hangX/lerobot
1
0
Fork 0
forked from tangger/lerobot
Code Pull Requests Activity

precommit

Browse Source
This commit is contained in:
mshukor
2025-06-02 20:51:37 +02:00
parent c2d645717a
commit 345bc8b2d0
1 changed files with 7 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
lerobot/common/policies/smolvla/configuration_smolvla.py
View File

@@ -82,10 +82,10 @@ class SmolVLAConfig(PreTrainedConfig):
scheduler_decay_steps: int = 30_000
scheduler_decay_lr: float = 2.5e-6
vlm_model_name: str = "HuggingFaceTB/SmolVLM2-500M-Video-Instruct" # Select the VLM backbone.
load_vlm_weights: bool = False # Set to True in case of training the expert from scratch. True when init from pretrained SmolVLA weights
vlm_model_name: str = "HuggingFaceTB/SmolVLM2-500M-Video-Instruct" # Select the VLM backbone.
load_vlm_weights: bool = False # Set to True in case of training the expert from scratch. True when init from pretrained SmolVLA weights
add_image_special_tokens: bool = False # Whether to use special image tokens around image features.
add_image_special_tokens: bool = False # Whether to use special image tokens around image features.
attention_mode: str = "cross_attn"
@@ -93,10 +93,10 @@ class SmolVLAConfig(PreTrainedConfig):
pad_language_to: str = "longest" # "max_length"
num_expert_layers: int = -1 # Less or equal to 0 is the default where the action expert has the same number of layers of VLM. Otherwise the expert have less layers.
num_vlm_layers: int = 16 # Number of layers used in the VLM (first num_vlm_layers layers)
self_attn_every_n_layers: int = 2 # Interleave SA layers each self_attn_every_n_layers
expert_width_multiplier: float = 0.75 # The action expert hidden size (wrt to the VLM)
num_expert_layers: int = -1 # Less or equal to 0 is the default where the action expert has the same number of layers of VLM. Otherwise the expert have less layers.
num_vlm_layers: int = 16 # Number of layers used in the VLM (first num_vlm_layers layers)
self_attn_every_n_layers: int = 2 # Interleave SA layers each self_attn_every_n_layers
expert_width_multiplier: float = 0.75 # The action expert hidden size (wrt to the VLM)
def __post_init__(self):
super().__post_init__()