "translate": [ + 0.05, + 0.05 + ] + } + } + } + }, + "revision": "main", + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 + }, + "output_dir": "outputs/train/act_dining_cleanup_v2_2_20260609_033312", + "job_name": "act_dining_cleanup_v2_2_20260609_033312", + "resume": false, + "seed": 1000, + "num_workers": 4, + "batch_size": 16, + "steps": 200000, + "eval_freq": 0, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 20000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adam", + "lr": 0.0001, + "weight_decay": 1e-06, + "grad_clip_norm": 10.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": { + "type": "diffuser", + "num_warmup_steps": 500, + "name": "cosine" + }, + "eval": { + "n_episodes": 50, + "batch_size": 50, + "use_async_envs": false + }, + "wandb": { + "enable": false, + "disable_artifact": false, + "project": "lerobot", + "entity": null, + "notes": null, + "run_id": null, + "mode": null + }, + "peft": null, + "use_rabc": false, + "rabc_progress_path": null, + "rabc_kappa": 0.01, + "rabc_epsilon": 1e-06, + "rabc_head_mode": "sparse", + "rename_map": {}, + "checkpoint_path": null +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/060000/training_state/optimizer_param_groups.json b/training_runs/advanced-diffusion-v3/checkpoints/060000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..87110f9b83a553bef300349a7f7025e6d1b83492 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/060000/training_state/optimizer_param_groups.json @@ -0,0 +1,233 @@ +[ + { + "lr": 7.961176263324901e-05, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 1e-06, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": false, + "initial_lr": 0.0001, + "params": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19, + 20, + 21, + 22, + 23, + 24, + 25, + 26, + 27, + 28, + 29, + 30, + 31, + 32, + 33, + 34, + 35, + 36, + 37, + 38, + 39, + 40, + 41, + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61, + 62, + 63, + 64, + 65, + 66, + 67, + 68, + 69, + 70, + 71, + 72, + 73, + 74, + 75, + 76, + 77, + 78, + 79, + 80, + 81, + 82, + 83, + 84, + 85, + 86, + 87, + 88, + 89, + 90, + 91, + 92, + 93, + 94, + 95, + 96, + 97, + 98, + 99, + 100, + 101, + 102, + 103, + 104, + 105, + 106, + 107, + 108, + 109, + 110, + 111, + 112, + 113, + 114, + 115, + 116, + 117, + 118, + 119, + 120, + 121, + 122, + 123, + 124, + 125, + 126, + 127, + 128, + 129, + 130, + 131, + 132, + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152, + 153, + 154, + 155, + 156, + 157, + 158, + 159, + 160, + 161, + 162, + 163, + 164, + 165, + 166, + 167, + 168, + 169, + 170, + 171, + 172, + 173, + 174, + 175, + 176, + 177, + 178, + 179, + 180, + 181, + 182, + 183, + 184, + 185, + 186, + 187, + 188, + 189, + 190, + 191, + 192, + 193, + 194, + 195, + 196, + 197, + 198, + 199, + 200, + 201, + 202, + 203, + 204, + 205, + 206, + 207, + 208, + 209, + 210, + 211 + ] + } +] \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/060000/training_state/optimizer_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/060000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..8cf4468f6fa9eba4e4ebbea4c87a09ccda4f2a9c --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/060000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f577c82a44368f4bc601f80d6322e1898f301493d67800739dae27cb4b494cef +size 2134472696 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/060000/training_state/rng_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/060000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..0534be8bf2eabc089910900953d6b79d83ce6d67 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/060000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:916a0ec657a5a06edc9c7101136ea8f89c986d5e830428767ff56720f6c98399 +size 15708 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/060000/training_state/scheduler_state.json b/training_runs/advanced-diffusion-v3/checkpoints/060000/training_state/scheduler_state.json new file mode 100644 index 0000000000000000000000000000000000000000..90a7b933eaf672331cef07dfd60c101926b5e7c1 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/060000/training_state/scheduler_state.json @@ -0,0 +1,15 @@ +{ + "base_lrs": [ + 0.0001 + ], + "last_epoch": 60000, + "_step_count": 60001, + "_is_initial": false, + "_get_lr_called_within_step": false, + "_last_lr": [ + 7.961176263324901e-05 + ], + "lr_lambdas": [ + null + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/060000/training_state/training_step.json b/training_runs/advanced-diffusion-v3/checkpoints/060000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..d360b6037f1c8c2c85d38e951160eb02ace507dd --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/060000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 60000 +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/config.json b/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..51a27627806ad961fce2c5204c9d1fa286e4c5ff --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/config.json @@ -0,0 +1,92 @@ +{ + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/model.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..f1ed89f8fb19c428083739b6874e32f4a00ab770 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c9c4b8693cabaabe0e2346cd48026bb7f06782efa085a2e55495466f1cd6736c +size 1067235520 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/policy_postprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..7f3d1455ad86832c79e49187567da5d47aa068e7 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/policy_postprocessor.json @@ -0,0 +1,32 @@ +{ + "name": "policy_postprocessor", + "steps": [ + { + "registry_name": "unnormalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors" + }, + { + "registry_name": "device_processor", + "config": { + "device": "cpu", + "float_dtype": null + } + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/policy_preprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..c7ddebe581f2032f63d45ea5161a28bc1de8194a --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/policy_preprocessor.json @@ -0,0 +1,64 @@ +{ + "name": "policy_preprocessor", + "steps": [ + { + "registry_name": "rename_observations_processor", + "config": { + "rename_map": {} + } + }, + { + "registry_name": "to_batch_processor", + "config": {} + }, + { + "registry_name": "device_processor", + "config": { + "device": "cuda", + "float_dtype": null + } + }, + { + "registry_name": "normalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/train_config.json b/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..6dbcbb45cc6ab3592379df4548bee3518cae49c4 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/080000/pretrained_model/train_config.json @@ -0,0 +1,226 @@ +{ + "dataset": { + "repo_id": "AI-Final/AI-aiCapstoneData-lerobot-advanced-v2-2", + "root": null, + "episodes": null, + "image_transforms": { + "enable": false, + "max_num_transforms": 3, + "random_order": false, + "tfs": { + "brightness": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "brightness": [ + 0.8, + 1.2 + ] + } + }, + "contrast": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "contrast": [ + 0.8, + 1.2 + ] + } + }, + "saturation": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "saturation": [ + 0.5, + 1.5 + ] + } + }, + "hue": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "hue": [ + -0.05, + 0.05 + ] + } + }, + "sharpness": { + "weight": 1.0, + "type": "SharpnessJitter", + "kwargs": { + "sharpness": [ + 0.5, + 1.5 + ] + } + }, + "affine": { + "weight": 1.0, + "type": "RandomAffine", + "kwargs": { + "degrees": [ + -5.0, + 5.0 + ], + "translate": [ + 0.05, + 0.05 + ] + } + } + } + }, + "revision": "main", + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 + }, + "output_dir": "outputs/train/act_dining_cleanup_v2_2_20260609_033312", + "job_name": "act_dining_cleanup_v2_2_20260609_033312", + "resume": false, + "seed": 1000, + "num_workers": 4, + "batch_size": 16, + "steps": 200000, + "eval_freq": 0, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 20000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adam", + "lr": 0.0001, + "weight_decay": 1e-06, + "grad_clip_norm": 10.