README.md

Implementation of some unbalanced loss for NLP task like focal_loss, dice_loss, DSC Loss, GHM Loss et.al and adversarial training like FGM, FGSM, PGD, FreeAT.

Loss Summary

Here is a loss implementation repository included unbalanced loss

Loss Name	paper	Notes
Weighted CE Loss	UNet Architectures in Multiplanar Volumetric Segmentation -- Validated on Three Knee MRI Cohorts
Focal Loss	Focal Loss for Dense Object Detection
Dice Loss	V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation
DSC Loss	Dice Loss for Data-imbalanced NLP Tasks
GHM Loss	Gradient Harmonized Single-stage Detector
Label Smoothing	When Does Label Smoothing Help?

How to use?

You can find all the loss usage information in test_loss.py.

Here is a simple demo of usage:

import torch
from unbalanced_loss.focal_loss import MultiFocalLoss

batch_size, num_class = 64, 10
Loss_Func = MultiFocalLoss(num_class=num_class, gamma=2.0, reduction='mean')

logits = torch.rand(batch_size, num_class, requires_grad=True)  # (batch_size, num_classes)
targets = torch.randint(0, num_class, size=(batch_size,))  # (batch_size, )

loss = Loss_Func(logits, targets)
loss.backward()

Adversarial Training Summary

Here is a Summary of Adversarial Training implementation.
you can find more details in adversarial_training/README.md

Adversarial Training	paper	Notes
FGM	Fast Gradient Method
FGSM	Fast Gradient Sign Method
PGD	Towards Deep Learning Models Resistant to Adversarial Attacks
FreeAT	Free Adversarial Training
FreeLB	Free Large Batch Adversarial Training

How to use?

You can find a simple demo for bert classification in test_bert.py.

Here is a simple demo of usage:
You just need to rewrite train function according to input for your model in file PGD.py, then you can use adversarial training like below.

import transformers
from model import bert_classification
from adversarial_training.PGD import PGD

batch_size, num_class = 64, 10
# model = your_model()
model = bert_classification()
AT_Model = PGD(model)
optimizer = transformers.AdamW(model.parameters(), lr=0.001)

# rewrite your train function in pgd.py
outputs, loss = AT_Model.train_bert(token, segment, mask, label, optimizer)

Adversarial Training Results

here are some results tested on THNews classification task based on bert.
you can find run the code as below:

cd scripts
sh run_at.sh

Adversarial Training	Time Cost(s/epoch )	best_acc
Normal(not add attack)	23.77	0.773
FGSM	45.95	0.7936
FGM	47.28	0.8008
PGD(k=3)	87.50	0.7963
FreeAT(k=3)	93.26	0.7896