Skip to content

Trainer - Generic

Category: trainer
Version: 1.0.0
UUID: 07b5408761a7f9e79e850badd149d4a1b600352bf6ec9655d6429937bbea3c84
Execution context: leip_af
Choice Priority: 500.0

Value Parameters

Name Synonyms Type Values Help
The accelerator to use (CPU, GPU, TPU, ...) trainer.accelerator choice cpu
gpu
tpu
ipu
auto
The devices to use trainer.devices multi-choice 0
1
2
3
Dry run trainer.fast_dev_run scalar bool
Gradient clipping algorithm trainer.gradient_clip_algorithm choice norm
value
Refer here for more details.
Gradient clipping value trainer.gradient_clip_value scalar float Refer here for more details.
Log data every N steps trainer.logging_frequency,
trainer.log_every_n_steps
scalar int Refer here for more details.
Maximum number of epochs train.num_epochs scalar int, min: 1
Maximum training time train.max_time scalar string Time before training stops, formatted as a string "DD:HH:MM:SS". A checkpoint will be saved when training stops, even if an epoch didn't complete. Refer (here)[https://lightning.ai/docs/pytorch/stable/common/trainer.html#max-time]
Trainer numerical precision trainer.precision choice 32
16
16-mixed
bf16
Refer here for more details.
The strategy to use for multi-device computation trainer.strategy choice ddp
ddp_notebook
ddp_fork

This component fits into

Name UUID Synonyms
Full Recipe 4511dd... trainer