DeepFellow DOCS

FineTuningJob

A fine-tuning job represents a request to fine-tune a model.

A fine-tuning job represents a request to fine-tune a model.

PropertyTypeRequiredDescription
idstringYesThe job identifier, which can be referenced in the API endpoints.
project_idstringYesThe project identifier this job belongs to.
objectstringNoThe object type, which is always 'fine_tuning.job'. Always fine_tuning.job. Default: "fine_tuning.job".
modelstringYesThe base model being fine-tuned.
statusserver__models__fine_tuning__StatusYesThe current status of the fine-tuning job.
training_filestringYesThe file ID used for training.
validation_filestring | nullNoThe file ID used for validation.
hyperparametersHyperparametersNoThe hyperparameters used for the fine-tuning job. Default: {"n_epochs":"auto","batch_size":"auto","learning_rate_multiplier":"auto"}.
methodMethod-Output | nullNoThe method used for fine-tuning.
created_atintegerNoThe Unix timestamp (in seconds) for when the job was created. Default: 1786012207.
finished_atinteger | nullNoThe Unix timestamp (in seconds) for when the job finished.
fine_tuned_modelstring | nullNoThe name of the fine-tuned model that is being created. Null if not yet available.
suffixstring | nullNoA string added to the fine-tuned model name.
errorError | nullNoError information if the job failed.
trained_tokensinteger | nullNoThe total number of billable tokens processed by this fine-tuning job. The value will be null if the fine-tuning job is still running.
worker_job_idstring | nullNoThe job ID assigned by the fine-tuning worker service.
worker_idstring | nullNoThe ID of the fine_tuning_workers document for the worker processing this job.
worker_sync_failuresintegerNoNumber of consecutive poll failures against the assigned worker. Default: 0.

We use cookies on our website. We use them to ensure proper functioning of the site and, if you agree, for purposes such as analytics, marketing, and targeting ads.