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": { + "type": "diffuser", + "num_warmup_steps": 500, + "name": "cosine" + }, + "eval": { + "n_episodes": 50, + "batch_size": 50, + "use_async_envs": false + }, + "wandb": { + "enable": false, + "disable_artifact": false, + "project": "lerobot", + "entity": null, + "notes": null, + "run_id": null, + "mode": null + }, + "peft": null, + "use_rabc": false, + "rabc_progress_path": null, + "rabc_kappa": 0.01, + "rabc_epsilon": 1e-06, + "rabc_head_mode": "sparse", + "rename_map": {}, + "checkpoint_path": null +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/080000/training_state/optimizer_param_groups.json b/training_runs/advanced-diffusion-v3/checkpoints/080000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..cabdfe39a8af11c485de021f8ebff4c6750d9d1f --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/080000/training_state/optimizer_param_groups.json @@ -0,0 +1,233 @@ +[ + { + "lr": 6.567532550188907e-05, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 1e-06, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": false, + "initial_lr": 0.0001, + "params": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19, + 20, + 21, + 22, + 23, + 24, + 25, + 26, + 27, + 28, + 29, + 30, + 31, + 32, + 33, + 34, + 35, + 36, + 37, + 38, + 39, + 40, + 41, + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61, + 62, + 63, + 64, + 65, + 66, + 67, + 68, + 69, + 70, + 71, + 72, + 73, + 74, + 75, + 76, + 77, + 78, + 79, + 80, + 81, + 82, + 83, + 84, + 85, + 86, + 87, + 88, + 89, + 90, + 91, + 92, + 93, + 94, + 95, + 96, + 97, + 98, + 99, + 100, + 101, + 102, + 103, + 104, + 105, + 106, + 107, + 108, + 109, + 110, + 111, + 112, + 113, + 114, + 115, + 116, + 117, + 118, + 119, + 120, + 121, + 122, + 123, + 124, + 125, + 126, + 127, + 128, + 129, + 130, + 131, + 132, + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152, + 153, + 154, + 155, + 156, + 157, + 158, + 159, + 160, + 161, + 162, + 163, + 164, + 165, + 166, + 167, + 168, + 169, + 170, + 171, + 172, + 173, + 174, + 175, + 176, + 177, + 178, + 179, + 180, + 181, + 182, + 183, + 184, + 185, + 186, + 187, + 188, + 189, + 190, + 191, + 192, + 193, + 194, + 195, + 196, + 197, + 198, + 199, + 200, + 201, + 202, + 203, + 204, + 205, + 206, + 207, + 208, + 209, + 210, + 211 + ] + } +] \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/080000/training_state/optimizer_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/080000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..2c31290016ad20b06179f81a78bb380c8ceaa63d --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/080000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d1d7ba6269dc157908c48dcd8189860d2a56bb481ccd26135ab50be49054fae9 +size 2134472696 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/080000/training_state/rng_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/080000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..53ac3e912d9a2825dc863443dbfe857e09715f72 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/080000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ae7f82adb79c890e0a941471aa22f03f0a029151dae62d5266360131ffb49ba5 +size 15708 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/080000/training_state/scheduler_state.json b/training_runs/advanced-diffusion-v3/checkpoints/080000/training_state/scheduler_state.json new file mode 100644 index 0000000000000000000000000000000000000000..c44746ed1d22f5bb663a3bc294e0e59a2711b22a --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/080000/training_state/scheduler_state.json @@ -0,0 +1,15 @@ +{ + "base_lrs": [ + 0.0001 + ], + "last_epoch": 80000, + "_step_count": 80001, + "_is_initial": false, + "_get_lr_called_within_step": false, + "_last_lr": [ + 6.567532550188907e-05 + ], + "lr_lambdas": [ + null + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/080000/training_state/training_step.json b/training_runs/advanced-diffusion-v3/checkpoints/080000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..384eace4ecc2a6cba352aa7cf27f04405b7319c3 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/080000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 80000 +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/config.json b/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..51a27627806ad961fce2c5204c9d1fa286e4c5ff --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/config.json @@ -0,0 +1,92 @@ +{ + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/model.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..489d86335c111efb3675974d087f44e31cea8ee6 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a4dd721ad4cb9f6d3822c0feeb83a1cfe1094bce59f60fa58979dc9800d67138 +size 1067235520 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/policy_postprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..7f3d1455ad86832c79e49187567da5d47aa068e7 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/policy_postprocessor.json @@ -0,0 +1,32 @@ +{ + "name": "policy_postprocessor", + "steps": [ + { + "registry_name": "unnormalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors" + }, + { + "registry_name": "device_processor", + "config": { + "device": "cpu", + "float_dtype": null + } + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/policy_preprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..c7ddebe581f2032f63d45ea5161a28bc1de8194a --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/policy_preprocessor.json @@ -0,0 +1,64 @@ +{ + "name": "policy_preprocessor", + "steps": [ + { + "registry_name": "rename_observations_processor", + "config": { + "rename_map": {} + } + }, + { + "registry_name": "to_batch_processor", + "config": {} + }, + { + "registry_name": "device_processor", + "config": { + "device": "cuda", + "float_dtype": null + } + }, + { + "registry_name": "normalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/train_config.json b/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..6dbcbb45cc6ab3592379df4548bee3518cae49c4 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/100000/pretrained_model/train_config.json @@ -0,0 +1,226 @@ +{ + "dataset": { + "repo_id": "AI-Final/AI-aiCapstoneData-lerobot-advanced-v2-2", + "root": null, + "episodes": null, + "image_transforms": { + "enable": false, + "max_num_transforms": 3, + "random_order": false, + "tfs": { + "brightness": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "brightness": [ + 0.8, + 1.2 + ] + } + }, + "contrast": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "contrast": [ + 0.8, + 1.2 + ] + } + }, + "saturation": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "saturation": [ + 0.5, + 1.5 + ] + } + }, + "hue": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "hue": [ + -0.05, + 0.05 + ] + } + }, + "sharpness": { + "weight": 1.0, + "type": "SharpnessJitter", + "kwargs": { + "sharpness": [ + 0.5, + 1.5 + ] + } + }, + "affine": { + "weight": 1.0, + "type": "RandomAffine", + "kwargs": { + "degrees": [ + -5.0, + 5.0 + ], + "translate": [ + 0.05, + 0.05 + ] + } + } + } + }, + "revision": "main", + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 + }, + "output_dir": "outputs/train/act_dining_cleanup_v2_2_20260609_033312", + "job_name": "act_dining_cleanup_v2_2_20260609_033312", + "resume": false, + "seed": 1000, + "num_workers": 4, + "batch_size": 16, + "steps": 200000, + "eval_freq": 0, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 20000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adam", + "lr": 0.0001, + "weight_decay": 1e-06, + "grad_clip_norm": 10.