NgtjddlmZddlmZddlZddlmZddlmZddlmZm Z Gdd eZ dS) ) annotations)IterableN)Tensor)SentenceTransformer)BatchHardTripletLoss$BatchHardTripletLossDistanceFunctioncVeZdZejfdfd Zdd Zdd ZeddZ xZ S)BatchHardSoftMarginTripletLossmodelrreturnNonecft|||_||_dS)a BatchHardSoftMarginTripletLoss takes a batch with (sentence, label) pairs and computes the loss for all possible, valid triplets, i.e., anchor and positive must have the same label, anchor and negative a different label. The labels must be integers, with same label indicating sentences from the same class. Your train dataset must contain at least 2 examples per label class. This soft-margin variant does not require setting a margin. Args: model: SentenceTransformer model distance_metric: Function that returns a distance between two embeddings. The class SiameseDistanceMetric contains pre-defined metrics that can be used. Definitions: :Easy triplets: Triplets which have a loss of 0 because ``distance(anchor, positive) + margin < distance(anchor, negative)``. :Hard triplets: Triplets where the negative is closer to the anchor than the positive, i.e., ``distance(anchor, negative) < distance(anchor, positive)``. :Semi-hard triplets: Triplets where the negative is not closer to the anchor than the positive, but which still have a positive loss, i.e., ``distance(anchor, positive) < distance(anchor, negative) + margin``. References: * Source: https://github.com/NegatioN/OnlineMiningTripletLoss/blob/master/online_triplet_loss/losses.py * Paper: In Defense of the Triplet Loss for Person Re-Identification, https://arxiv.org/abs/1703.07737 * Blog post: https://omoindrot.github.io/triplet-loss Requirements: 1. Each sentence must be labeled with a class. 2. Your dataset must contain at least 2 examples per labels class. 3. Your dataset should contain hard positives and negatives. Inputs: +------------------+--------+ | Texts | Labels | +==================+========+ | single sentences | class | +------------------+--------+ Recommendations: - Use ``BatchSamplers.GROUP_BY_LABEL`` (:class:`docs `) to ensure that each batch contains 2+ examples per label class. Relations: * :class:`BatchHardTripletLoss` uses a user-specified margin, while this loss does not require setting a margin. Example: :: from sentence_transformers import SentenceTransformer, SentenceTransformerTrainer, losses from datasets import Dataset model = SentenceTransformer("microsoft/mpnet-base") # E.g. 0: sports, 1: economy, 2: politics train_dataset = Dataset.from_dict({ "sentence": [ "He played a great game.", "The stock is up 20%", "They won 2-1.", "The last goal was amazing.", "They all voted against the bill.", ], "label": [0, 1, 0, 0, 2], }) loss = losses.BatchHardSoftMarginTripletLoss(model) trainer = SentenceTransformerTrainer( model=model, train_dataset=train_dataset, loss=loss, ) trainer.train() N)super__init__sentence_embedderdistance_metric)selfr r __class__s w/var/www/html/ai-engine/env/lib/python3.11/site-packages/sentence_transformers/losses/BatchHardSoftMarginTripletLoss.pyrz'BatchHardSoftMarginTripletLoss.__init__s4T !&.sentence_featuresIterable[dict[str, Tensor]]labelsrcp||dd}|||S)Nrsentence_embedding)r#batch_hard_triplet_soft_margin_loss)rrrreps rforwardz&BatchHardSoftMarginTripletLoss.forward\s7$$%6q%9::;OP77DDDr embeddingsc||}tj|}||z}|dd\}}tj|}|dd\} }|| d|z zz} | dd\} }tjtj || z } | } | S)a6Build the triplet loss over a batch of embeddings. For each anchor, we get the hardest positive and hardest negative to form a triplet. Args: labels: labels of the batch, of size (batch_size,) embeddings: tensor of shape (batch_size, embed_dim) squared: Boolean. If true, output is the pairwise squared euclidean distance matrix. If false, output is the pairwise euclidean distance matrix. Returns: Label_Sentence_Triplet: scalar tensor containing the triplet loss rT)keepdimg?) rr get_anchor_positive_triplet_maskfloatmax get_anchor_negative_triplet_maskmintorchlog1pexpmean)rrr pairwise_distmask_anchor_positiveanchor_positive_disthardest_positive_dist_mask_anchor_negativemax_anchor_negative_distanchor_negative_disthardest_negative_disttl triplet_losss rrzBBatchHardSoftMarginTripletLoss.batch_hard_triplet_soft_margin_lossbs,,Z88  4TU[\\bbdd 4mC$8#;#;At#;#L#L q 4TU[\\bbdd'4&7&74&7&H&H# !,/G3QeKe/ff$8#;#;At#;#L#L q [#8;P#PQQ R Rwwyy rstrcdS)Na @misc{hermans2017defense, title={In Defense of the Triplet Loss for Person Re-Identification}, author={Alexander Hermans and Lucas Beyer and Bastian Leibe}, year={2017}, eprint={1703.07737}, archivePrefix={arXiv}, primaryClass={cs.CV} } )rs rcitationz'BatchHardSoftMarginTripletLoss.citations   r)r rr r)rrrrr r)rrr rr r)r r7) __name__ __module__ __qualname__r eucledian_distancerrrpropertyr: __classcell__)rs@rr r s:^:qL/L/L/L/L/L/L/\EEEE ))))V   X     rr ) __future__rcollections.abcrr(r)sentence_transformers.SentenceTransformerrrr r r9rrrDs""""""$$$$$$ IIIIII\\\\\\\\KKKKK%9KKKKKr