ez_transfer

ez_transfer.base_model

class easytransfer.base_model(**kwargs)

build_logits(features, mode)

Given features, this method take care of building graph for train/eval/predict

Parameters:

• features -- either raw text features or numerical features such as input_ids, input_mask ...
• mode -- tf.estimator.ModeKeys.TRAIN | tf.estimator.ModeKeys.EVAL | tf.estimator.ModeKeys.PREDICT

Returns:

logits, labels

Examples:

def build_logits(self, features, mode=None):
    preprocessor = preprocessors.get_preprocessor(self.pretrain_model_name_or_path)
    model = model_zoo.get_pretrained_model(self.pretrain_model_name_or_path)

    dense = layers.Dense(self.num_labels,
                         kernel_initializer=layers.get_initializer(0.02),
                         name='dense')

    input_ids, input_mask, segment_ids, label_ids = preprocessor(features)
    outputs = model([input_ids, input_mask, segment_ids], mode=mode)
    pooled_output = outputs[1]

    logits = dense(pooled_output)
    return logits, label_ids

build_loss(logits, labels)

Build loss function

Parameters:

• logits -- logits returned from build_logits
• labels -- labels returned from build_logits

Returns:

loss

Examples:

def build_loss(self, logits, labels):
    return softmax_cross_entropy(labels, depth=self.config.num_labels, logits=logits)

build_eval_metrics(logits, labels)

Build evaluation metrics

Parameters:

• logits -- logits returned from build_logits
• labels -- labels returned from build_logits

Returns:

metric_dict

Examples:

def build_eval_metrics(self, logits, labels):
    predictions = tf.argmax(logits, axis=-1, output_type=tf.int32)
    info_dict = {
        "predictions": predictions,
        "labels": labels,
    }
    evaluator = PyEvaluator()
    labels = [i for i in range(self.num_labels)]
    metric_dict = evaluator.get_metric_ops(info_dict, labels)
    ret_metrics = evaluator.evaluate(labels)
    tf.summary.scalar("eval accuracy", ret_metrics['py_accuracy'])
    tf.summary.scalar("eval F1 micro score", ret_metrics['py_micro_f1'])
    tf.summary.scalar("eval F1 macro score", ret_metrics['py_macro_f1'])
    return metric_dict

build_predictions(logits)

Build predictions

Parameters:logits -- logits returned from build_logits Returns:predictions

Examples:

def build_predictions(self, output):
    logits, _ = output
    predictions = dict()
    predictions["predictions"] = tf.argmax(logits, axis=-1, output_type=tf.int32)
    return predictions