bridge.training

Subpackages

• bridge.training.tokenizers
  • bridge.training.tokenizers.bert_tokenization
  • bridge.training.tokenizers.config
  • bridge.training.tokenizers.gpt2_tokenization
  • bridge.training.tokenizers.multimodal_tokenizer
  • bridge.training.tokenizers.tokenizer
• bridge.training.post_training
  • bridge.training.post_training.distillation
  • bridge.training.post_training.checkpointing
• bridge.training.utils
  • bridge.training.utils.train_utils
  • bridge.training.utils.packed_seq_utils
  • bridge.training.utils.log_utils
  • bridge.training.utils.pg_utils
  • bridge.training.utils.theoretical_memory_utils
  • bridge.training.utils.omegaconf_utils
  • bridge.training.utils.flop_utils
  • bridge.training.utils.batch_utils
  • bridge.training.utils.checkpoint_utils
  • bridge.training.utils.sig_utils
  • bridge.training.utils.padding_utils
  • bridge.training.utils.moe_token_drop
  • bridge.training.utils.config_utils
  • bridge.training.utils.weight_decay_utils
  • bridge.training.utils.visual_inputs
  • bridge.training.utils.wandb_utils
• bridge.training.mlm_compat
  • bridge.training.mlm_compat.activations
  • bridge.training.mlm_compat.arguments
  • bridge.training.mlm_compat.model

Submodules

• bridge.training.fault_tolerance
• bridge.training.pretrain
• bridge.training.tensor_inspect
• bridge.training.config
• bridge.training.gpt_step
• bridge.training.mixed_precision
• bridge.training.initialize
• bridge.training.inprocess_restart
• bridge.training.flex_dispatcher_backend
• bridge.training.setup
• bridge.training.distill
• bridge.training.nvrx_straggler
• bridge.training.train
• bridge.training.model_load_save
• bridge.training.forward_step_func_types
• bridge.training.llava_step
• bridge.training.losses
• bridge.training.eval
• bridge.training.optim
• bridge.training.vlm_step
• bridge.training.finetune
• bridge.training.comm_overlap
• bridge.training.profiling
• bridge.training.state
• bridge.training.checkpointing