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": { + "type": "diffuser", + "num_warmup_steps": 500, + "name": "cosine" + }, + "eval": { + "n_episodes": 50, + "batch_size": 50, + "use_async_envs": false + }, + "wandb": { + "enable": false, + "disable_artifact": false, + "project": "lerobot", + "entity": null, + "notes": null, + "run_id": null, + "mode": null + }, + "peft": null, + "use_rabc": false, + "rabc_progress_path": null, + "rabc_kappa": 0.01, + "rabc_epsilon": 1e-06, + "rabc_head_mode": "sparse", + "rename_map": {}, + "checkpoint_path": null +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/100000/training_state/optimizer_param_groups.json b/training_runs/advanced-diffusion-v3/checkpoints/100000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..8d9846da446d9efe948aa1ad31a630a03ffac311 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/100000/training_state/optimizer_param_groups.json @@ -0,0 +1,233 @@ +[ + { + "lr": 5.0196841136498776e-05, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 1e-06, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": false, + "initial_lr": 0.0001, + "params": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19, + 20, + 21, + 22, + 23, + 24, + 25, + 26, + 27, + 28, + 29, + 30, + 31, + 32, + 33, + 34, + 35, + 36, + 37, + 38, + 39, + 40, + 41, + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61, + 62, + 63, + 64, + 65, + 66, + 67, + 68, + 69, + 70, + 71, + 72, + 73, + 74, + 75, + 76, + 77, + 78, + 79, + 80, + 81, + 82, + 83, + 84, + 85, + 86, + 87, + 88, + 89, + 90, + 91, + 92, + 93, + 94, + 95, + 96, + 97, + 98, + 99, + 100, + 101, + 102, + 103, + 104, + 105, + 106, + 107, + 108, + 109, + 110, + 111, + 112, + 113, + 114, + 115, + 116, + 117, + 118, + 119, + 120, + 121, + 122, + 123, + 124, + 125, + 126, + 127, + 128, + 129, + 130, + 131, + 132, + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152, + 153, + 154, + 155, + 156, + 157, + 158, + 159, + 160, + 161, + 162, + 163, + 164, + 165, + 166, + 167, + 168, + 169, + 170, + 171, + 172, + 173, + 174, + 175, + 176, + 177, + 178, + 179, + 180, + 181, + 182, + 183, + 184, + 185, + 186, + 187, + 188, + 189, + 190, + 191, + 192, + 193, + 194, + 195, + 196, + 197, + 198, + 199, + 200, + 201, + 202, + 203, + 204, + 205, + 206, + 207, + 208, + 209, + 210, + 211 + ] + } +] \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/100000/training_state/optimizer_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/100000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..82945afece957790feea39176402b77a71e1261b --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/100000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9b44341603a1bf75c9c7d2d6e35e7c0a17cdfe27f9d7ccc399e63373d441f5a6 +size 2134472696 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/100000/training_state/rng_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/100000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..0210ba871c9050c98e3c3191e991b469a9358218 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/100000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4d067aecc57744c9280b429286c19942c76cd7a0e8205be089b8fc6776459a45 +size 15708 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/100000/training_state/scheduler_state.json b/training_runs/advanced-diffusion-v3/checkpoints/100000/training_state/scheduler_state.json new file mode 100644 index 0000000000000000000000000000000000000000..f0b0d9e99258dc69279ec9748dbf7f82a5cd25cc --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/100000/training_state/scheduler_state.json @@ -0,0 +1,15 @@ +{ + "base_lrs": [ + 0.0001 + ], + "last_epoch": 100000, + "_step_count": 100001, + "_is_initial": false, + "_get_lr_called_within_step": false, + "_last_lr": [ + 5.0196841136498776e-05 + ], + "lr_lambdas": [ + null + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/100000/training_state/training_step.json b/training_runs/advanced-diffusion-v3/checkpoints/100000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..8eb73c13d28bca88058c08796abbf931c3f9b012 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/100000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 100000 +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/config.json b/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..51a27627806ad961fce2c5204c9d1fa286e4c5ff --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/config.json @@ -0,0 +1,92 @@ +{ + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/model.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..0256b888842410592305eb71c12ed61f11f75a7e --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b342d052305a8e394cabe0c39fca3dd386905d06716a821b20910c329f305ce3 +size 1067235520 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/policy_postprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..7f3d1455ad86832c79e49187567da5d47aa068e7 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/policy_postprocessor.json @@ -0,0 +1,32 @@ +{ + "name": "policy_postprocessor", + "steps": [ + { + "registry_name": "unnormalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors" + }, + { + "registry_name": "device_processor", + "config": { + "device": "cpu", + "float_dtype": null + } + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/policy_preprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..c7ddebe581f2032f63d45ea5161a28bc1de8194a --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/policy_preprocessor.json @@ -0,0 +1,64 @@ +{ + "name": "policy_preprocessor", + "steps": [ + { + "registry_name": "rename_observations_processor", + "config": { + "rename_map": {} + } + }, + { + "registry_name": "to_batch_processor", + "config": {} + }, + { + "registry_name": "device_processor", + "config": { + "device": "cuda", + "float_dtype": null + } + }, + { + "registry_name": "normalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/train_config.json b/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..6dbcbb45cc6ab3592379df4548bee3518cae49c4 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/120000/pretrained_model/train_config.json @@ -0,0 +1,226 @@ +{ + "dataset": { + "repo_id": "AI-Final/AI-aiCapstoneData-lerobot-advanced-v2-2", + "root": null, + "episodes": null, + "image_transforms": { + "enable": false, + "max_num_transforms": 3, + "random_order": false, + "tfs": { + "brightness": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "brightness": [ + 0.8, + 1.2 + ] + } + }, + "contrast": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "contrast": [ + 0.8, + 1.2 + ] + } + }, + "saturation": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "saturation": [ + 0.5, + 1.5 + ] + } + }, + "hue": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "hue": [ + -0.05, + 0.05 + ] + } + }, + "sharpness": { + "weight": 1.0, + "type": "SharpnessJitter", + "kwargs": { + "sharpness": [ + 0.5, + 1.5 + ] + } + }, + "affine": { + "weight": 1.0, + "type": "RandomAffine", + "kwargs": { + "degrees": [ + -5.0, + 5.0 + ], + "translate": [ + 0.05, + 0.05 + ] + } + } + } + }, + "revision": "main", + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 + }, + "output_dir": "outputs/train/act_dining_cleanup_v2_2_20260609_033312", + "job_name": "act_dining_cleanup_v2_2_20260609_033312", + "resume": false, + "seed": 1000, + "num_workers": 4, + "batch_size": 16, + "steps": 200000, + "eval_freq": 0, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 20000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adam", + "lr": 0.0001, + "weight_decay": 1e-06, + "grad_clip_norm": 10.