diff --git a/checkpoints/005000/pretrained_model/config.json b/checkpoints/005000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..8f6bc13d7275d238cd636521a3c6fcebf19f46f9 --- /dev/null +++ b/checkpoints/005000/pretrained_model/config.json @@ -0,0 +1,71 @@ +{ + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 +} \ No newline at end of file diff --git a/checkpoints/005000/pretrained_model/model.safetensors b/checkpoints/005000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..5b97efb78de2a776cb294d1abc6d3ffd9f80029e --- /dev/null +++ b/checkpoints/005000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cf5bfc717b4f3e0be4c24a62da797df7fd8def88c90e32d5a5787c50c426bb07 +size 206413008 diff --git a/checkpoints/005000/pretrained_model/policy_postprocessor.json b/checkpoints/005000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..9ccb342474cecbf2bdbb034d27169a351ad96d2a --- /dev/null +++ b/checkpoints/005000/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": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "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/checkpoints/005000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/005000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/005000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/005000/pretrained_model/policy_preprocessor.json b/checkpoints/005000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..d4cf9cc28b244fdc3cd036343eb06e691d8ac81c --- /dev/null +++ b/checkpoints/005000/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": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/005000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/005000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/005000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/005000/pretrained_model/train_config.json b/checkpoints/005000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a83affbd21b6568c590999eaa904b7b2ca8b1033 --- /dev/null +++ b/checkpoints/005000/pretrained_model/train_config.json @@ -0,0 +1,201 @@ +{ + "dataset": { + "repo_id": "TANAY779/record_actV2", + "root": "/home/.cache/huggingface/hub/datasets--TANAY779--record_actV2/snapshots/aefc5fdee896149dad7930032211f70707c24c7a", + "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": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 + }, + "output_dir": "outputs/train/act_so101_v2_TORCHCODEC", + "job_name": "act_so101_v2_TORCHCODEC", + "resume": false, + "seed": 1000, + "num_workers": 2, + "batch_size": 8, + "steps": 80000, + "eval_freq": 20000, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 5000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adamw", + "lr": 1e-05, + "weight_decay": 0.0001, + "grad_clip_norm": 10.0, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": null, + "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/checkpoints/005000/training_state/optimizer_param_groups.json b/checkpoints/005000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..33506df097b87c64b2aa15765ab89706780c0dc0 --- /dev/null +++ b/checkpoints/005000/training_state/optimizer_param_groups.json @@ -0,0 +1,191 @@ +[ + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "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 + ] + }, + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "params": [ + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152 + ] + } +] \ No newline at end of file diff --git a/checkpoints/005000/training_state/optimizer_state.safetensors b/checkpoints/005000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..5255c9c85edec360c7d6fbf49af9bb60a0783c2d --- /dev/null +++ b/checkpoints/005000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:45d4e3ae0da309f84629c303b99a0d11c670c1dc82c40a604d9cb1dbddf49053 +size 412530932 diff --git a/checkpoints/005000/training_state/rng_state.safetensors b/checkpoints/005000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e06b115594e4f0073e4f1411882dad579fc66cec --- /dev/null +++ b/checkpoints/005000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:70c9fc782d8dc2a4ba195a1b264e947db4a0cbb67d9897ff47bd8944df7484ca +size 15708 diff --git a/checkpoints/005000/training_state/training_step.json b/checkpoints/005000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..592449d3fc8b35c30c604eb1dabe60537e8224a0 --- /dev/null +++ b/checkpoints/005000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 5000 +} \ No newline at end of file diff --git a/checkpoints/010000/pretrained_model/config.json b/checkpoints/010000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..8f6bc13d7275d238cd636521a3c6fcebf19f46f9 --- /dev/null +++ b/checkpoints/010000/pretrained_model/config.json @@ -0,0 +1,71 @@ +{ + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 +} \ No newline at end of file diff --git a/checkpoints/010000/pretrained_model/model.safetensors b/checkpoints/010000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..08d3bcd310af4ac37dd7ee06d1747cce12440f4e --- /dev/null +++ b/checkpoints/010000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3ba3169a5d69e6a2e098c18877953a4289cf719c30e3a883c7eaae8128ad7100 +size 206413008 diff --git a/checkpoints/010000/pretrained_model/policy_postprocessor.json b/checkpoints/010000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..9ccb342474cecbf2bdbb034d27169a351ad96d2a --- /dev/null +++ b/checkpoints/010000/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": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "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/checkpoints/010000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/010000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/010000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/010000/pretrained_model/policy_preprocessor.json b/checkpoints/010000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..d4cf9cc28b244fdc3cd036343eb06e691d8ac81c --- /dev/null +++ b/checkpoints/010000/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": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/010000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/010000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/010000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/010000/pretrained_model/train_config.json b/checkpoints/010000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a83affbd21b6568c590999eaa904b7b2ca8b1033 --- /dev/null +++ b/checkpoints/010000/pretrained_model/train_config.json @@ -0,0 +1,201 @@ +{ + "dataset": { + "repo_id": "TANAY779/record_actV2", + "root": "/home/.cache/huggingface/hub/datasets--TANAY779--record_actV2/snapshots/aefc5fdee896149dad7930032211f70707c24c7a", + "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": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 + }, + "output_dir": "outputs/train/act_so101_v2_TORCHCODEC", + "job_name": "act_so101_v2_TORCHCODEC", + "resume": false, + "seed": 1000, + "num_workers": 2, + "batch_size": 8, + "steps": 80000, + "eval_freq": 20000, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 5000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adamw", + "lr": 1e-05, + "weight_decay": 0.0001, + "grad_clip_norm": 10.0, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": null, + "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/checkpoints/010000/training_state/optimizer_param_groups.json b/checkpoints/010000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..33506df097b87c64b2aa15765ab89706780c0dc0 --- /dev/null +++ b/checkpoints/010000/training_state/optimizer_param_groups.json @@ -0,0 +1,191 @@ +[ + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "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 + ] + }, + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "params": [ + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152 + ] + } +] \ No newline at end of file diff --git a/checkpoints/010000/training_state/optimizer_state.safetensors b/checkpoints/010000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..2e4f22e34f5322d335afecc1f23318cd9c51288b --- /dev/null +++ b/checkpoints/010000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:48e263cf164af4b2e3339bba0cea2b94757ec2e9a03cac5dae209ca942be8cb4 +size 412530932 diff --git a/checkpoints/010000/training_state/rng_state.safetensors b/checkpoints/010000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..438db712805cb626383e966f5b585b44fdb3319b --- /dev/null +++ b/checkpoints/010000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:438ad611d4451427708fbb6ac1575768a005b406592b6a1a70a3e6a2c14303f2 +size 15708 diff --git a/checkpoints/010000/training_state/training_step.json b/checkpoints/010000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..7cb7c0986e9e7461ca851ce71e95d235ae3d2732 --- /dev/null +++ b/checkpoints/010000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 10000 +} \ No newline at end of file diff --git a/checkpoints/015000/pretrained_model/config.json b/checkpoints/015000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..8f6bc13d7275d238cd636521a3c6fcebf19f46f9 --- /dev/null +++ b/checkpoints/015000/pretrained_model/config.json @@ -0,0 +1,71 @@ +{ + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 +} \ No newline at end of file diff --git a/checkpoints/015000/pretrained_model/model.safetensors b/checkpoints/015000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c43e2bf81320476799c5a1aacd186177088b7baf --- /dev/null +++ b/checkpoints/015000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5dae177128b56b28b33f8ee29f2311287789b52c03c91762c2a56ad31b82435d +size 206413008 diff --git a/checkpoints/015000/pretrained_model/policy_postprocessor.json b/checkpoints/015000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..9ccb342474cecbf2bdbb034d27169a351ad96d2a --- /dev/null +++ b/checkpoints/015000/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": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "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/checkpoints/015000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/015000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/015000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/015000/pretrained_model/policy_preprocessor.json b/checkpoints/015000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..d4cf9cc28b244fdc3cd036343eb06e691d8ac81c --- /dev/null +++ b/checkpoints/015000/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": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/015000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/015000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/015000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/015000/pretrained_model/train_config.json b/checkpoints/015000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a83affbd21b6568c590999eaa904b7b2ca8b1033 --- /dev/null +++ b/checkpoints/015000/pretrained_model/train_config.json @@ -0,0 +1,201 @@ +{ + "dataset": { + "repo_id": "TANAY779/record_actV2", + "root": "/home/.cache/huggingface/hub/datasets--TANAY779--record_actV2/snapshots/aefc5fdee896149dad7930032211f70707c24c7a", + "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": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 + }, + "output_dir": "outputs/train/act_so101_v2_TORCHCODEC", + "job_name": "act_so101_v2_TORCHCODEC", + "resume": false, + "seed": 1000, + "num_workers": 2, + "batch_size": 8, + "steps": 80000, + "eval_freq": 20000, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 5000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adamw", + "lr": 1e-05, + "weight_decay": 0.0001, + "grad_clip_norm": 10.0, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": null, + "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/checkpoints/015000/training_state/optimizer_param_groups.json b/checkpoints/015000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..33506df097b87c64b2aa15765ab89706780c0dc0 --- /dev/null +++ b/checkpoints/015000/training_state/optimizer_param_groups.json @@ -0,0 +1,191 @@ +[ + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "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 + ] + }, + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "params": [ + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152 + ] + } +] \ No newline at end of file diff --git a/checkpoints/015000/training_state/optimizer_state.safetensors b/checkpoints/015000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..624dfb48f5eec33059980af1f7616fcea0b42e6d --- /dev/null +++ b/checkpoints/015000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1b56dfe7945fcef6d94cb8fadfa2e4a23f3198cd44d33bc8644f19f604c94276 +size 412530932 diff --git a/checkpoints/015000/training_state/rng_state.safetensors b/checkpoints/015000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..8a931f39593b069dcfe5401d0b10cdf41a1dfd45 --- /dev/null +++ b/checkpoints/015000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a444ee91099e5d18c9ae955618a72f8f967ebce9d0918e5f2634bee55549c982 +size 15708 diff --git a/checkpoints/015000/training_state/training_step.json b/checkpoints/015000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..5cec056c8ba7f3c2e865a0f73ae59975a0503067 --- /dev/null +++ b/checkpoints/015000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 15000 +} \ No newline at end of file diff --git a/checkpoints/020000/pretrained_model/config.json b/checkpoints/020000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..8f6bc13d7275d238cd636521a3c6fcebf19f46f9 --- /dev/null +++ b/checkpoints/020000/pretrained_model/config.json @@ -0,0 +1,71 @@ +{ + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 +} \ No newline at end of file diff --git a/checkpoints/020000/pretrained_model/model.safetensors b/checkpoints/020000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..3a5b5bc655c368890e9f304b0965277961e17f0b --- /dev/null +++ b/checkpoints/020000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9f9b6d8efc76e552deb10bf13238adc633fc1a1de7dcdbd85e48d2d6256f77 +size 206413008 diff --git a/checkpoints/020000/pretrained_model/policy_postprocessor.json b/checkpoints/020000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..9ccb342474cecbf2bdbb034d27169a351ad96d2a --- /dev/null +++ b/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": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "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/checkpoints/020000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/020000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/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:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/020000/pretrained_model/policy_preprocessor.json b/checkpoints/020000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..d4cf9cc28b244fdc3cd036343eb06e691d8ac81c --- /dev/null +++ b/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": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/020000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/020000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/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:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/020000/pretrained_model/train_config.json b/checkpoints/020000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a83affbd21b6568c590999eaa904b7b2ca8b1033 --- /dev/null +++ b/checkpoints/020000/pretrained_model/train_config.json @@ -0,0 +1,201 @@ +{ + "dataset": { + "repo_id": "TANAY779/record_actV2", + "root": "/home/.cache/huggingface/hub/datasets--TANAY779--record_actV2/snapshots/aefc5fdee896149dad7930032211f70707c24c7a", + "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": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 + }, + "output_dir": "outputs/train/act_so101_v2_TORCHCODEC", + "job_name": "act_so101_v2_TORCHCODEC", + "resume": false, + "seed": 1000, + "num_workers": 2, + "batch_size": 8, + "steps": 80000, + "eval_freq": 20000, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 5000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adamw", + "lr": 1e-05, + "weight_decay": 0.0001, + "grad_clip_norm": 10.0, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": null, + "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/checkpoints/020000/training_state/optimizer_param_groups.json b/checkpoints/020000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..33506df097b87c64b2aa15765ab89706780c0dc0 --- /dev/null +++ b/checkpoints/020000/training_state/optimizer_param_groups.json @@ -0,0 +1,191 @@ +[ + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "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 + ] + }, + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "params": [ + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152 + ] + } +] \ No newline at end of file diff --git a/checkpoints/020000/training_state/optimizer_state.safetensors b/checkpoints/020000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..04a6cc7e96dd0d80837a64d3ddc72b1400a1ef48 --- /dev/null +++ b/checkpoints/020000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:266c06836bedd7f119a3f96a4e0ecc8d78c22b29bd663de8e4f0ed2f531bf9dc +size 412530932 diff --git a/checkpoints/020000/training_state/rng_state.safetensors b/checkpoints/020000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..1f38f0c38fdd35317064434ebb792bc615cad872 --- /dev/null +++ b/checkpoints/020000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ffc4282043c36454ecc8e81e161dc29b612f804e9102b2fec3704cd06840cbd5 +size 15708 diff --git a/checkpoints/020000/training_state/training_step.json b/checkpoints/020000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..dc9bb47026c5d5237ca6fc5dbff6020dd122ea05 --- /dev/null +++ b/checkpoints/020000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 20000 +} \ No newline at end of file diff --git a/checkpoints/025000/pretrained_model/config.json b/checkpoints/025000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..8f6bc13d7275d238cd636521a3c6fcebf19f46f9 --- /dev/null +++ b/checkpoints/025000/pretrained_model/config.json @@ -0,0 +1,71 @@ +{ + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 +} \ No newline at end of file diff --git a/checkpoints/025000/pretrained_model/model.safetensors b/checkpoints/025000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d0d9effebf48df850ca778e7e8f0a9d30a97d313 --- /dev/null +++ b/checkpoints/025000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2d9026a56f8fc47978470bb1e0bceb068b00afe8ad81e42ae69192f995d63c46 +size 206413008 diff --git a/checkpoints/025000/pretrained_model/policy_postprocessor.json b/checkpoints/025000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..9ccb342474cecbf2bdbb034d27169a351ad96d2a --- /dev/null +++ b/checkpoints/025000/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": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "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/checkpoints/025000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/025000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/025000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/025000/pretrained_model/policy_preprocessor.json b/checkpoints/025000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..d4cf9cc28b244fdc3cd036343eb06e691d8ac81c --- /dev/null +++ b/checkpoints/025000/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": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/025000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/025000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/025000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/025000/pretrained_model/train_config.json b/checkpoints/025000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a83affbd21b6568c590999eaa904b7b2ca8b1033 --- /dev/null +++ b/checkpoints/025000/pretrained_model/train_config.json @@ -0,0 +1,201 @@ +{ + "dataset": { + "repo_id": "TANAY779/record_actV2", + "root": "/home/.cache/huggingface/hub/datasets--TANAY779--record_actV2/snapshots/aefc5fdee896149dad7930032211f70707c24c7a", + "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": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 + }, + "output_dir": "outputs/train/act_so101_v2_TORCHCODEC", + "job_name": "act_so101_v2_TORCHCODEC", + "resume": false, + "seed": 1000, + "num_workers": 2, + "batch_size": 8, + "steps": 80000, + "eval_freq": 20000, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 5000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adamw", + "lr": 1e-05, + "weight_decay": 0.0001, + "grad_clip_norm": 10.0, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": null, + "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/checkpoints/025000/training_state/optimizer_param_groups.json b/checkpoints/025000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..33506df097b87c64b2aa15765ab89706780c0dc0 --- /dev/null +++ b/checkpoints/025000/training_state/optimizer_param_groups.json @@ -0,0 +1,191 @@ +[ + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "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 + ] + }, + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "params": [ + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152 + ] + } +] \ No newline at end of file diff --git a/checkpoints/025000/training_state/optimizer_state.safetensors b/checkpoints/025000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..33eca4424df96b9430f33ac950f5d1dda386cbab --- /dev/null +++ b/checkpoints/025000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4d2bae9ef72761f10c7ca7aac7f8f5768814b3a086cf66af120c3a8eb123c839 +size 412530932 diff --git a/checkpoints/025000/training_state/rng_state.safetensors b/checkpoints/025000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..03b9a833af0de8323729ea42f3c931768ae3b33c --- /dev/null +++ b/checkpoints/025000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:97afaa1cc4905fbc35b2fb6053eb7a201865f8960ec89a79b4c16a6f6699e65d +size 15708 diff --git a/checkpoints/025000/training_state/training_step.json b/checkpoints/025000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..ac6903ae05e7bbaec1e2da710fdbe91d8dc3f37d --- /dev/null +++ b/checkpoints/025000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 25000 +} \ No newline at end of file diff --git a/checkpoints/030000/pretrained_model/config.json b/checkpoints/030000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..8f6bc13d7275d238cd636521a3c6fcebf19f46f9 --- /dev/null +++ b/checkpoints/030000/pretrained_model/config.json @@ -0,0 +1,71 @@ +{ + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 +} \ No newline at end of file diff --git a/checkpoints/030000/pretrained_model/model.safetensors b/checkpoints/030000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..1cd13cfd75eb8a91d610768396c9cef4293e2de2 --- /dev/null +++ b/checkpoints/030000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b6695b14766689f4307b3a37064cb7d24832f1449e6f4b3b56d0fe1795272507 +size 206413008 diff --git a/checkpoints/030000/pretrained_model/policy_postprocessor.json b/checkpoints/030000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..9ccb342474cecbf2bdbb034d27169a351ad96d2a --- /dev/null +++ b/checkpoints/030000/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": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "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/checkpoints/030000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/030000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/030000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/030000/pretrained_model/policy_preprocessor.json b/checkpoints/030000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..d4cf9cc28b244fdc3cd036343eb06e691d8ac81c --- /dev/null +++ b/checkpoints/030000/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": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/030000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/030000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/030000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/030000/pretrained_model/train_config.json b/checkpoints/030000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a83affbd21b6568c590999eaa904b7b2ca8b1033 --- /dev/null +++ b/checkpoints/030000/pretrained_model/train_config.json @@ -0,0 +1,201 @@ +{ + "dataset": { + "repo_id": "TANAY779/record_actV2", + "root": "/home/.cache/huggingface/hub/datasets--TANAY779--record_actV2/snapshots/aefc5fdee896149dad7930032211f70707c24c7a", + "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": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 + }, + "output_dir": "outputs/train/act_so101_v2_TORCHCODEC", + "job_name": "act_so101_v2_TORCHCODEC", + "resume": false, + "seed": 1000, + "num_workers": 2, + "batch_size": 8, + "steps": 80000, + "eval_freq": 20000, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 5000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adamw", + "lr": 1e-05, + "weight_decay": 0.0001, + "grad_clip_norm": 10.0, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": null, + "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/checkpoints/030000/training_state/optimizer_param_groups.json b/checkpoints/030000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..33506df097b87c64b2aa15765ab89706780c0dc0 --- /dev/null +++ b/checkpoints/030000/training_state/optimizer_param_groups.json @@ -0,0 +1,191 @@ +[ + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "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 + ] + }, + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "params": [ + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152 + ] + } +] \ No newline at end of file diff --git a/checkpoints/030000/training_state/optimizer_state.safetensors b/checkpoints/030000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..5adf83f133925a5a8151ec0f240e68e8c79452de --- /dev/null +++ b/checkpoints/030000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:54f7b2b5a87291c21f01bea59b1d3d8258725ab52868a948aba0b3ee46a98d95 +size 412530932 diff --git a/checkpoints/030000/training_state/rng_state.safetensors b/checkpoints/030000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..6b2ee81418aad974e59dcd2017545925247c0c91 --- /dev/null +++ b/checkpoints/030000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5fdcfd4158d21f79400b3c52b55631699b76724b2a84065ac087a9042def1995 +size 15708 diff --git a/checkpoints/030000/training_state/training_step.json b/checkpoints/030000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..f4945f660f45b332883dccfccf18d8b8815d916a --- /dev/null +++ b/checkpoints/030000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 30000 +} \ No newline at end of file diff --git a/checkpoints/035000/pretrained_model/config.json b/checkpoints/035000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..8f6bc13d7275d238cd636521a3c6fcebf19f46f9 --- /dev/null +++ b/checkpoints/035000/pretrained_model/config.json @@ -0,0 +1,71 @@ +{ + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 +} \ No newline at end of file diff --git a/checkpoints/035000/pretrained_model/model.safetensors b/checkpoints/035000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b0e567e30da688b5f6cbf13c928add4049006ed6 --- /dev/null +++ b/checkpoints/035000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9525420676090927f9975ca1964a4c7dc1192c0fec7d2f3424c3e148ddfb195c +size 206413008 diff --git a/checkpoints/035000/pretrained_model/policy_postprocessor.json b/checkpoints/035000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..9ccb342474cecbf2bdbb034d27169a351ad96d2a --- /dev/null +++ b/checkpoints/035000/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": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "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/checkpoints/035000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/035000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/035000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/035000/pretrained_model/policy_preprocessor.json b/checkpoints/035000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..d4cf9cc28b244fdc3cd036343eb06e691d8ac81c --- /dev/null +++ b/checkpoints/035000/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": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/035000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/035000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/035000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/035000/pretrained_model/train_config.json b/checkpoints/035000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a83affbd21b6568c590999eaa904b7b2ca8b1033 --- /dev/null +++ b/checkpoints/035000/pretrained_model/train_config.json @@ -0,0 +1,201 @@ +{ + "dataset": { + "repo_id": "TANAY779/record_actV2", + "root": "/home/.cache/huggingface/hub/datasets--TANAY779--record_actV2/snapshots/aefc5fdee896149dad7930032211f70707c24c7a", + "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": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 + }, + "output_dir": "outputs/train/act_so101_v2_TORCHCODEC", + "job_name": "act_so101_v2_TORCHCODEC", + "resume": false, + "seed": 1000, + "num_workers": 2, + "batch_size": 8, + "steps": 80000, + "eval_freq": 20000, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 5000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adamw", + "lr": 1e-05, + "weight_decay": 0.0001, + "grad_clip_norm": 10.0, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": null, + "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/checkpoints/035000/training_state/optimizer_param_groups.json b/checkpoints/035000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..33506df097b87c64b2aa15765ab89706780c0dc0 --- /dev/null +++ b/checkpoints/035000/training_state/optimizer_param_groups.json @@ -0,0 +1,191 @@ +[ + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "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 + ] + }, + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "params": [ + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152 + ] + } +] \ No newline at end of file diff --git a/checkpoints/035000/training_state/optimizer_state.safetensors b/checkpoints/035000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..7d23ad9f97fe655ab3920f7300cc3ff38ba72c0f --- /dev/null +++ b/checkpoints/035000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:46a2042528bd71e438ecb4cef9d2ce015b2ff3d14dbfc688ede302548e51e9fa +size 412530932 diff --git a/checkpoints/035000/training_state/rng_state.safetensors b/checkpoints/035000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..8565c55d935f055081b5f8d4232b622b8a5e9a42 --- /dev/null +++ b/checkpoints/035000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:19f70a7129cbb1a5a933a870e553798851d7f793ced917511ecf43084661750b +size 15708 diff --git a/checkpoints/035000/training_state/training_step.json b/checkpoints/035000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..60080824ceb9805d29c885917dc77e0277e52eb2 --- /dev/null +++ b/checkpoints/035000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 35000 +} \ No newline at end of file diff --git a/checkpoints/040000/pretrained_model/config.json b/checkpoints/040000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..8f6bc13d7275d238cd636521a3c6fcebf19f46f9 --- /dev/null +++ b/checkpoints/040000/pretrained_model/config.json @@ -0,0 +1,71 @@ +{ + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 +} \ No newline at end of file diff --git a/checkpoints/040000/pretrained_model/model.safetensors b/checkpoints/040000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..743de58481ce56b4d9adb269ba2d71aa97d07c80 --- /dev/null +++ b/checkpoints/040000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:22736d9e2a00212b088be1dae5c6f5711bc51981819c95dfd95d9aab9e8dcd6b +size 206413008 diff --git a/checkpoints/040000/pretrained_model/policy_postprocessor.json b/checkpoints/040000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..9ccb342474cecbf2bdbb034d27169a351ad96d2a --- /dev/null +++ b/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": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "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/checkpoints/040000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/040000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/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:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/040000/pretrained_model/policy_preprocessor.json b/checkpoints/040000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..d4cf9cc28b244fdc3cd036343eb06e691d8ac81c --- /dev/null +++ b/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": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/040000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/040000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/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:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/040000/pretrained_model/train_config.json b/checkpoints/040000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a83affbd21b6568c590999eaa904b7b2ca8b1033 --- /dev/null +++ b/checkpoints/040000/pretrained_model/train_config.json @@ -0,0 +1,201 @@ +{ + "dataset": { + "repo_id": "TANAY779/record_actV2", + "root": "/home/.cache/huggingface/hub/datasets--TANAY779--record_actV2/snapshots/aefc5fdee896149dad7930032211f70707c24c7a", + "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": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 + }, + "output_dir": "outputs/train/act_so101_v2_TORCHCODEC", + "job_name": "act_so101_v2_TORCHCODEC", + "resume": false, + "seed": 1000, + "num_workers": 2, + "batch_size": 8, + "steps": 80000, + "eval_freq": 20000, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 5000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adamw", + "lr": 1e-05, + "weight_decay": 0.0001, + "grad_clip_norm": 10.0, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": null, + "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/checkpoints/040000/training_state/optimizer_param_groups.json b/checkpoints/040000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..33506df097b87c64b2aa15765ab89706780c0dc0 --- /dev/null +++ b/checkpoints/040000/training_state/optimizer_param_groups.json @@ -0,0 +1,191 @@ +[ + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "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 + ] + }, + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "params": [ + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152 + ] + } +] \ No newline at end of file diff --git a/checkpoints/040000/training_state/optimizer_state.safetensors b/checkpoints/040000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e74cdd37e4bfcc230935e4a8635ab4de9b8a9346 --- /dev/null +++ b/checkpoints/040000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2e3d3dea60dd981279aec8e4b506976dd6718f71ab52f8324ce7b6f0632ef53d +size 412530932 diff --git a/checkpoints/040000/training_state/rng_state.safetensors b/checkpoints/040000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..42d17e974a9679d5aee0ace12948ae86886ff2f2 --- /dev/null +++ b/checkpoints/040000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:72d3210a429d7cee9d09a60c7fa6c95fcf74c9ed8f4ec3c81498cfbdb3ec454c +size 15708 diff --git a/checkpoints/040000/training_state/training_step.json b/checkpoints/040000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..fe40d8ff4312c7e5a8fba9bcc932a43a1384ba77 --- /dev/null +++ b/checkpoints/040000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 40000 +} \ No newline at end of file diff --git a/checkpoints/045000/pretrained_model/config.json b/checkpoints/045000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..8f6bc13d7275d238cd636521a3c6fcebf19f46f9 --- /dev/null +++ b/checkpoints/045000/pretrained_model/config.json @@ -0,0 +1,71 @@ +{ + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 +} \ No newline at end of file diff --git a/checkpoints/045000/pretrained_model/model.safetensors b/checkpoints/045000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b0e91d0b032f2d46332b5b646b11405eb3a15354 --- /dev/null +++ b/checkpoints/045000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c80a47acde6895da2186f430d47055116ed4d220da7922c02ad42f82dd401149 +size 206413008 diff --git a/checkpoints/045000/pretrained_model/policy_postprocessor.json b/checkpoints/045000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..9ccb342474cecbf2bdbb034d27169a351ad96d2a --- /dev/null +++ b/checkpoints/045000/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": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "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/checkpoints/045000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/045000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/045000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/045000/pretrained_model/policy_preprocessor.json