manashmandal

def build_deep_autoencoder(img_shape,code_size=32):
    """PCA's deeper brother. See instructions above"""
    H,W,C = img_shape
    
    encoder = keras.models.Sequential()
    encoder.add(L.InputLayer(img_shape))

    encoder.add(L.Flatten())
    encoder.add(L.Dense(code_size*8, activation='relu'))
    encoder.add(L.Dense(code_size*4, activation='tanh'))

manashmandal

def visualize(img,encoder,decoder):
    """Draws original, encoded and decoded images"""
    code = encoder.predict(img[None])[0]
    reco = decoder.predict(code[None])[0]

    plt.subplot(1,3,1)
    plt.title("Original")
    plt.imshow(img)

    plt.subplot(1,3,2)

manashmandal

def build_pca_autoencoder(img_shape,code_size=32):
    """
    Here we define a simple linear autoencoder as described above.
    We also flatten and un-flatten data to be compatible with image shapes
    """
    
    encoder = keras.models.Sequential()
    encoder.add(L.InputLayer(img_shape))
    encoder.add(L.Flatten())                  #flatten image to vector
    encoder.add(L.Dense(code_size))           #actual encoder

manashmandal

import numpy as np
import os
import cv2
import pandas as pd
import tarfile
import tqdm


ATTRS_NAME = "./data/lfw_attributes.txt"  # http://www.cs.columbia.edu/CAVE/databases/pubfig/download/lfw_attributes.txt
RAW_IMAGES_NAME = "./data/lfw.tgz"  # http://vis-www.cs.umass.edu/lfw/lfw.tgz

manashmandal

train_X = np.empty((18000, 200, 200))

for i in range(0,18000):
    
    if i >= 3600 and i < 5400:
        img = np.rot90(org_train_x[i].reshape(28,28),k =3, axes=(0,1))
        img = np.fliplr(img)
    elif i >= 9000 and i < 10800:
        img = np.rot90(org_train_x[i].reshape(28,28),k =3, axes=(0,1))
        img = np.fliplr(img)

manashmandal

import numpy as np
from sklearn.model_selection import train_test_split

def get_train_test(split_ratio=0.6, random_state=42):
    # Get available labels
    labels, indices, _ = get_labels(DATA_PATH)

    # Getting first arrays
    X = np.load(labels[0] + '.npy')
    y = np.zeros(X.shape[0])

manashmandal

import numpy as np
import os

DATA_PATH = "./data/"

# Input: Folder Path
# Output: Tuple (Label, Indices of the labels, one-hot encoded labels)
def get_labels(path=DATA_PATH):
    labels = os.listdir(path)
    label_indices = np.arange(0, len(labels))

manashmandal

import numpy as np
import librosa

def wav2mfcc(file_path, max_pad_len=11):
    wave, sr = librosa.load(file_path, mono=True, sr=None)
    wave = wave[::3]
    mfcc = librosa.feature.mfcc(wave, sr=16000)
    pad_width = max_pad_len - mfcc.shape[1]
    mfcc = np.pad(mfcc, pad_width=((0, 0), (0, pad_width)), mode='constant')
    return mfcc

manashmandal

pinMode()
digitalRead()
digitalWrite()

manashmandal

import spacy
import pandas as pd
import argparse
import re
nlp = spacy.load('en')


# Get the name of the file
parser = argparse.ArgumentParser()
parser.add_argument("file_path", help="Enter csv filepath here")