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": { + "type": "diffuser", + "num_warmup_steps": 500, + "name": "cosine" + }, + "eval": { + "n_episodes": 50, + "batch_size": 50, + "use_async_envs": false + }, + "wandb": { + "enable": false, + "disable_artifact": false, + "project": "lerobot", + "entity": null, + "notes": null, + "run_id": null, + "mode": null + }, + "peft": null, + "use_rabc": false, + "rabc_progress_path": null, + "rabc_kappa": 0.01, + "rabc_epsilon": 1e-06, + "rabc_head_mode": "sparse", + "rename_map": {}, + "checkpoint_path": null +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/120000/training_state/optimizer_param_groups.json b/training_runs/advanced-diffusion-v3/checkpoints/120000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..8069d753aa6b96f1c61c4492eef24924489b3fd1 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/120000/training_state/optimizer_param_groups.json @@ -0,0 +1,233 @@ +[ + { + "lr": 3.4698992686449127e-05, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 1e-06, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": false, + "initial_lr": 0.0001, + "params": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19, + 20, + 21, + 22, + 23, + 24, + 25, + 26, + 27, + 28, + 29, + 30, + 31, + 32, + 33, + 34, + 35, + 36, + 37, + 38, + 39, + 40, + 41, + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61, + 62, + 63, + 64, + 65, + 66, + 67, + 68, + 69, + 70, + 71, + 72, + 73, + 74, + 75, + 76, + 77, + 78, + 79, + 80, + 81, + 82, + 83, + 84, + 85, + 86, + 87, + 88, + 89, + 90, + 91, + 92, + 93, + 94, + 95, + 96, + 97, + 98, + 99, + 100, + 101, + 102, + 103, + 104, + 105, + 106, + 107, + 108, + 109, + 110, + 111, + 112, + 113, + 114, + 115, + 116, + 117, + 118, + 119, + 120, + 121, + 122, + 123, + 124, + 125, + 126, + 127, + 128, + 129, + 130, + 131, + 132, + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152, + 153, + 154, + 155, + 156, + 157, + 158, + 159, + 160, + 161, + 162, + 163, + 164, + 165, + 166, + 167, + 168, + 169, + 170, + 171, + 172, + 173, + 174, + 175, + 176, + 177, + 178, + 179, + 180, + 181, + 182, + 183, + 184, + 185, + 186, + 187, + 188, + 189, + 190, + 191, + 192, + 193, + 194, + 195, + 196, + 197, + 198, + 199, + 200, + 201, + 202, + 203, + 204, + 205, + 206, + 207, + 208, + 209, + 210, + 211 + ] + } +] \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/120000/training_state/optimizer_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/120000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..7eb9db5068bd9617c70b7fdf2ae933239434f6ae --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/120000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a36a4fb944ad83910e73e0316791936dfb75ba124bf71a7ae0a22bb08fe60fe6 +size 2134472696 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/120000/training_state/rng_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/120000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d00b57968d0553fab84ce9e35db792d853a01771 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/120000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7fddcda91faa6f1619b64ace3b00b8d749f64228381ebff8933272e2781a18c3 +size 15708 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/120000/training_state/scheduler_state.json b/training_runs/advanced-diffusion-v3/checkpoints/120000/training_state/scheduler_state.json new file mode 100644 index 0000000000000000000000000000000000000000..d6455435f8b29ad5daaacb689ba88a6e947b6861 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/120000/training_state/scheduler_state.json @@ -0,0 +1,15 @@ +{ + "base_lrs": [ + 0.0001 + ], + "last_epoch": 120000, + "_step_count": 120001, + "_is_initial": false, + "_get_lr_called_within_step": false, + "_last_lr": [ + 3.4698992686449127e-05 + ], + "lr_lambdas": [ + null + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/120000/training_state/training_step.json b/training_runs/advanced-diffusion-v3/checkpoints/120000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..6bbb8719463d819c865c104597046a0165eea278 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/120000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 120000 +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/config.json b/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..51a27627806ad961fce2c5204c9d1fa286e4c5ff --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/config.json @@ -0,0 +1,92 @@ +{ + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/model.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..038ff17ac0f5a44cf68a4abf1a00bd172d4956c4 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c007bcd40434e5f8f2224e735d3b5498ca9c6ca2c687d248ceefdee860ae6157 +size 1067235520 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/policy_postprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..7f3d1455ad86832c79e49187567da5d47aa068e7 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/policy_postprocessor.json @@ -0,0 +1,32 @@ +{ + "name": "policy_postprocessor", + "steps": [ + { + "registry_name": "unnormalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors" + }, + { + "registry_name": "device_processor", + "config": { + "device": "cpu", + "float_dtype": null + } + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/policy_preprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..c7ddebe581f2032f63d45ea5161a28bc1de8194a --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/policy_preprocessor.json @@ -0,0 +1,64 @@ +{ + "name": "policy_preprocessor", + "steps": [ + { + "registry_name": "rename_observations_processor", + "config": { + "rename_map": {} + } + }, + { + "registry_name": "to_batch_processor", + "config": {} + }, + { + "registry_name": "device_processor", + "config": { + "device": "cuda", + "float_dtype": null + } + }, + { + "registry_name": "normalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/train_config.json b/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..6dbcbb45cc6ab3592379df4548bee3518cae49c4 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/140000/pretrained_model/train_config.json @@ -0,0 +1,226 @@ +{ + "dataset": { + "repo_id": "AI-Final/AI-aiCapstoneData-lerobot-advanced-v2-2", + "root": null, + "episodes": null, + "image_transforms": { + "enable": false, + "max_num_transforms": 