b/checkpoints/045000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..d4cf9cc28b244fdc3cd036343eb06e691d8ac81c --- /dev/null +++ b/checkpoints/045000/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": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/045000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/045000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/045000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/045000/pretrained_model/train_config.json b/checkpoints/045000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a83affbd21b6568c590999eaa904b7b2ca8b1033 --- /dev/null +++ b/checkpoints/045000/pretrained_model/train_config.json @@ -0,0 +1,201 @@ +{ + "dataset": { + "repo_id": "TANAY779/record_actV2", + "root": "/home/.cache/huggingface/hub/datasets--TANAY779--record_actV2/snapshots/aefc5fdee896149dad7930032211f70707c24c7a", + "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": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 + }, + "output_dir": "outputs/train/act_so101_v2_TORCHCODEC", + "job_name": "act_so101_v2_TORCHCODEC", + "resume": false, + "seed": 1000, + "num_workers": 2, + "batch_size": 8, + "steps": 80000, + "eval_freq": 20000, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 5000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adamw", + "lr": 1e-05, + "weight_decay": 0.0001, + "grad_clip_norm": 10.0, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": null, + "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/checkpoints/045000/training_state/optimizer_param_groups.json b/checkpoints/045000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..33506df097b87c64b2aa15765ab89706780c0dc0 --- /dev/null +++ b/checkpoints/045000/training_state/optimizer_param_groups.json @@ -0,0 +1,191 @@ +[ + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "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 + ] + }, + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "params": [ + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152 + ] + } +] \ No newline at end of file diff --git a/checkpoints/045000/training_state/optimizer_state.safetensors b/checkpoints/045000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..55e35e8195a2c1056f14e9ae5aef708255f968ca --- /dev/null +++ b/checkpoints/045000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9c88dfaffe6cf3c02ebde6e9d0d2747a56c68432f8ff9d634fa1a74dcb427857 +size 412530932 diff --git a/checkpoints/045000/training_state/rng_state.safetensors b/checkpoints/045000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..70a1a7695a40caaced751ce10cbdcd1a66d3eec4 --- /dev/null +++ b/checkpoints/045000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:799d777019d7638fc716f5cb67cd57926f1a2294226f2b4a070b6e35718e14b0 +size 15708 diff --git a/checkpoints/045000/training_state/training_step.json b/checkpoints/045000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..472b301a48ca897c9f2b7da46adc0312cf9a91d9 --- /dev/null +++ b/checkpoints/045000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 45000 +} \ No newline at end of file diff --git a/checkpoints/050000/pretrained_model/config.json b/checkpoints/050000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..8f6bc13d7275d238cd636521a3c6fcebf19f46f9 --- /dev/null +++ b/checkpoints/050000/pretrained_model/config.json @@ -0,0 +1,71 @@ +{ + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 +} \ No newline at end of file diff --git a/checkpoints/050000/pretrained_model/model.safetensors b/checkpoints/050000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..ee4f5b6b4912bd5a0bfc6811402ae36fdf6c6f2d --- /dev/null +++ b/checkpoints/050000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:91489c4d26519818273731f255273a7cc22b72ce59322e2444d75e3821391ce3 +size 206413008 diff --git a/checkpoints/050000/pretrained_model/policy_postprocessor.json b/checkpoints/050000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..9ccb342474cecbf2bdbb034d27169a351ad96d2a --- /dev/null +++ b/checkpoints/050000/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": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "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/checkpoints/050000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/050000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/050000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/050000/pretrained_model/policy_preprocessor.json b/checkpoints/050000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..d4cf9cc28b244fdc3cd036343eb06e691d8ac81c --- /dev/null +++ b/checkpoints/050000/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": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/050000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/050000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/050000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/050000/pretrained_model/train_config.json b/checkpoints/050000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a83affbd21b6568c590999eaa904b7b2ca8b1033 --- /dev/null +++ b/checkpoints/050000/pretrained_model/train_config.json @@ -0,0 +1,201 @@ +{ + "dataset": { + "repo_id": "TANAY779/record_actV2", + "root": "/home/.cache/huggingface/hub/datasets--TANAY779--record_actV2/snapshots/aefc5fdee896149dad7930032211f70707c24c7a", + "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": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 + }, + "output_dir": "outputs/train/act_so101_v2_TORCHCODEC", + "job_name": "act_so101_v2_TORCHCODEC", + "resume": false, + "seed": 1000, + "num_workers": 2, + "batch_size": 8, + "steps": 80000, + "eval_freq": 20000, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 5000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adamw", + "lr": 1e-05, + "weight_decay": 0.0001, + "grad_clip_norm": 10.0, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": null, + "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/checkpoints/050000/training_state/optimizer_param_groups.json b/checkpoints/050000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..33506df097b87c64b2aa15765ab89706780c0dc0 --- /dev/null +++ b/checkpoints/050000/training_state/optimizer_param_groups.json @@ -0,0 +1,191 @@ +[ + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "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 + ] + }, + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "params": [ + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152 + ] + } +] \ No newline at end of file diff --git a/checkpoints/050000/training_state/optimizer_state.safetensors b/checkpoints/050000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..0b6500580f0eb088d97613bfbfda9a82dbe05dcf --- /dev/null +++ b/checkpoints/050000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4ad5c9b24656e2476b8f2c6356132dc98d298f3f1ddc90d3e154125aeffbd3b4 +size 412530932 diff --git a/checkpoints/050000/training_state/rng_state.safetensors b/checkpoints/050000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..f7e8b5803029f3bc041e045ca32a6d879b70b960 --- /dev/null +++ b/checkpoints/050000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:216ffb36ce1d87c6a7af5006cd68244e4ebbe2140a037e8bca8dcaa4af48bee4 +size 15708 diff --git a/checkpoints/050000/training_state/training_step.json b/checkpoints/050000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..b3c03563f17d6d98552117ccf00705b693aa3fbd --- /dev/null +++ b/checkpoints/050000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 50000 +} \ No newline at end of file diff --git a/checkpoints/055000/pretrained_model/config.json b/checkpoints/055000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..8f6bc13d7275d238cd636521a3c6fcebf19f46f9 --- /dev/null +++ b/checkpoints/055000/pretrained_model/config.json @@ -0,0 +1,71 @@ +{ + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 +} \ No newline at end of file diff --git a/checkpoints/055000/pretrained_model/model.safetensors b/checkpoints/055000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..da5f4bee04890cd2d128d0d535bf8ac2d797e91d --- /dev/null +++ b/checkpoints/055000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bc0f44c80a193a67b95c4f9ad041e8965c91df961a84a3400546951b1ce248a6 +size 206413008 diff --git a/checkpoints/055000/pretrained_model/policy_postprocessor.json b/checkpoints/055000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..9ccb342474cecbf2bdbb034d27169a351ad96d2a --- /dev/null +++ b/checkpoints/055000/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": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "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/checkpoints/055000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/055000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/055000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/055000/pretrained_model/policy_preprocessor.json b/checkpoints/055000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..d4cf9cc28b244fdc3cd036343eb06e691d8ac81c --- /dev/null +++ b/checkpoints/055000/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": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/055000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/055000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/055000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/055000/pretrained_model/train_config.json