CreateFineTuneRequest class abstract

CreateFineTuneRequest

Properties:

  • trainingFile - The ID of an uploaded file that contains training data. See upload file for how to upload a file. Your dataset must be formatted as a JSONL file, where each training example is a JSON object with the keys "prompt" and "completion". Additionally, you must upload your file with the purpose fine-tune. See the fine-tuning guide for more details.
  • validationFile - The ID of an uploaded file that contains validation data. If you provide this file, the data is used to generate validation metrics periodically during fine-tuning. These metrics can be viewed in the fine-tuning results file. Your train and validation data should be mutually exclusive. Your dataset must be formatted as a JSONL file, where each validation example is a JSON object with the keys "prompt" and "completion". Additionally, you must upload your file with the purpose fine-tune. See the fine-tuning guide for more details.
  • model - The name of the base model to fine-tune. You can select one of "ada", "babbage", "curie", "davinci", or a fine-tuned model created after 2022-04-21. To learn more about these models, see the Models documentation.
  • nEpochs - The number of epochs to train the model for. An epoch refers to one full cycle through the training dataset.
  • batchSize - The batch size to use for training. The batch size is the number of training examples used to train a single forward and backward pass. By default, the batch size will be dynamically configured to be ~0.2% of the number of examples in the training set, capped at 256 - in general, we've found that larger batch sizes tend to work better for larger datasets.
  • learningRateMultiplier - The learning rate multiplier to use for training. The fine-tuning learning rate is the original learning rate used for pretraining multiplied by this value. By default, the learning rate multiplier is the 0.05, 0.1, or 0.2 depending on final batch_size (larger learning rates tend to perform better with larger batch sizes). We recommend experimenting with values in the range 0.02 to 0.2 to see what produces the best results.
  • promptLossWeight - The weight to use for loss on the prompt tokens. This controls how much the model tries to learn to generate the prompt (as compared to the completion which always has a weight of 1.0), and can add a stabilizing effect to training when completions are short. If prompts are extremely long (relative to completions), it may make sense to reduce this weight so as to avoid over-prioritizing learning the prompt.
  • computeClassificationMetrics - If set, we calculate classification-specific metrics such as accuracy and F-1 score using the validation set at the end of every epoch. These metrics can be viewed in the results file. In order to compute classification metrics, you must provide a validation_file. Additionally, you must specify classification_n_classes for multiclass classification or classification_positive_class for binary classification.
  • classificationNClasses - The number of classes in a classification task. This parameter is required for multiclass classification.
  • classificationPositiveClass - The positive class in binary classification. This parameter is needed to generate precision, recall, and F1 metrics when doing binary classification.
  • classificationBetas - If this is provided, we calculate F-beta scores at the specified beta values. The F-beta score is a generalization of F-1 score. This is only used for binary classification. With a beta of 1 (i.e. the F-1 score), precision and recall are given the same weight. A larger beta score puts more weight on recall and less on precision. A smaller beta score puts more weight on precision and less on recall.
  • suffix - A string of up to 40 characters that will be added to your fine-tuned model name. For example, a suffix of "custom-model-name" would produce a model name like ada:ft-your-org:custom-model-name-2022-02-15-04-21-04.
Annotations
  • @BuiltValue()

Constructors

CreateFineTuneRequest([void updates(CreateFineTuneRequestBuilder b)])
factory

Properties

batchSize → int?
The batch size to use for training. The batch size is the number of training examples used to train a single forward and backward pass. By default, the batch size will be dynamically configured to be ~0.2% of the number of examples in the training set, capped at 256 - in general, we've found that larger batch sizes tend to work better for larger datasets.
no setter
classificationBetas → BuiltList<num>?
If this is provided, we calculate F-beta scores at the specified beta values. The F-beta score is a generalization of F-1 score. This is only used for binary classification. With a beta of 1 (i.e. the F-1 score), precision and recall are given the same weight. A larger beta score puts more weight on recall and less on precision. A smaller beta score puts more weight on precision and less on recall.
no setter
classificationNClasses → int?
The number of classes in a classification task. This parameter is required for multiclass classification.
no setter
classificationPositiveClass → String?
The positive class in binary classification. This parameter is needed to generate precision, recall, and F1 metrics when doing binary classification.
no setter
computeClassificationMetrics → bool?
If set, we calculate classification-specific metrics such as accuracy and F-1 score using the validation set at the end of every epoch. These metrics can be viewed in the results file. In order to compute classification metrics, you must provide a validation_file. Additionally, you must specify classification_n_classes for multiclass classification or classification_positive_class for binary classification.
no setter
hashCode → int
The hash code for this object.
no setterinherited
learningRateMultiplier → num?
The learning rate multiplier to use for training. The fine-tuning learning rate is the original learning rate used for pretraining multiplied by this value. By default, the learning rate multiplier is the 0.05, 0.1, or 0.2 depending on final batch_size (larger learning rates tend to perform better with larger batch sizes). We recommend experimenting with values in the range 0.02 to 0.2 to see what produces the best results.
no setter
model → String?
The name of the base model to fine-tune. You can select one of "ada", "babbage", "curie", "davinci", or a fine-tuned model created after 2022-04-21. To learn more about these models, see the Models documentation.
no setter
nEpochs → int?
The number of epochs to train the model for. An epoch refers to one full cycle through the training dataset.
no setter
promptLossWeight → num?
The weight to use for loss on the prompt tokens. This controls how much the model tries to learn to generate the prompt (as compared to the completion which always has a weight of 1.0), and can add a stabilizing effect to training when completions are short. If prompts are extremely long (relative to completions), it may make sense to reduce this weight so as to avoid over-prioritizing learning the prompt.
no setter
runtimeType → Type
A representation of the runtime type of the object.
no setterinherited
suffix → String?
A string of up to 40 characters that will be added to your fine-tuned model name. For example, a suffix of "custom-model-name" would produce a model name like ada:ft-your-org:custom-model-name-2022-02-15-04-21-04.
no setter
trainingFile → String
The ID of an uploaded file that contains training data. See upload file for how to upload a file. Your dataset must be formatted as a JSONL file, where each training example is a JSON object with the keys "prompt" and "completion". Additionally, you must upload your file with the purpose fine-tune. See the fine-tuning guide for more details.
no setter
validationFile → String?
The ID of an uploaded file that contains validation data. If you provide this file, the data is used to generate validation metrics periodically during fine-tuning. These metrics can be viewed in the fine-tuning results file. Your train and validation data should be mutually exclusive. Your dataset must be formatted as a JSONL file, where each validation example is a JSON object with the keys "prompt" and "completion". Additionally, you must upload your file with the purpose fine-tune. See the fine-tuning guide for more details.
no setter

Methods

noSuchMethod(Invocation invocation) → dynamic
Invoked when a nonexistent method or property is accessed.
inherited
rebuild(dynamic updates(CreateFineTuneRequestBuilder)) → CreateFineTuneRequest
Rebuilds the instance.
inherited
toBuilder() → CreateFineTuneRequestBuilder
Converts the instance to a builder B.
inherited
toString() → String
A string representation of this object.
inherited

Operators

operator ==(Object other) → bool
The equality operator.
inherited

Static Properties

serializer → Serializer<CreateFineTuneRequest>
no setter