3, + "random_order": false, + "tfs": { + "brightness": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "brightness": [ + 0.8, + 1.2 + ] + } + }, + "contrast": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "contrast": [ + 0.8, + 1.2 + ] + } + }, + "saturation": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "saturation": [ + 0.5, + 1.5 + ] + } + }, + "hue": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "hue": [ + -0.05, + 0.05 + ] + } + }, + "sharpness": { + "weight": 1.0, + "type": "SharpnessJitter", + "kwargs": { + "sharpness": [ + 0.5, + 1.5 + ] + } + }, + "affine": { + "weight": 1.0, + "type": "RandomAffine", + "kwargs": { + "degrees": [ + -5.0, + 5.0 + ], + "translate": [ + 0.05, + 0.05 + ] + } + } + } + }, + "revision": "main", + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 + }, + "output_dir": "outputs/train/act_dining_cleanup_v2_2_20260609_033312", + "job_name": "act_dining_cleanup_v2_2_20260609_033312", + "resume": false, + "seed": 1000, + "num_workers": 4, + "batch_size": 16, + "steps": 200000, + "eval_freq": 0, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 20000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adam", + "lr": 0.0001, + "weight_decay": 1e-06, + "grad_clip_norm": 10.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": { + "type": "diffuser", + "num_warmup_steps": 500, + "name": "cosine" + }, + "eval": { + "n_episodes": 50, + "batch_size": 50, + "use_async_envs": false + }, + "wandb": { + "enable": false, + "disable_artifact": false, + "project": "lerobot", + "entity": null, + "notes": null, + "run_id": null, + "mode": null + }, + "peft": null, + "use_rabc": false, + "rabc_progress_path": null, + "rabc_kappa": 0.01, + "rabc_epsilon": 1e-06, + "rabc_head_mode": "sparse", + "rename_map": {}, + "checkpoint_path": null +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/140000/training_state/optimizer_param_groups.json b/training_runs/advanced-diffusion-v3/checkpoints/140000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..c41f017a28b1fb31fb3102b39d0bbe1ea08e238f --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/140000/training_state/optimizer_param_groups.json @@ -0,0 +1,233 @@ +[ + { + "lr": 2.070636822698877e-05, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 1e-06, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": false, + "initial_lr": 0.0001, + "params": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19, + 20, + 21, + 22, + 23, + 24, + 25, + 26, + 27, + 28, + 29, + 30, + 31, + 32, + 33, + 34, + 35, + 36, + 37, + 38, + 39, + 40, + 41, + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61, + 62, + 63, + 64, + 65, + 66, + 67, + 68, + 69, + 70, + 71, + 72, + 73, + 74, + 75, + 76, + 77, + 78, + 79, + 80, + 81, + 82, + 83, + 84, + 85, + 86, + 87, + 88, + 89, + 90, + 91, + 92, + 93, + 94, + 95, + 96, + 97, + 98, + 99, + 100, + 101, + 102, + 103, + 104, + 105, + 106, + 107, + 108, + 109, + 110, + 111, + 112, + 113, + 114, + 115, + 116, + 117, + 118, + 119, + 120, + 121, + 122, + 123, + 124, + 125, + 126, + 127, + 128, + 129, + 130, + 131, + 132, + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152, + 153, + 154, + 155, + 156, + 157, + 158, + 159, + 160, + 161, + 162, + 163, + 164, + 165, + 166, + 167, + 168, + 169, + 170, + 171, + 172, + 173, + 174, + 175, + 176, + 177, + 178, + 179, + 180, + 181, + 182, + 183, + 184, + 185, + 186, + 187, + 188, + 189, + 190, + 191, + 192, + 193, + 194, + 195, + 196, + 197, + 198, + 199, + 200, + 201, + 202, + 203, + 204, + 205, + 206, + 207, + 208, + 209, + 210, + 211 + ] + } +] \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/140000/training_state/optimizer_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/140000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..aeab683222d6495a9c178311fb96a6fd27abee3b --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/140000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4927046cfad993d1faff90bc77b061a7baff2f65144ebb8d6a1bb136adf4fe6a +size 2134472696 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/140000/training_state/rng_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/140000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..23c236d484e7fe755e9ee5dc56c459aa4f711c04 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/140000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:84908b09351fbced0bd4a907ec60c7a8e8cffe38750b5a7210a371ae07630fab +size 15708 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/140000/training_state/scheduler_state.json b/training_runs/advanced-diffusion-v3/checkpoints/140000/training_state/scheduler_state.json new file mode 100644 index 0000000000000000000000000000000000000000..7347da9db9bc9ac0bd7b99d35be76b26a9016d27 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/140000/training_state/scheduler_state.json @@ -0,0 +1,15 @@ +{ + "base_lrs": [ + 0.0001 + ], + "last_epoch": 140000, + "_step_count": 140001, + "_is_initial": false, + "_get_lr_called_within_step": false, + "_last_lr": [ + 2.070636822698877e-05 + ], + "lr_lambdas": [ + null + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/140000/training_state/training_step.json b/training_runs/advanced-diffusion-v3/checkpoints/140000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..f3695dfae1578a72a581cd61c827ae9be8dcaa64 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/140000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 140000 +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/config.json b/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..51a27627806ad961fce2c5204c9d1fa286e4c5ff --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/config.json @@ -0,0 +1,92 @@ +{ + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/model.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..ee65053338d96648ef7492d4b4806d5a8c5c1268 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9857d6e47af0c5cad5e104d522f1585748802e219c57e30390cdf61342a81950 +size 1067235520 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/policy_postprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..7f3d1455ad86832c79e49187567da5d47aa068e7 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/policy_postprocessor.json @@ -0,0 +1,32 @@ +{ + "name": "policy_postprocessor", + "steps": [ + { + "registry_name": "unnormalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors" + }, + { + "registry_name": "device_processor", + "config": { + "device": "cpu", + "float_dtype": null + } + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/policy_preprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..c7ddebe581f2032f63d45ea5161a28bc1de8194a --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/policy_preprocessor.json @@ -0,0 +1,64 @@ +{ + "name": "policy_preprocessor", + "steps": [ + { + "registry_name": "rename_observations_processor", + "config": { + "rename_map": {} + } + }, + { + "registry_name": "to_batch_processor", + "config": {} + }, + { + "registry_name": "device_processor", + "config": { + "device": "cuda", + "float_dtype": null + } + }, + { + "registry_name": "normalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/train_config.json b/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..6dbcbb45cc6ab3592379df4548bee3518cae49c4 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/160000/pretrained_model/train_config.json @@ -0,0 +1,226 @@ +{ + "dataset": { + "repo_id": "AI-Final/AI-aiCapstoneData-lerobot-advanced-v2-2", + "root": null, + "episodes": null, + "image_transforms": { + "enable": false, + "max_num_transforms": 3, + "random_order": false, + "tfs": { + "brightness": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "brightness": [ + 0.8, + 1.2 + ] + } + }, + "contrast": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "contrast": [ + 0.8, + 1.2 + ] + } + }, + "saturation": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "saturation": [ + 0.5, + 1.5 + ] + } + }, + "hue": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "hue": [ + -0.05, + 0.05 + ] + } + }, + "sharpness": { + "weight": 1.0, + "type": "SharpnessJitter", + "kwargs": { + "sharpness": [ + 0.5, + 1.5 + ] + } + }, + "affine": { + "weight": 1.0, + "type": "RandomAffine", + "kwargs": { + "degrees": [ + -5.0, + 5.0 + ], + "translate": [ + 0.05, + 0.05 + ] + } + } + } + }, + "revision": "main", + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 + }, + "output_dir": "outputs/train/act_dining_cleanup_v2_2_20260609_033312", + "job_name": "act_dining_cleanup_v2_2_20260609_033312", + "resume": false, + "seed": 1000, + "num_workers": 4, + "batch_size": 16, + "steps": 200000, + "eval_freq": 0, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 20000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adam", + "lr": 0.0001, + "weight_decay": 1e-06, + "grad_clip_norm": 10.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": { + "type": "diffuser", + "num_warmup_steps": 500, + "name": "cosine" + }, + "eval": { + "n_episodes": 50, + "batch_size": 50, + "use_async_envs": false + }, + "wandb": { + "enable": false, + "disable_artifact": false, + "project": "lerobot", + "entity": null, + "notes": null, + "run_id": null, + "mode": null + }, + "peft": null, + "use_rabc": false, + "rabc_progress_path": null, + "rabc_kappa": 0.01, + "rabc_epsilon": 1e-06, + "rabc_head_mode": "sparse", + "rename_map": {}, + "checkpoint_path": null +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/160000/training_state/optimizer_param_groups.json b/training_runs/advanced-diffusion-v3/checkpoints/160000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..bcefd7fe1be68294f45a54c9846d7f586697a6ec --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/160000/training_state/optimizer_param_groups.json @@ -0,0 +1,233 @@ +[ + { + "lr": 9.595480663188527e-06, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 1e-06, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": false, + "initial_lr": 0.0001, + "params": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19, + 20, + 21, + 22, + 23, + 24, + 25, + 26, + 27, + 28, + 29, + 30, + 31, + 32, + 33, + 34, + 35, + 36, + 37, + 38, + 39, + 40, + 41, + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61, + 62, + 63, + 64, + 65, + 66, + 67, + 68, + 69, + 70, + 71, + 72, + 73, + 74, + 75, + 76, + 77, + 78, + 79, + 80, + 81, + 82, + 83, + 84, + 85, + 86, + 87, + 88, + 89, + 90, + 91, + 92, + 93, + 94, + 95, + 96, + 97, + 98, + 99, + 100, + 101, + 102, + 103, + 104, + 105, + 106, + 107, + 108, + 109, + 110, + 111, + 112, + 113, + 114, + 115, + 116, + 117, + 118, + 119, + 120, + 121, + 122, + 123, + 124, + 125, + 126, + 127, + 128, + 129, + 130, + 131, + 132, + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152, + 153, + 154, + 155, + 156, + 157, + 158, + 159, + 160, + 161, + 162, + 163, + 164, + 165, + 166, + 167, + 168, + 169, + 170, + 171, + 172, + 173, + 174, + 175, + 176, + 177, + 178, + 179, + 180, + 181, + 182, + 183, + 184, + 185, + 186, + 187, + 188, + 189, + 190, + 191, + 192, + 193, + 194, + 195, + 196, + 197, + 198, + 199, + 200, + 201, + 202, + 203, + 204, + 205, + 206, + 207, + 208, + 209, + 210, + 211 + ] + } +] \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/160000/training_state/optimizer_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/160000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..840e165477e4d69d32347012915cc6cf7adb0d77 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/160000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:33da73c7685709a3e19911fe289d58f68e2d30bb8662df9870580a8242b69b1b +size 2134472696 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/160000/training_state/rng_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/160000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..0cd1a96edbc068230608661714c33edf08c19255 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/160000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c00ecb2fc30f6a15c15ba1424e2ba8d7679a420f8f3c87a7af1f21ba0867f45a +size 15708 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/160000/training_state/scheduler_state.json b/training_runs/advanced-diffusion-v3/checkpoints/160000/training_state/scheduler_state.json new file mode 100644 index 0000000000000000000000000000000000000000..959c7fc6d427ec1af8694cdf354a7d6354d8c9dc --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/160000/training_state/scheduler_state.json @@ -0,0 +1,15 @@ +{ + "base_lrs": [ + 0.0001 + ], + "last_epoch": 160000, + "_step_count": 160001, + "_is_initial": false, + "_get_lr_called_within_step": false, + "_last_lr": [ + 9.595480663188527e-06 + ], + "lr_lambdas": [ + null + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/160000/training_state/training_step.json b/training_runs/advanced-diffusion-v3/checkpoints/160000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..18f112ab3b2ba641c0049e21daf4fb129368a0a0 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/160000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 160000 +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/config.json b/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..51a27627806ad961fce2c5204c9d1fa286e4c5ff --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/config.json @@ -0,0 +1,92 @@ +{ + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/model.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..65ecc721128915619da25e0373ccaf317c392e3e --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:13b8e88b9ceb238a87a6c710ac9f0871e05d6d9585277786f61d01717777bb7b +size 1067235520 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/policy_postprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..7f3d1455ad86832c79e49187567da5d47aa068e7 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/policy_postprocessor.json @@ -0,0 +1,32 @@ +{ + "name": "policy_postprocessor", + "steps": [ + { + "registry_name": "unnormalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors" + }, + { + "registry_name": "device_processor", + "config": { + "device": "cpu", + "float_dtype": null + } + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/policy_preprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..c7ddebe581f2032f63d45ea5161a28bc1de8194a --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/policy_preprocessor.json @@ -0,0 +1,64 @@ +{ + "name": "policy_preprocessor", + "steps": [ + { + "registry_name": "rename_observations_processor", + "config": { + "rename_map": {} + } + }, + { + "registry_name": "to_batch_processor", + "config": {} + }, + { + "registry_name": "device_processor", + "config": { + "device": "cuda", + "float_dtype": null + } + }, + { + "registry_name": "normalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/train_config.json