Model

Start Training

Start model training. Training duration is typically 2-4 hours. The associated dataset must have a status of 'completed' before training can begin.

Default Training Parameters: If parameters are not provided, defaults are automatically selected:

  • rank: 128
  • learning_rate: 0.0001
  • lr_scheduler: "constant_with_warmup"
  • total_training_steps: 750
  • checkpoint_interval: 500

Constraints: Advanced training parameters are only supported when the model's training_mode is set to 'expert'.

post/tailored-gen/models/{model_id}/start_training

Request

  • Base URL: https://engine.prod.bria-api.com/v1
  • URL: https://engine.prod.bria-api.com/v1/tailored-gen/models/{model_id}/start_training
  • Auth: none declared. Required headers: api_token

Path parameters

model_idinteger required

Headers

api_tokenstring required

Request body

lr_scheduler'linear' | 'cosine' | 'cosine_with_restarts' | 'polynomial' | 'constant' | 'constant_with_warmup'

The learning rate scheduler.

learning_ratenumber float

The initial learning rate. Default is 0.0001.

total_training_stepsinteger

Total number of training optimization steps. Default is 750.

rankinteger

The rank (latent dimension). Default is 128.

checkpoint_intervalinteger

Interval (in steps) at which checkpoints are saved. Default is 500.

lr_warmup_stepsinteger

Number of warmup steps. Relevant only when lr_scheduler is constant_with_warmup.

Response

Training successfully started

idinteger
namestring
descriptionstring
status'InProgress'
dataset_idinteger
training_modestring
project_idinteger
created_atstring date-time
updated_atstring date-time
active_model_versionstring
checkpoint_stepsstring[]

Changes

No recorded changes to this operation across all 1 revision of this API.