update models finetuned on pick_place_banana

.gitignore

@@ -13,4 +13,5 @@
 **/*epoch_16_*
 **/*epoch_17_*
 **/*epoch_18_*
-**/*epoch_19_*
+**/*epoch_19_*
+**/visualization

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_Fanuc_Epoch5/saved_epoch_10_step_540/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_Fanuc_Epoch5/saved_epoch_10_step_540/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f704bfadc60d96f6bcafe0e9b80dadf3c659c0d4665b0ef58c85427a0e5bc95
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_Fanuc_Epoch5/saved_epoch_15_step_810/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_Fanuc_Epoch5/saved_epoch_15_step_810/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec9b4c23940d2eb1175cbda54fe810af3d39e7778dd660470145429ee23a5d43
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_Fanuc_Epoch5/saved_epoch_20_step_1080/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_Fanuc_Epoch5/saved_epoch_20_step_1080/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2385b0eae0c771291de9c58a4c979049ff289e7d0d1956e84fa330d8cd156f42
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_Fanuc_Epoch5/saved_epoch_3_step_162/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_Fanuc_Epoch5/saved_epoch_3_step_162/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4cbb4cf064dc77dd37317cc524552b611841b4fcbf314c81f7a0db4f04043d1a
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_Fanuc_Epoch5/saved_epoch_5_step_270/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_Fanuc_Epoch5/saved_epoch_5_step_270/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4f8dac4011db9dcd93bf07fd3766eab0e3c700bde2028a11320096283cc26e0
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_oxe_Epoch10/saved_epoch_10_step_540/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_oxe_Epoch10/saved_epoch_10_step_540/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72fae29554f32cba3ede74177ef8c837e91d345546d305bd12aa3cbb0e3aa28c
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_oxe_Epoch10/saved_epoch_15_step_810/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_oxe_Epoch10/saved_epoch_15_step_810/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6de82933ce3faadbf3c06e9f5f982f0680a1eb9ca3a7010d4e2a9fc34969b537
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_oxe_Epoch10/saved_epoch_20_step_1080/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_oxe_Epoch10/saved_epoch_20_step_1080/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba9d7d947eb6d07c2d4f0e8f78be60499185edb0d12304168d30e37ed739608c
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_oxe_Epoch10/saved_epoch_3_step_162/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_oxe_Epoch10/saved_epoch_3_step_162/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d4369ddbdc36d530dbcfde5effa280813f70c91e28efb7c13b6f68b75f3e893
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_oxe_Epoch10/saved_epoch_5_step_270/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_oxe_Epoch10/saved_epoch_5_step_270/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf1ddaaa98cea087ae1064951334de1f7359f7dbc5500c841e23816509cae240
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc150_Epoch5/saved_epoch_10_step_540/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc150_Epoch5/saved_epoch_10_step_540/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:715b0692f14f95bb83bae997a2248c39fd06eff3aff862d187f790efd65a5831
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc150_Epoch5/saved_epoch_15_step_810/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc150_Epoch5/saved_epoch_15_step_810/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a28840fd46bb84673f0e5d979d3d84259e4c22c3d4678ae4a19596df90d1b62d
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc150_Epoch5/saved_epoch_20_step_1080/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc150_Epoch5/saved_epoch_20_step_1080/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d466d4900e79127f030c43509de8b74e576f672e01ef770f40f9ef317c4af72e
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc150_Epoch5/saved_epoch_3_step_162/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc150_Epoch5/saved_epoch_3_step_162/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c7cd9c5936e534314d9b76209762add3d6fc772b7845e75241b7f45be035336
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc150_Epoch5/saved_epoch_5_step_270/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc150_Epoch5/saved_epoch_5_step_270/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:253c956b97671f3e9184bc32ba461949f6c4b9dd1620f450af9ead795eea5ddd
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc50_Epoch5/saved_epoch_10_step_540/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc50_Epoch5/saved_epoch_10_step_540/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6cd82e22f6b606aeff52633b8a4350818d076cfebb6625d1499dbd9cb53091e
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc50_Epoch5/saved_epoch_15_step_810/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc50_Epoch5/saved_epoch_15_step_810/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:305a0c7df6ce16e9acedfaea57a6c1c5b490bfb00077e0d39b5926a74e46fc48
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc50_Epoch5/saved_epoch_20_step_1080/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc50_Epoch5/saved_epoch_20_step_1080/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af5e68c0e60f2fe82e21907feabfcbc27eae4c1c79196cb65c3c03ed8c965913
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc50_Epoch5/saved_epoch_3_step_162/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc50_Epoch5/saved_epoch_3_step_162/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dfc74649f99f330c39d68b27bdd82354b51021ffb171f85ad7f537e974919901
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc50_Epoch5/saved_epoch_5_step_270/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: false
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredFalse_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_rtx_fanuc50_Epoch5/saved_epoch_5_step_270/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf3d796875e99441e61ea613d024fe9faae9f8eb9f8df8186092c8caca1c27fb
+size 363796998

