utils.py

from __future__ import absolute_import
import os
import sys
import errno
import pickle
import shutil
import json
import os.path as osp
import numpy as np
import torch
import scipy
sys.path.append('.')
from sklearn.decomposition import PCA, KernelPCA


def mkdir_if_missing(directory):
    if not osp.exists(directory):
        try:
            os.makedirs(directory)
        except OSError as e:
            if e.errno != errno.EEXIST:
                raise

def pca(teacher, student):
  pca = PCA(n_components=200)
  pca.fit(teacher)
  max_component = pca.components_.T
  teacher = np.dot(teacher, max_component)
  student = np.dot(student, max_component)
  return  student, teacher

def Hamming_Score(y_true, y_pred, torch=True,cate=False):
    """
    torch = true mean y_pred is torch tensor
    if torch=false mean y_pred=numpy
    """
    acc_list = []
    if torch:
        from sklearn.metrics import accuracy_score
    for i in range(len(y_true)):
        if torch:

            summary = y_true[i] == y_pred[i].double()

            num = np.sum(summary.numpy())
        else:
            summary = y_true[i] == y_pred[i]
            num = np.sum(summary)
        tmp_a = num / float(len(y_true[i]))
        acc_list.append(tmp_a)
    #print('mean score from hamming',np.mean(acc_list))
    return np.mean(acc_list)

def hamming_precision(y_true, y_pred,torch = True, cate = True):
    acc_list = []
    if torch:
        from sklearn.metrics import accuracy_score
        y_true = y_true.numpy()
        y_pred = y_pred.numpy()
    for i in range(len(y_true)):
        
        set_true = set( np.where(y_true[i]==1)[0] )
        set_pred = set( np.where(y_pred[i]==1)[0] )
        tmp_a = None
        if len(set_true) == 0 and len(set_pred) == 0:
            tmp_a = 1
        else:
            tmp_a = len(set_true.intersection(set_pred))/\
            float( len(set_true.union(set_pred)) )
        acc_list.append(tmp_a)
    return np.mean(acc_list)


def save_checkpoint(state, is_best, fpath='checkpoint.pth.tar'):
    mkdir_if_missing(osp.dirname(fpath))
    torch.save(state, fpath)
    if is_best:
        shutil.copy(fpath, osp.join(osp.dirname(fpath), 'best_model.pth.tar'))