API Reference
- Description
- tft_init_controller
- tft_start_controller
- tft_destroy_controller
- tft_init_processor
- tft_start_processor
- tft_destroy_processor
- tft_start_updating_os
- tft_start_copy_os
- tft_end_updating_os
- tft_set_optimizer_replica
- tft_exception_handler
- tft_set_step_args
- tft_register_rename_handler
- tft_register_save_ckpt_handler
- tft_register_exit_handler
- tft_register_stop_handler
- tft_register_clean_handler
- tft_register_rebuild_group_handler
- tft_register_repair_handler
- tft_register_rollback_handler
- tft_register_stream_sync_handler
- tft_register_zit_upgrade_rollback_handler
- tft_register_zit_upgrade_repair_handler
- tft_register_zit_upgrade_rebuild_handler
- tft_register_zit_downgrade_rebuild_handler
- tft_report_error
- tft_wait_next_action
- tft_get_repair_step
- tft_get_repair_type
- tft_is_reboot_node
- tft_get_reboot_type
- tft_reset_limit_step
- tft_set_dp_group_info
- tft_report_load_ckpt_step
- tft_register_decrypt_handler
- tft_notify_controller_dump
- tft_notify_controller_stop_train
- tft_notify_controller_on_global_rank
- tft_notify_controller_prepare_action
- tft_notify_controller_change_strategy
- tft_register_mindx_callback
- tft_query_high_availability_switch
- tft_can_do_uce_repair
- tft_set_update_start_time
- tft_set_update_end_time
- tft_pause_train
- OptimizerType
- Action
- ReportState
- RepairType
Parent topic: Fault Recovery Acceleration