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredTrue_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_Fanuc_Epoch5/saved_epoch_10_step_540/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: true
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredTrue_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_Fanuc_Epoch5/saved_epoch_10_step_540/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8708a8e684adfa47387ef41d6520e702f8258124c3e4435df22744c7db16636b
+size 364193934

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredTrue_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_Fanuc_Epoch5/saved_epoch_15_step_810/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: true
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredTrue_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_Fanuc_Epoch5/saved_epoch_15_step_810/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48405e90b5f6a3379f59f8c266b4397e696b570204c1e836741bf42db4d7707e
+size 364193934

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredTrue_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_Fanuc_Epoch5/saved_epoch_20_step_1080/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: true
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredTrue_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_Fanuc_Epoch5/saved_epoch_20_step_1080/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b71b25e135213f3d299b6679df2cb86f0b890ba55b75fb38fa4453ef43574ceb
+size 364193934

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredTrue_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_Fanuc_Epoch5/saved_epoch_3_step_162/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: true
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredTrue_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_Fanuc_Epoch5/saved_epoch_3_step_162/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:814c4f08fbba8a02082225b71d7b3f21fda04426d5f91e039644739ebfd926f1
+size 364193934

moto_gpt_finetuned_on_pick_place_banana_dataset/data_pick_place_banana_dataset-model_actPredTrueDiscrete_motionPredTrue_visionMaeLarge_seq2_chunk9_maskProb0.5-train_lr0.001_bs32_binaryGripperTrue-aug_shiftTrue_resizedCropFalse-resume_from_predLatentOnly_Fanuc_Epoch5/saved_epoch_5_step_270/config.yaml

@@ -0,0 +1,34 @@
+_target_: moto_gpt.src.models.moto_gpt.MotoGPT
+model_lang:
+  _target_: transformers.T5EncoderModel.from_pretrained
+  pretrained_model_name_or_path: t5-base
+model_vision:
+  _target_: moto_gpt.src.models.mae_model.MaeEncoder
+  use_obs_feature: true
+  pretrained_model_name_or_path: facebook/vit-mae-large
+model_causal_transformer:
+  _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Model
+  config:
+    _target_: moto_gpt.src.models.trajectory_gpt2.GPT2Config
+    vocab_size: 1
+    n_embd: 768
+    n_layer: 12
+    n_head: 12
+    activation_function: relu
+    dropout: 0.1
+    n_positions: 1024
+act_dim: 7
+hidden_size: 768
+sequence_length: 2
+chunk_size: 9
+per_latent_motion_len: 8
+latent_motion_codebook_size: 128
+latent_motion_pred: true
+act_pred: true
+img_feat_dim: 1024
+patch_feat_dim: 1024
+lang_feat_dim: 768
+mask_latent_motion_probability: 0.5
+freeze_lang: true
+freeze_vision: true
+pred_discrete_arm_action: true