b/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..6dbcbb45cc6ab3592379df4548bee3518cae49c4 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/180000/pretrained_model/train_config.json @@ -0,0 +1,226 @@ +{ + "dataset": { + "repo_id": "AI-Final/AI-aiCapstoneData-lerobot-advanced-v2-2", + "root": null, + "episodes": null, + "image_transforms": { + "enable": false, + "max_num_transforms": 3, + "random_order": false, + "tfs": { + "brightness": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "brightness": [ + 0.8, + 1.2 + ] + } + }, + "contrast": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "contrast": [ + 0.8, + 1.2 + ] + } + }, + "saturation": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "saturation": [ + 0.5, + 1.5 + ] + } + }, + "hue": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "hue": [ + -0.05, + 0.05 + ] + } + }, + "sharpness": { + "weight": 1.0, + "type": "SharpnessJitter", + "kwargs": { + "sharpness": [ + 0.5, + 1.5 + ] + } + }, + "affine": { + "weight": 1.0, + "type": "RandomAffine", + "kwargs": { + "degrees": [ + -5.0, + 5.0 + ], + "translate": [ + 0.05, + 0.05 + ] + } + } + } + }, + "revision": "main", + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 + }, + "output_dir": "outputs/train/act_dining_cleanup_v2_2_20260609_033312", + "job_name": "act_dining_cleanup_v2_2_20260609_033312", + "resume": false, + "seed": 1000, + "num_workers": 4, + "batch_size": 16, + "steps": 200000, + "eval_freq": 0, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 20000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adam", + "lr": 0.0001, + "weight_decay": 1e-06, + "grad_clip_norm": 10.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": { + "type": "diffuser", + "num_warmup_steps": 500, + "name": "cosine" + }, + "eval": { + "n_episodes": 50, + "batch_size": 50, + "use_async_envs": false + }, + "wandb": { + "enable": false, + "disable_artifact": false, + "project": "lerobot", + "entity": null, + "notes": null, + "run_id": null, + "mode": null + }, + "peft": null, + "use_rabc": false, + "rabc_progress_path": null, + "rabc_kappa": 0.01, + "rabc_epsilon": 1e-06, + "rabc_head_mode": "sparse", + "rename_map": {}, + "checkpoint_path": null +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/180000/training_state/optimizer_param_groups.json b/training_runs/advanced-diffusion-v3/checkpoints/180000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..6060808e87eb30472c6fc7f26236065822d1b137 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/180000/training_state/optimizer_param_groups.json @@ -0,0 +1,233 @@ +[ + { + "lr": 2.4593544067785466e-06, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 1e-06, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": false, + "initial_lr": 0.0001, + "params": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19, + 20, + 21, + 22, + 23, + 24, + 25, + 26, + 27, + 28, + 29, + 30, + 31, + 32, + 33, + 34, + 35, + 36, + 37, + 38, + 39, + 40, + 41, + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61, + 62, + 63, + 64, + 65, + 66, + 67, + 68, + 69, + 70, + 71, + 72, + 73, + 74, + 75, + 76, + 77, + 78, + 79, + 80, + 81, + 82, + 83, + 84, + 85, + 86, + 87, + 88, + 89, + 90, + 91, + 92, + 93, + 94, + 95, + 96, + 97, + 98, + 99, + 100, + 101, + 102, + 103, + 104, + 105, + 106, + 107, + 108, + 109, + 110, + 111, + 112, + 113, + 114, + 115, + 116, + 117, + 118, + 119, + 120, + 121, + 122, + 123, + 124, + 125, + 126, + 127, + 128, + 129, + 130, + 131, + 132, + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152, + 153, + 154, + 155, + 156, + 157, + 158, + 159, + 160, + 161, + 162, + 163, + 164, + 165, + 166, + 167, + 168, + 169, + 170, + 171, + 172, + 173, + 174, + 175, + 176, + 177, + 178, + 179, + 180, + 181, + 182, + 183, + 184, + 185, + 186, + 187, + 188, + 189, + 190, + 191, + 192, + 193, + 194, + 195, + 196, + 197, + 198, + 199, + 200, + 201, + 202, + 203, + 204, + 205, + 206, + 207, + 208, + 209, + 210, + 211 + ] + } +] \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/180000/training_state/optimizer_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/180000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..a5540e82911a64ffc8b954d4384dc80be7e3b9ef --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/180000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:76ad9554b7accf4cab2f977cee54d25ec05c7f7f94a20aac5c98f0fd966e5f55 +size 2134472696 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/180000/training_state/rng_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/180000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c6b82c7d4ca0190289ea16e043e13edb511549aa --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/180000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8cea999b0709510ee7de3f810c3154782cb1e64da7850d854e061d39d20e9cf0 +size 15708 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/180000/training_state/scheduler_state.json b/training_runs/advanced-diffusion-v3/checkpoints/180000/training_state/scheduler_state.json new file mode 100644 index 0000000000000000000000000000000000000000..7fc8da19266860db5160db88bf4588d2bbf714d1 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/180000/training_state/scheduler_state.json @@ -0,0 +1,15 @@ +{ + "base_lrs": [ + 0.0001 + ], + "last_epoch": 180000, + "_step_count": 180001, + "_is_initial": false, + "_get_lr_called_within_step": false, + "_last_lr": [ + 2.4593544067785466e-06 + ], + "lr_lambdas": [ + null + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/180000/training_state/training_step.json b/training_runs/advanced-diffusion-v3/checkpoints/180000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..3da013f3092359f6c4ebb97dfc69e8300dbb32e4 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/180000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 180000 +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/config.json b/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..51a27627806ad961fce2c5204c9d1fa286e4c5ff --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/config.json @@ -0,0 +1,92 @@ +{ + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/model.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d104fff2d42a80b5a529eb1ed335518d3ad14b11 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ef28ed8ba505fd50570d8b128b4ae9c1ba88ae3043457a64e6ff19185d4ea9e9 +size 1067235520 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/policy_postprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..7f3d1455ad86832c79e49187567da5d47aa068e7 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/policy_postprocessor.json @@ -0,0 +1,32 @@ +{ + "name": "policy_postprocessor", + "steps": [ + { + "registry_name": "unnormalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors" + }, + { + "registry_name": "device_processor", + "config": { + "device": "cpu", + "float_dtype": null + } + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/policy_preprocessor.json b/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..c7ddebe581f2032f63d45ea5161a28bc1de8194a --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/policy_preprocessor.json @@ -0,0 +1,64 @@ +{ + "name": "policy_preprocessor", + "steps": [ + { + "registry_name": "rename_observations_processor", + "config": { + "rename_map": {} + } + }, + { + "registry_name": "to_batch_processor", + "config": {} + }, + { + "registry_name": "device_processor", + "config": { + "device": "cuda", + "float_dtype": null + } + }, + { + "registry_name": "normalizer_processor", + "config": { + "eps": 1e-08, + "features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..704482aa795bd2559039661e2127bb612034ef17 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b34463feeba35ffc7fce28f390a6fdd072dd2d8036f7a62d3dc0de5cc7a3f6 +size 7756 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/train_config.json b/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..6dbcbb45cc6ab3592379df4548bee3518cae49c4 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/200000/pretrained_model/train_config.json @@ -0,0 +1,226 @@ +{ + "dataset": { + "repo_id": "AI-Final/AI-aiCapstoneData-lerobot-advanced-v2-2", + "root": null, + "episodes": null, + "image_transforms": { + "enable": false, + "max_num_transforms": 3, + "random_order": false, + "tfs": { + "brightness": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "brightness": [ + 0.8, + 1.2 + ] + } + }, + "contrast": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "contrast": [ + 0.8, + 1.2 + ] + } + }, + "saturation": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "saturation": [ + 0.5, + 1.5 + ] + } + }, + "hue": { + "weight": 1.0, + "type": "ColorJitter", + "kwargs": { + "hue": [ + -0.05, + 0.05 + ] + } + }, + "sharpness": { + "weight": 1.0, + "type": "SharpnessJitter", + "kwargs": { + "sharpness": [ + 0.5, + 1.5 + ] + } + }, + "affine": { + "weight": 1.0, + "type": "RandomAffine", + "kwargs": { + "degrees": [ + -5.0, + 5.0 + ], + "translate": [ + 0.05, + 0.05 + ] + } + } + } + }, + "revision": "main", + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "diffusion", + "n_obs_steps": 2, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 9 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + }, + "observation.images.front": { + "type": "VISUAL", + "shape": [ + 3, + 480, + 640 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 8 + ] + } + }, + "device": "cuda", + "use_amp": false, + "use_peft": false, + "push_to_hub": true, + "repo_id": "AI-Final/advanced-diffusion-v3", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "horizon": 16, + "n_action_steps": 8, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MIN_MAX", + "ACTION": "MIN_MAX" + }, + "drop_n_last_frames": 7, + "vision_backbone": "resnet18", + "resize_shape": null, + "crop_ratio": 1.0, + "crop_shape": null, + "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, + "compile_model": false, + "compile_mode": "reduce-overhead", + "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 + }, + "output_dir": "outputs/train/act_dining_cleanup_v2_2_20260609_033312", + "job_name": "act_dining_cleanup_v2_2_20260609_033312", + "resume": false, + "seed": 1000, + "num_workers": 4, + "batch_size": 16, + "steps": 200000, + "eval_freq": 0, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 20000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adam", + "lr": 0.0001, + "weight_decay": 1e-06, + "grad_clip_norm": 10.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": { + "type": "diffuser", + "num_warmup_steps": 500, + "name": "cosine" + }, + "eval": { + "n_episodes": 50, + "batch_size": 50, + "use_async_envs": false + }, + "wandb": { + "enable": false, + "disable_artifact": false, + "project": "lerobot", + "entity": null, + "notes": null, + "run_id": null, + "mode": null + }, + "peft": null, + "use_rabc": false, + "rabc_progress_path": null, + "rabc_kappa": 0.01, + "rabc_epsilon": 1e-06, + "rabc_head_mode": "sparse", + "rename_map": {}, + "checkpoint_path": null +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/200000/training_state/optimizer_param_groups.json b/training_runs/advanced-diffusion-v3/checkpoints/200000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..5ad4a0652f3e90db3740c824423694349bd048f0 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/200000/training_state/optimizer_param_groups.json @@ -0,0 +1,233 @@ +[ + { + "lr": 0.0, + "betas": [ + 0.95, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 1e-06, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": false, + "initial_lr": 0.0001, + "params": [ + 0, + 1, + 2, + 3, + 4, + 5, + 6, + 7, + 8, + 9, + 10, + 11, + 12, + 13, + 14, + 15, + 16, + 17, + 18, + 19, + 20, + 21, + 22, + 23, + 24, + 25, + 26, + 27, + 28, + 29, + 30, + 31, + 32, + 33, + 34, + 35, + 36, + 37, + 38, + 39, + 40, + 41, + 42, + 43, + 44, + 45, + 46, + 47, + 48, + 49, + 50, + 51, + 52, + 53, + 54, + 55, + 56, + 57, + 58, + 59, + 60, + 61, + 62, + 63, + 64, + 65, + 66, + 67, + 68, + 69, + 70, + 71, + 72, + 73, + 74, + 75, + 76, + 77, + 78, + 79, + 80, + 81, + 82, + 83, + 84, + 85, + 86, + 87, + 88, + 89, + 90, + 91, + 92, + 93, + 94, + 95, + 96, + 97, + 98, + 99, + 100, + 101, + 102, + 103, + 104, + 105, + 106, + 107, + 108, + 109, + 110, + 111, + 112, + 113, + 114, + 115, + 116, + 117, + 118, + 119, + 120, + 121, + 122, + 123, + 124, + 125, + 126, + 127, + 128, + 129, + 130, + 131, + 132, + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152, + 153, + 154, + 155, + 156, + 157, + 158, + 159, + 160, + 161, + 162, + 163, + 164, + 165, + 166, + 167, + 168, + 169, + 170, + 171, + 172, + 173, + 174, + 175, + 176, + 177, + 178, + 179, + 180, + 181, + 182, + 183, + 184, + 185, + 186, + 187, + 188, + 189, + 190, + 191, + 192, + 193, + 194, + 195, + 196, + 197, + 198, + 199, + 200, + 201, + 202, + 203, + 204, + 205, + 206, + 207, + 208, + 209, + 210, + 211 + ] + } +] \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/200000/training_state/optimizer_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/200000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..422a4a70bd1041246ab3b313d8752584cddc055d --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/200000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b8227671d8b899efb78823ee6891026793f553f4872511b95392e9acdbc0eaea +size 2134472696 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/200000/training_state/rng_state.safetensors b/training_runs/advanced-diffusion-v3/checkpoints/200000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..223d9b9c879f0c854f44dca337c3901a4bb9677e --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/200000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ef131120ff908970c66db47b853a16b1318fec8e94e0db609f7cb8cd9c53b1f9 +size 15708 diff --git a/training_runs/advanced-diffusion-v3/checkpoints/200000/training_state/scheduler_state.json b/training_runs/advanced-diffusion-v3/checkpoints/200000/training_state/scheduler_state.json new file mode 100644 index 0000000000000000000000000000000000000000..ebeb5e218c0a3e73559a74ead254ad4f8e30632e --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/200000/training_state/scheduler_state.json @@ -0,0 +1,15 @@ +{ + "base_lrs": [ + 0.0001 + ], + "last_epoch": 200000, + "_step_count": 200001, + "_is_initial": false, + "_get_lr_called_within_step": false, + "_last_lr": [ + 0.0 + ], + "lr_lambdas": [ + null + ] +} \ No newline at end of file diff --git a/training_runs/advanced-diffusion-v3/checkpoints/200000/training_state/training_step.json b/training_runs/advanced-diffusion-v3/checkpoints/200000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..bcef8d8b05253341579ace0987c46cee6885c961 --- /dev/null +++ b/training_runs/advanced-diffusion-v3/checkpoints/200000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 200000 +} \ No newline at end of file