hengck23

from common import *

import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F
from my_tabnet.sparsemax import Sparsemax

class Identity(torch.nn.Module):
    def forward(self, x):

hengck23

from common import *
num_class = 266


# https://github.com/qiuqiangkong/audioset_tagging_cnn/blob/8d9999d72b282d2dc50a5b5f668dd91369f853c5/pytorch/models.py
# https://www.kaggle.com/hidehisaarai1213/introduction-to-sound-event-detection
# https://github.com/qiuqiangkong/sed_from_wekaly_labelled_data/blob/master/spectrogram_to_wave.py
class ConvBlock(nn.Module):
    def __init__(self, in_channel, out_channel, pool_size=1):
        super(ConvBlock, self).__init__()

hengck23

from common import *

# https://scale.com/blog/pytorch-improvements
# Making Pytorch Transformer Twice as Fast on Sequence Generation.

# https://towardsdatascience.com/how-to-code-the-transformer-in-pytorch-24db27c8f9ec
class FeedForward(nn.Module):
    def __init__(self, dim, ff_dim=2048, dropout=0.1):
        super().__init__()
        self.linear1 = nn.Linear(dim, ff_dim)

hengck23

class ConvBn1d(nn.Module):
    def __init__(self, in_dim, out_dim, kernel_size, padding, stride=1, pool=None):
        super().__init__()
        self.conv = nn.Conv1d(in_dim, out_dim, kernel_size=kernel_size, padding=padding, stride=stride)
        self.bn   = nn.BatchNorm1d(out_dim)
        self.relu = nn.SiLU(inplace=True)

        if pool is None:
            self.pool=nn.Identity()
        else:

hengck23

from common import *
from scipy.signal import get_window
from scipy import signal

#https://gist.github.com/keunwoochoi/be18701219fb671f6c74b3d6e0740513


def next_pow2(x):
    return int(np.ceil(np.log2(x)))

hengck23

from common import *
from ventilator import *
import faiss

data_dir = root_dir+'/data'
train_df = pd.read_csv(data_dir + '/train.csv')
test_df  = pd.read_csv(data_dir + '/test.csv')


if 1:  #check statistics