"""
Author: Yonglong Tian (yonglong@mit.edu)
Date: May 07, 2020
"""
from __future__ import print_function

import torch
import torch.nn as nn
import numpy as np


class SupConLoss(nn.Module):
    """Supervised Contrastive Learning: https://arxiv.org/pdf/2004.11362.pdf.
    It also supports the unsupervised contrastive loss in SimCLR"""
    def __init__(self, temperature=0.07, contrast_mode='all',
                 base_temperature=0.07):
        super(SupConLoss, self).__init__()
        self.temperature = temperature
        self.contrast_mode = contrast_mode
        self.base_temperature = base_temperature

    def forward(self, features, labels=None, mask=None):
        """Compute loss for model. If both `labels` and `mask` are None,
        it degenerates to SimCLR unsupervised loss:
        https://arxiv.org/pdf/2002.05709.pdf

        Args:
            features: hidden vector of shape [bsz, n_views, ...].
            labels: ground truth of shape [bsz].
            mask: contrastive mask of shape [bsz, bsz], mask_{i,j}=1 if sample j
                has the same class as sample i. Can be asymmetric.
        Returns:
            A loss scalar.
        """
        device = (torch.device('cuda')
                  if features.is_cuda
                  else torch.device('cpu'))

        if len(features.shape) < 3:
            raise ValueError('`features` needs to be [bsz, n_views, ...],'
                             'at least 3 dimensions are required')
        if len(features.shape) > 3:
            features = features.view(features.shape[0], features.shape[1], -1)

        batch_size = features.shape[0]
        if labels is not None and mask is not None:
            raise ValueError('Cannot define both `labels` and `mask`')
        elif labels is None and mask is None:
            mask = torch.eye(batch_size, dtype=torch.float32).to(device)
        elif labels is not None:
            labels = labels.contiguous().view(-1, 1)
            if labels.shape[0] != batch_size:
                raise ValueError('Num of labels does not match num of features')
            mask = torch.eq(labels, labels.T).float().to(device)
        else:
            mask = mask.float().to(device)

        contrast_count = features.shape[1]
        contrast_feature = torch.cat(torch.unbind(features, dim=1), dim=0)
        if self.contrast_mode == 'one':
            anchor_feature = features[:, 0]
            anchor_count = 1
        elif self.contrast_mode == 'all':
            anchor_feature = contrast_feature
            anchor_count = contrast_count
        else:
            raise ValueError('Unknown mode: {}'.format(self.contrast_mode))

        # compute logits
        anchor_dot_contrast = torch.div(
            torch.matmul(anchor_feature, contrast_feature.T),
            self.temperature)
        # for numerical stability
        logits_max, _ = torch.max(anchor_dot_contrast, dim=1, keepdim=True)
        logits = anchor_dot_contrast - logits_max.detach()

        # tile mask
        mask = mask.repeat(anchor_count, contrast_count)
        # mask-out self-contrast cases
        logits_mask = torch.scatter(
            torch.ones_like(mask),
            1,
            torch.arange(batch_size * anchor_count).view(-1, 1).to(device),
            0
        )
        mask = mask * logits_mask

        # compute log_prob
        exp_logits = torch.exp(logits) * logits_mask
        log_prob = logits - torch.log(exp_logits.sum(1, keepdim=True))

        # compute mean of log-likelihood over positive
        mean_log_prob_pos = (mask * log_prob).sum(1) / mask.sum(1)

        # loss
        loss = - (self.temperature / self.base_temperature) * mean_log_prob_pos
        loss = loss.view(anchor_count, batch_size).mean()

        return loss


class SupConLoss1View(nn.Module):
    """Supervised Contrastive Learning: https://arxiv.org/pdf/2004.11362.pdf.
    It also supports the unsupervised contrastive loss in SimCLR"""
    def __init__(self, temperature=0.07, base_temperature=0.07):
        super(SupConLoss1View, self).__init__()
        self.temperature = temperature
        self.base_temperature = base_temperature

    def forward(self, features, labels):
        device = (torch.device('cuda')
                  if features.is_cuda
                  else torch.device('cpu'))

        if len(features.shape) != 2:
            raise ValueError('`features` needs to be [bsz, ndim]')

        batch_size = features.shape[0]
        labels = labels.contiguous().view(-1, 1)
        if labels.shape[0] != batch_size:
            raise ValueError('Num of labels does not match num of features')
        mask = torch.eq(labels, labels.T).float().to(device)

        cross = torch.matmul(features, features.T)