b/checkpoints/055000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a83affbd21b6568c590999eaa904b7b2ca8b1033 --- /dev/null +++ b/checkpoints/055000/pretrained_model/train_config.json @@ -0,0 +1,201 @@ +{ + "dataset": { + "repo_id": "TANAY779/record_actV2", + "root": "/home/.cache/huggingface/hub/datasets--TANAY779--record_actV2/snapshots/aefc5fdee896149dad7930032211f70707c24c7a", + "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": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 + }, + "output_dir": "outputs/train/act_so101_v2_TORCHCODEC", + "job_name": "act_so101_v2_TORCHCODEC", + "resume": false, + "seed": 1000, + "num_workers": 2, + "batch_size": 8, + "steps": 80000, + "eval_freq": 20000, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 5000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adamw", + "lr": 1e-05, + "weight_decay": 0.0001, + "grad_clip_norm": 10.0, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": null, + "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/checkpoints/055000/training_state/optimizer_param_groups.json b/checkpoints/055000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..33506df097b87c64b2aa15765ab89706780c0dc0 --- /dev/null +++ b/checkpoints/055000/training_state/optimizer_param_groups.json @@ -0,0 +1,191 @@ +[ + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "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 + ] + }, + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "params": [ + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152 + ] + } +] \ No newline at end of file diff --git a/checkpoints/055000/training_state/optimizer_state.safetensors b/checkpoints/055000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..776c620e8d0a010004634dc084931c1dda558ef4 --- /dev/null +++ b/checkpoints/055000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:017a64e2750ab5b88cd0e25a516c1d5c1ab71314cb2ce513153c5885b7db1801 +size 412530932 diff --git a/checkpoints/055000/training_state/rng_state.safetensors b/checkpoints/055000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..84ec33ccb0873ea6f3df43f11516f6fa26e7d4c7 --- /dev/null +++ b/checkpoints/055000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2971417abb0ecef1b4dde48aebd910e360bcf08a030c0e3ed27c2b0bf7f79f34 +size 15708 diff --git a/checkpoints/055000/training_state/training_step.json b/checkpoints/055000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..57c2477e950fdfdedf6a402eec3c3f0fe2d039ed --- /dev/null +++ b/checkpoints/055000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 55000 +} \ No newline at end of file diff --git a/checkpoints/060000/pretrained_model/config.json b/checkpoints/060000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..8f6bc13d7275d238cd636521a3c6fcebf19f46f9 --- /dev/null +++ b/checkpoints/060000/pretrained_model/config.json @@ -0,0 +1,71 @@ +{ + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 +} \ No newline at end of file diff --git a/checkpoints/060000/pretrained_model/model.safetensors b/checkpoints/060000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..7152def07d0af28d548e92b3d0bf509df60754f3 --- /dev/null +++ b/checkpoints/060000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3d3a4b6c79e37a7f1beeb8a696129670672bb463b7394b6e898ea57da534855f +size 206413008 diff --git a/checkpoints/060000/pretrained_model/policy_postprocessor.json b/checkpoints/060000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..9ccb342474cecbf2bdbb034d27169a351ad96d2a --- /dev/null +++ b/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": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "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/checkpoints/060000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/060000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/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:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/060000/pretrained_model/policy_preprocessor.json b/checkpoints/060000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..d4cf9cc28b244fdc3cd036343eb06e691d8ac81c --- /dev/null +++ b/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": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/060000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/060000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/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:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/060000/pretrained_model/train_config.json b/checkpoints/060000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a83affbd21b6568c590999eaa904b7b2ca8b1033 --- /dev/null +++ b/checkpoints/060000/pretrained_model/train_config.json @@ -0,0 +1,201 @@ +{ + "dataset": { + "repo_id": "TANAY779/record_actV2", + "root": "/home/.cache/huggingface/hub/datasets--TANAY779--record_actV2/snapshots/aefc5fdee896149dad7930032211f70707c24c7a", + "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": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 + }, + "output_dir": "outputs/train/act_so101_v2_TORCHCODEC", + "job_name": "act_so101_v2_TORCHCODEC", + "resume": false, + "seed": 1000, + "num_workers": 2, + "batch_size": 8, + "steps": 80000, + "eval_freq": 20000, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 5000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adamw", + "lr": 1e-05, + "weight_decay": 0.0001, + "grad_clip_norm": 10.0, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": null, + "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/checkpoints/060000/training_state/optimizer_param_groups.json b/checkpoints/060000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..33506df097b87c64b2aa15765ab89706780c0dc0 --- /dev/null +++ b/checkpoints/060000/training_state/optimizer_param_groups.json @@ -0,0 +1,191 @@ +[ + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "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 + ] + }, + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "params": [ + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152 + ] + } +] \ No newline at end of file diff --git a/checkpoints/060000/training_state/optimizer_state.safetensors b/checkpoints/060000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..f44e54fc5eb0a5f220b77fe215ccf2d954b1d5e3 --- /dev/null +++ b/checkpoints/060000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:10ac198b409bd01fee39610c9f952be8fd23495ffc1c6c02fe2b1e7ffeafbed1 +size 412530932 diff --git a/checkpoints/060000/training_state/rng_state.safetensors b/checkpoints/060000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..45e021fcc56cac7c57ab39321df645f4dc596d4c --- /dev/null +++ b/checkpoints/060000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fb116d03b0c57ffbeb7370607fb9e5d53008c828c44e56e5ba415948691be472 +size 15708 diff --git a/checkpoints/060000/training_state/training_step.json b/checkpoints/060000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..d360b6037f1c8c2c85d38e951160eb02ace507dd --- /dev/null +++ b/checkpoints/060000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 60000 +} \ No newline at end of file diff --git a/checkpoints/065000/pretrained_model/config.json b/checkpoints/065000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..8f6bc13d7275d238cd636521a3c6fcebf19f46f9 --- /dev/null +++ b/checkpoints/065000/pretrained_model/config.json @@ -0,0 +1,71 @@ +{ + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 +} \ No newline at end of file diff --git a/checkpoints/065000/pretrained_model/model.safetensors b/checkpoints/065000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..076966d10dd10818b6d0650a262f67564e19aabf --- /dev/null +++ b/checkpoints/065000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a9edddc4d3d6927e499547de499b39ba2223172b6e2509badc3d265fea9eaf55 +size 206413008 diff --git a/checkpoints/065000/pretrained_model/policy_postprocessor.json b/checkpoints/065000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..9ccb342474cecbf2bdbb034d27169a351ad96d2a --- /dev/null +++ b/checkpoints/065000/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": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "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/checkpoints/065000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/065000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/065000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/065000/pretrained_model/policy_preprocessor.json b/checkpoints/065000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..d4cf9cc28b244fdc3cd036343eb06e691d8ac81c --- /dev/null +++ b/checkpoints/065000/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": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/065000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/065000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/065000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/065000/pretrained_model/train_config.json b/checkpoints/065000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a83affbd21b6568c590999eaa904b7b2ca8b1033 --- /dev/null +++ b/checkpoints/065000/pretrained_model/train_config.json @@ -0,0 +1,201 @@ +{ + "dataset": { + "repo_id": "TANAY779/record_actV2", + "root": "/home/.cache/huggingface/hub/datasets--TANAY779--record_actV2/snapshots/aefc5fdee896149dad7930032211f70707c24c7a", + "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": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 + }, + "output_dir": "outputs/train/act_so101_v2_TORCHCODEC", + "job_name": "act_so101_v2_TORCHCODEC", + "resume": false, + "seed": 1000, + "num_workers": 2, + "batch_size": 8, + "steps": 80000, + "eval_freq": 20000, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 5000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adamw", + "lr": 1e-05, + "weight_decay": 0.0001, + "grad_clip_norm": 10.0, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": null, + "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/checkpoints/065000/training_state/optimizer_param_groups.json b/checkpoints/065000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..33506df097b87c64b2aa15765ab89706780c0dc0 --- /dev/null +++ b/checkpoints/065000/training_state/optimizer_param_groups.json @@ -0,0 +1,191 @@ +[ + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "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 + ] + }, + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "params": [ + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152 + ] + } +] \ No newline at end of file diff --git a/checkpoints/065000/training_state/optimizer_state.safetensors b/checkpoints/065000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..2411b36ee3420fe285d90f1639543f25719c5936 --- /dev/null +++ b/checkpoints/065000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a03174820dfaddfba49096491431897ee575db26f98430e0a1c7586e602916ad +size 412530932 diff --git a/checkpoints/065000/training_state/rng_state.safetensors b/checkpoints/065000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..09c23c9b2c32fe9050035040fce2535ce852d640 --- /dev/null +++ b/checkpoints/065000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:30bd702b84ef5356f4e6940203a811b7f248d35660f3eea36053d350cc8cfa96 +size 15708 diff --git a/checkpoints/065000/training_state/training_step.json b/checkpoints/065000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..29c11af0e7f84ef97479db014fb5c2ce76abe095 --- /dev/null +++ b/checkpoints/065000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 65000 +} \ No newline at end of file diff --git a/checkpoints/070000/pretrained_model/config.json b/checkpoints/070000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..8f6bc13d7275d238cd636521a3c6fcebf19f46f9 --- /dev/null +++ b/checkpoints/070000/pretrained_model/config.json @@ -0,0 +1,71 @@ +{ + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 +} \ No newline at end of file diff --git a/checkpoints/070000/pretrained_model/model.safetensors b/checkpoints/070000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..f467d6c1576e0e3c8d28447968608c5aeae8f958 --- /dev/null +++ b/checkpoints/070000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3317b9224ad34f29da5cedb818ad14d46c58a4242260e0b29a18057185bab141 +size 206413008 diff --git a/checkpoints/070000/pretrained_model/policy_postprocessor.json b/checkpoints/070000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..9ccb342474cecbf2bdbb034d27169a351ad96d2a --- /dev/null +++ b/checkpoints/070000/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": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "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/checkpoints/070000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/070000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/070000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/070000/pretrained_model/policy_preprocessor.json b/checkpoints/070000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..d4cf9cc28b244fdc3cd036343eb06e691d8ac81c --- /dev/null +++ b/checkpoints/070000/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": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/070000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/070000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/070000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/070000/pretrained_model/train_config.json b/checkpoints/070000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a83affbd21b6568c590999eaa904b7b2ca8b1033 --- /dev/null +++ b/checkpoints/070000/pretrained_model/train_config.json @@ -0,0 +1,201 @@ +{ + "dataset": { + "repo_id": "TANAY779/record_actV2", + "root": "/home/.cache/huggingface/hub/datasets--TANAY779--record_actV2/snapshots/aefc5fdee896149dad7930032211f70707c24c7a", + "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": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 + }, + "output_dir": "outputs/train/act_so101_v2_TORCHCODEC", + "job_name": "act_so101_v2_TORCHCODEC", + "resume": false, + "seed": 1000, + "num_workers": 2, + "batch_size": 8, + "steps": 80000, + "eval_freq": 20000, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 5000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adamw", + "lr": 1e-05, + "weight_decay": 0.0001, + "grad_clip_norm": 10.0, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": null, + "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/checkpoints/070000/training_state/optimizer_param_groups.json b/checkpoints/070000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..33506df097b87c64b2aa15765ab89706780c0dc0 --- /dev/null +++ b/checkpoints/070000/training_state/optimizer_param_groups.json @@ -0,0 +1,191 @@ +[ + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "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 + ] + }, + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "params": [ + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152 + ] + } +] \ No newline at end of file diff --git a/checkpoints/070000/training_state/optimizer_state.safetensors b/checkpoints/070000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..3e590748c68713af81c3bf588f09f8ba9c25f59b --- /dev/null +++ b/checkpoints/070000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eaa6ad368ea733b48d1d2f75aac2ccb5015a1404176927155d756e695b6ccf68 +size 412530932 diff --git a/checkpoints/070000/training_state/rng_state.safetensors b/checkpoints/070000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..ffeacb4ad21a1519cca3e8f8f9e6a0a0aa63164a --- /dev/null +++ b/checkpoints/070000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5c1d76c34f405f072d2dc9f9faa91cab630f76cdd1452ef0d209d0e5af8cc561 +size 15708 diff --git a/checkpoints/070000/training_state/training_step.json b/checkpoints/070000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..1ec0c1eb0f525e34fa3a57743f750b2b34b5b77b --- /dev/null +++ b/checkpoints/070000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 70000 +} \ No newline at end of file diff --git a/checkpoints/075000/pretrained_model/config.json b/checkpoints/075000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..8f6bc13d7275d238cd636521a3c6fcebf19f46f9 --- /dev/null +++ b/checkpoints/075000/pretrained_model/config.json @@ -0,0 +1,71 @@ +{ + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 +} \ No newline at end of file diff --git a/checkpoints/075000/pretrained_model/model.safetensors b/checkpoints/075000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..19eb8adbeea6457c788444dd95680af8c74f2259 --- /dev/null +++ b/checkpoints/075000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0d0ce952b70712c4a509a2a9f0326977ccbd3996d5e9bac6617e01e1e8008b80 +size 206413008 diff --git a/checkpoints/075000/pretrained_model/policy_postprocessor.json b/checkpoints/075000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..9ccb342474cecbf2bdbb034d27169a351ad96d2a --- /dev/null +++ b/checkpoints/075000/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": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "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/checkpoints/075000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/075000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/075000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/075000/pretrained_model/policy_preprocessor.json b/checkpoints/075000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..d4cf9cc28b244fdc3cd036343eb06e691d8ac81c --- /dev/null +++ b/checkpoints/075000/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": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/075000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/075000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/checkpoints/075000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/075000/pretrained_model/train_config.json b/checkpoints/075000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a83affbd21b6568c590999eaa904b7b2ca8b1033 --- /dev/null +++ b/checkpoints/075000/pretrained_model/train_config.json @@ -0,0 +1,201 @@ +{ + "dataset": { + "repo_id": "TANAY779/record_actV2", + "root": "/home/.cache/huggingface/hub/datasets--TANAY779--record_actV2/snapshots/aefc5fdee896149dad7930032211f70707c24c7a", + "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": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 + }, + "output_dir": "outputs/train/act_so101_v2_TORCHCODEC", + "job_name": "act_so101_v2_TORCHCODEC", + "resume": false, + "seed": 1000, + "num_workers": 2, + "batch_size": 8, + "steps": 80000, + "eval_freq": 20000, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 5000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adamw", + "lr": 1e-05, + "weight_decay": 0.0001, + "grad_clip_norm": 10.0, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": null, + "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/checkpoints/075000/training_state/optimizer_param_groups.json b/checkpoints/075000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..33506df097b87c64b2aa15765ab89706780c0dc0 --- /dev/null +++ b/checkpoints/075000/training_state/optimizer_param_groups.json @@ -0,0 +1,191 @@ +[ + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "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 + ] + }, + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "params": [ + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152 + ] + } +] \ No newline at end of file diff --git a/checkpoints/075000/training_state/optimizer_state.safetensors b/checkpoints/075000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..37c173c34f326484b2bee44d3b2dc38f7b6ed564 --- /dev/null +++ b/checkpoints/075000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0089f46ade9dd771faf02526a0aa714cd6bca6d0ca10b6c3ed282b8581314b2a +size 412530932 diff --git a/checkpoints/075000/training_state/rng_state.safetensors b/checkpoints/075000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..84cf25aae31ab5169ec341b27d5b1c00d391e57e --- /dev/null +++ b/checkpoints/075000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3938af993f47c816afdc2acd858266ad694cba887edb6c4e437e6de88daa529b +size 15708 diff --git a/checkpoints/075000/training_state/training_step.json b/checkpoints/075000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..19674b7211fb0b7620c8755c555877b160e5a745 --- /dev/null +++ b/checkpoints/075000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 75000 +} \ No newline at end of file diff --git a/checkpoints/080000/pretrained_model/config.json b/checkpoints/080000/pretrained_model/config.json new file mode 100644 index 0000000000000000000000000000000000000000..8f6bc13d7275d238cd636521a3c6fcebf19f46f9 --- /dev/null +++ b/checkpoints/080000/pretrained_model/config.json @@ -0,0 +1,71 @@ +{ + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 +} \ No newline at end of file diff --git a/checkpoints/080000/pretrained_model/model.safetensors b/checkpoints/080000/pretrained_model/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b9cabd0757d19ff9ea13695d7d832238c8c7ab8f --- /dev/null +++ b/checkpoints/080000/pretrained_model/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9f36c12974fb6053d591cba1d75430bf10a776538c1e930c87ee2efd76eb6d1b +size 206413008 diff --git a/checkpoints/080000/pretrained_model/policy_postprocessor.json b/checkpoints/080000/pretrained_model/policy_postprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..9ccb342474cecbf2bdbb034d27169a351ad96d2a --- /dev/null +++ b/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": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "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/checkpoints/080000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors b/checkpoints/080000/pretrained_model/policy_postprocessor_step_0_unnormalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/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:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/080000/pretrained_model/policy_preprocessor.json b/checkpoints/080000/pretrained_model/policy_preprocessor.json new file mode 100644 index 0000000000000000000000000000000000000000..d4cf9cc28b244fdc3cd036343eb06e691d8ac81c --- /dev/null +++ b/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": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "norm_map": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + } + }, + "state_file": "policy_preprocessor_step_3_normalizer_processor.safetensors" + } + ] +} \ No newline at end of file diff --git a/checkpoints/080000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors b/checkpoints/080000/pretrained_model/policy_preprocessor_step_3_normalizer_processor.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb20cf0d901a0bff0205022f9af3b7e038bc82d9 --- /dev/null +++ b/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:dc715c57d972f144155d6d22f5d608fd760e1002ccdb2ad5fc97f2a4c0461a4b +size 7584 diff --git a/checkpoints/080000/pretrained_model/train_config.json b/checkpoints/080000/pretrained_model/train_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a83affbd21b6568c590999eaa904b7b2ca8b1033 --- /dev/null +++ b/checkpoints/080000/pretrained_model/train_config.json @@ -0,0 +1,201 @@ +{ + "dataset": { + "repo_id": "TANAY779/record_actV2", + "root": "/home/.cache/huggingface/hub/datasets--TANAY779--record_actV2/snapshots/aefc5fdee896149dad7930032211f70707c24c7a", + "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": null, + "use_imagenet_stats": true, + "video_backend": "torchcodec", + "streaming": false + }, + "env": null, + "policy": { + "type": "act", + "n_obs_steps": 1, + "input_features": { + "observation.state": { + "type": "STATE", + "shape": [ + 6 + ] + }, + "observation.images.overhead": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + }, + "observation.images.wrist": { + "type": "VISUAL", + "shape": [ + 3, + 960, + 1280 + ] + } + }, + "output_features": { + "action": { + "type": "ACTION", + "shape": [ + 6 + ] + } + }, + "device": "cuda", + "use_amp": true, + "use_peft": false, + "push_to_hub": true, + "repo_id": "TANAY779/act_so101_v2_TORCHCODEC", + "private": null, + "tags": null, + "license": null, + "pretrained_path": null, + "chunk_size": 30, + "n_action_steps": 30, + "normalization_mapping": { + "VISUAL": "MEAN_STD", + "STATE": "MEAN_STD", + "ACTION": "MEAN_STD" + }, + "vision_backbone": "resnet18", + "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1", + "replace_final_stride_with_dilation": false, + "pre_norm": false, + "dim_model": 512, + "n_heads": 8, + "dim_feedforward": 3200, + "feedforward_activation": "relu", + "n_encoder_layers": 4, + "n_decoder_layers": 1, + "use_vae": true, + "latent_dim": 32, + "n_vae_encoder_layers": 4, + "temporal_ensemble_coeff": null, + "dropout": 0.1, + "kl_weight": 10.0, + "optimizer_lr": 1e-05, + "optimizer_weight_decay": 0.0001, + "optimizer_lr_backbone": 1e-05 + }, + "output_dir": "outputs/train/act_so101_v2_TORCHCODEC", + "job_name": "act_so101_v2_TORCHCODEC", + "resume": false, + "seed": 1000, + "num_workers": 2, + "batch_size": 8, + "steps": 80000, + "eval_freq": 20000, + "log_freq": 200, + "tolerance_s": 0.0001, + "save_checkpoint": true, + "save_freq": 5000, + "use_policy_training_preset": true, + "optimizer": { + "type": "adamw", + "lr": 1e-05, + "weight_decay": 0.0001, + "grad_clip_norm": 10.0, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08 + }, + "scheduler": null, + "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/checkpoints/080000/training_state/optimizer_param_groups.json b/checkpoints/080000/training_state/optimizer_param_groups.json new file mode 100644 index 0000000000000000000000000000000000000000..33506df097b87c64b2aa15765ab89706780c0dc0 --- /dev/null +++ b/checkpoints/080000/training_state/optimizer_param_groups.json @@ -0,0 +1,191 @@ +[ + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "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 + ] + }, + { + "lr": 1e-05, + "betas": [ + 0.9, + 0.999 + ], + "eps": 1e-08, + "weight_decay": 0.0001, + "amsgrad": false, + "maximize": false, + "foreach": null, + "capturable": false, + "differentiable": false, + "fused": null, + "decoupled_weight_decay": true, + "params": [ + 133, + 134, + 135, + 136, + 137, + 138, + 139, + 140, + 141, + 142, + 143, + 144, + 145, + 146, + 147, + 148, + 149, + 150, + 151, + 152 + ] + } +] \ No newline at end of file diff --git a/checkpoints/080000/training_state/optimizer_state.safetensors b/checkpoints/080000/training_state/optimizer_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e60c5b935fe879f2378c1c91e0f12ab1f4419b17 --- /dev/null +++ b/checkpoints/080000/training_state/optimizer_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:355c6615f9bef4deed257ac4c2c945ee000abcbd2f113531302bd469eaf6c7b4 +size 412530932 diff --git a/checkpoints/080000/training_state/rng_state.safetensors b/checkpoints/080000/training_state/rng_state.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..9230826466b375f13b1be079ab32cf4eb5495171 --- /dev/null +++ b/checkpoints/080000/training_state/rng_state.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0142d7d9eb8a964b99d588bb069647611bd78f931efbb4cf55e9e4ad7db96083 +size 15708 diff --git a/checkpoints/080000/training_state/training_step.json b/checkpoints/080000/training_state/training_step.json new file mode 100644 index 0000000000000000000000000000000000000000..384eace4ecc2a6cba352aa7cf27f04405b7319c3 --- /dev/null +++ b/checkpoints/080000/training_state/training_step.json @@ -0,0 +1,3 @@ +{ + "step": 80000 +} \ No newline at end of file