distilbert-base-multilingual-cased-fine-ptbr

This model is a fine-tuned version of distilbert-base-multilingual-cased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7726
• Accuracy: EvaluationModule(name: "accuracy", module_type: "metric", features: {'predictions': Value(dtype='int32', id=None), 'references': Value(dtype='int32', id=None)}, usage: """ Args: predictions (list of int): Predicted labels. references (list of int): Ground truth labels. normalize (boolean): If set to False, returns the number of correctly classified samples. Otherwise, returns the fraction of correctly classified samples. Defaults to True. sample_weight (list of float): Sample weights Defaults to None.

Returns: accuracy (float or int): Accuracy score. Minimum possible value is 0. Maximum possible value is 1.0, or the number of examples input, if normalize is set to True.. A higher score means higher accuracy.

Examples:

Example 1-A simple example
    >>> accuracy_metric = evaluate.load("accuracy")
    >>> results = accuracy_metric.compute(references=[0, 1, 2, 0, 1, 2], predictions=[0, 1, 1, 2, 1, 0])
    >>> print(results)
    {'accuracy': 0.5}

Example 2-The same as Example 1, except with `normalize` set to `False`.
    >>> accuracy_metric = evaluate.load("accuracy")
    >>> results = accuracy_metric.compute(references=[0, 1, 2, 0, 1, 2], predictions=[0, 1, 1, 2, 1, 0], normalize=False)
    >>> print(results)
    {'accuracy': 3.0}

Example 3-The same as Example 1, except with `sample_weight` set.
    >>> accuracy_metric = evaluate.load("accuracy")
    >>> results = accuracy_metric.compute(references=[0, 1, 2, 0, 1, 2], predictions=[0, 1, 1, 2, 1, 0], sample_weight=[0.5, 2, 0.7, 0.5, 9, 0.4])
    >>> print(results)
    {'accuracy': 0.8778625954198473}

""", stored examples: 0)

• F1: EvaluationModule(name: "f1", module_type: "metric", features: {'predictions': Value(dtype='int32', id=None), 'references': Value(dtype='int32', id=None)}, usage: """ Args: predictions (list of int): Predicted labels. references (list of int): Ground truth labels. labels (list of int): The set of labels to include when average is not set to 'binary', and the order of the labels if average is None. Labels present in the data can be excluded, for example to calculate a multiclass average ignoring a majority negative class. Labels not present in the data will result in 0 components in a macro average. For multilabel targets, labels are column indices. By default, all labels in predictions and references are used in sorted order. Defaults to None. pos_label (int): The class to be considered the positive class, in the case where average is set to binary. Defaults to 1. average (string): This parameter is required for multiclass/multilabel targets. If set to None, the scores for each class are returned. Otherwise, this determines the type of averaging performed on the data. Defaults to 'binary'.

    - 'binary': Only report results for the class specified by `pos_label`. This is applicable only if the classes found in `predictions` and `references` are binary.
    - 'micro': Calculate metrics globally by counting the total true positives, false negatives and false positives.
    - 'macro': Calculate metrics for each label, and find their unweighted mean. This does not take label imbalance into account.
    - 'weighted': Calculate metrics for each label, and find their average weighted by support (the number of true instances for each label). This alters `'macro'` to account for label imbalance. This option can result in an F-score that is not between precision and recall.
    - 'samples': Calculate metrics for each instance, and find their average (only meaningful for multilabel classification).
  

  sample_weight (list of float): Sample weights Defaults to None.

Returns: f1 (float or array of float): F1 score or list of f1 scores, depending on the value passed to average. Minimum possible value is 0. Maximum possible value is 1. Higher f1 scores are better.

Examples:

Example 1-A simple binary example
    >>> f1_metric = evaluate.load("f1")
    >>> results = f1_metric.compute(references=[0, 1, 0, 1, 0], predictions=[0, 0, 1, 1, 0])
    >>> print(results)
    {'f1': 0.5}

Example 2-The same simple binary example as in Example 1, but with `pos_label` set to `0`.
    >>> f1_metric = evaluate.load("f1")
    >>> results = f1_metric.compute(references=[0, 1, 0, 1, 0], predictions=[0, 0, 1, 1, 0], pos_label=0)
    >>> print(round(results['f1'], 2))
    0.67

Example 3-The same simple binary example as in Example 1, but with `sample_weight` included.
    >>> f1_metric = evaluate.load("f1")
    >>> results = f1_metric.compute(references=[0, 1, 0, 1, 0], predictions=[0, 0, 1, 1, 0], sample_weight=[0.9, 0.5, 3.9, 1.2, 0.3])
    >>> print(round(results['f1'], 2))
    0.35

Example 4-A multiclass example, with different values for the `average` input.
    >>> predictions = [0, 2, 1, 0, 0, 1]
    >>> references = [0, 1, 2, 0, 1, 2]
    >>> results = f1_metric.compute(predictions=predictions, references=references, average="macro")
    >>> print(round(results['f1'], 2))
    0.27
    >>> results = f1_metric.compute(predictions=predictions, references=references, average="micro")
    >>> print(round(results['f1'], 2))
    0.33
    >>> results = f1_metric.compute(predictions=predictions, references=references, average="weighted")
    >>> print(round(results['f1'], 2))
    0.27
    >>> results = f1_metric.compute(predictions=predictions, references=references, average=None)
    >>> print(results)
    {'f1': array([0.8, 0. , 0. ])}

Example 5-A multi-label example
    >>> f1_metric = evaluate.load("f1", "multilabel")
    >>> results = f1_metric.compute(predictions=[[0, 1, 1], [1, 1, 0]], references=[[0, 1, 1], [0, 1, 0]], average="macro")
    >>> print(round(results['f1'], 2))
    0.67

""", stored examples: 0)

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 3

Training results

Framework versions

• Transformers 4.37.1
• Pytorch 2.1.2+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1