        # frobenius_loss = torch.norm(mask-cross)


        upper_diag = torch.triu_indices(batch_size,batch_size,+1)
        cross_upper = cross[upper_diag[0], upper_diag[1]]
        mask_upper = mask[upper_diag[0], upper_diag[1]]
        # npos = int(mask_upper.sum().item())
        # weight = torch.from_numpy(np.asarray([1-pos, pos], dtype=float)).to(device)
        #return torch.nn.functional.binary_cross_entropy_with_logits(cross_upper, mask_upper)
        #print('mask min-max:', mask.min(), mask.max())
        #print('cross min-max:', cross.min(), cross.max())
        #return torch.norm(cross-mask, p='fro')  # <-- diagonal signal (trivial) should be too strong
        pos_loss = mse(cross_upper, mask_upper, label=1)
        neg_loss = mse(cross_upper, mask_upper, label=0)
        # return frobenius_loss, neg_loss, pos_loss
        #return neg_loss, pos_loss
        # balanced_loss = pos_loss + neg_loss
        # return balanced_loss, neg_loss, pos_loss
        # loss = torch.nn.functional.binary_cross_entropy(cross_upper, mask_upper)
        # return loss, neg_loss, pos_loss
        # n=len(mask_upper)
        # return (1/n)*torch.sqrt(torch.sum((cross_upper-mask_upper)**2)), neg_loss, pos_loss
        return mse(cross_upper, mask_upper), neg_loss, pos_loss


class SupConLoss1ViewCrossEntropy(nn.Module):
    """Supervised Contrastive Learning: https://arxiv.org/pdf/2004.11362.pdf.
    It also supports the unsupervised contrastive loss in SimCLR"""
    def __init__(self):
        super(SupConLoss1ViewCrossEntropy, self).__init__()

    def forward(self, features, labels):
        device = (torch.device('cuda')
                  if features.is_cuda
                  else torch.device('cpu'))

        if len(features.shape) != 2:
            raise ValueError('`features` needs to be [bsz, ndim]')

        batch_size = features.shape[0]
        labels = labels.contiguous().view(-1, 1)
        if labels.shape[0] != batch_size:
            raise ValueError('Num of labels does not match num of features')

        mask = torch.eq(labels, labels.T).float().to(device)
        cross = torch.matmul(features, features.T)

        upper_diag = torch.triu_indices(batch_size,batch_size,+1)
        cross_upper = cross[upper_diag[0], upper_diag[1]]
        mask_upper = mask[upper_diag[0], upper_diag[1]]

        npos = int(mask_upper.sum().item())

        pos_loss = bin_cross_entropy(cross_upper, mask_upper, label=1)
        neg_loss = bin_cross_entropy(cross_upper, mask_upper, label=0)

        loss = bin_cross_entropy(cross_upper, mask_upper)
        return loss, neg_loss, pos_loss


def choice(tensor, k):
    perm = torch.randperm(tensor.size(0))
    idx = perm[:k]
    return tensor[idx]


def mse(input, target, label=None, k=-1):
    if label is not None:
        index = target==label
        input = input[index]
        target = target[index]
    if k>-1:
        input = choice(input, k)
        target = choice(target, k)

    return torch.mean((input - target) ** 2)

    # if label==0:
    #     return torch.mean(input**2)
    # else:
    #     return torch.mean((1-input)**2)
    # index = target==label
    # return torch.mean((input[index] - target[index]) ** 2)


def bin_cross_entropy(input, target, label=None):
    if label is None:
        return torch.nn.functional.binary_cross_entropy_with_logits(input, target)
    index = target == label
    return torch.nn.functional.binary_cross_entropy_with_logits(input[index], target[index])