Returns a list of non stopwords for the sentence, obtained after cleaning ponctuation and spaces

Parse the program of the given refexp and return the objects, along with the list of tokens they correspond to from the scene necessary to compute the

Print a progress bar during training. :param i: index of current iteration/epoch. :param max: max number of iterations/epochs. :param

Extract the bounding box of a given region as a list

This function removes all ponctuation.

Return a condensed report of the results as a dict of dict. report[k][cat] is the recall@k for the given category

This function disables printing when not in master process

PyTorch Implementation for tf.nn.sigmoid_cross_entropy_with_logits: max(x, 0) - x * z + log(1 + exp(-abs(x))) in https://www.

Loss used in RetinaNet for dense detection: https://arxiv.org/abs/1708.02002. Args: inputs: A float tensor of arbitrary shape.

This function simplifies doubled or more complex punctuation. The exception is '...'.

Apply exponential moving average update. The weights are updated in-place as follow: w_ema = w_ema * decay + (1 - decay) * w Args:

Accepts a list of bounding boxes in pascal format (xmin,ymin,xmax,ymax) Returns the list of boxes in coco format (xmin,ymin, width, height)

Convert [x1 y1 x2 y2] box format to [x y w h] format.

Args: index (int): Index Returns: tuple: Tuple (image, target). target is the object returned by ``coco.loadA

Args: index (int): Index Returns: tuple: Tuple (image, target). target is the object returned by ``coco.load