Skip to content

Instantly share code, notes, and snippets.

View ucalyptus2's full-sized avatar
🏠
Working from home

Sayantan Das ucalyptus2

🏠
Working from home
  • Canada
View GitHub Profile
<html>
<head>
<title>Sayantan Das</title>
</head>
<body bgcolor=blue><img src="http://mynameissayantan.webs.com/armenia_marmashen_2-wallpaper-1366x768.jpg"></img></body>
</html>
@ucalyptus2
ucalyptus2 / Hustlers.txt
Created January 2, 2018 12:14
Hustlers from 2016-17
Important people in my contacts whom I should focus on this 2018
------------------------------------------------------------------
Aayush Katiyar
Abhirup Coder
Abhishek Mukherjee
Abhishek Prasad
Aditya Mukherjee
rm /content/plots/roc/*.h5
rm /content/models/*.h5
zip -r $1.zip /content/plots/ /content/results/ /content/scores/ /content/models/
rm -rf /content/plots/ /content/results/ /content/scores/ /content/models/
mkdir /content/plots/ /content/results/ /content/scores/ /content/models/
import matplotlib.pyplot as plt
import os
import tensorflow as tf
from keras.utils import Sequence
from skimage.io import imread
from sklearn.utils import shuffle
import numpy as np
tf.logging.set_verbosity(tf.logging.ERROR)
from IPython import get_ipython
# coding: utf-8
"""
@Author: Sayantan Das
@Github: ucalyptus
"""
import ftplib
uid = 'ucalyptus'
import os
import numpy as np
def pad(img, crop_size):
h, w, c = img.shape
n_h = int(h/crop_size)
n_w = int(w/crop_size)
w_toadd = (n_w+1) * crop_size - w
import math
import tensorflow as tf
tf.config.experimental_run_functions_eagerly(True)
from tensorflow.python.keras.layers.convolutional import Conv
from tensorflow.python.keras.utils import conv_utils
class BiasHeUniform(tf.keras.initializers.VarianceScaling):
def __init__(self, seed=None):
super(BiasHeUniform, self).__init__(scale=1. / 3., mode='fan_in', distribution='uniform', seed=seed)
name: Upload Python Package
on:
release:
types: [created]
jobs:
deploy:
runs-on: ubuntu-latest
steps:
\documentclass[crop,tikz]{standalone}
\usepackage{tikz}
\usetikzlibrary{positioning}
\begin{document}
\begin{tikzpicture}
\node[circle, draw, thick] (z) {$\vec{a}_{real}$};
\node[circle, draw, thick, right=5em of z] (x) {$\vec{b}_{fake}$};
\draw[-stealth, thick] (z) -- node[above] {$G_{AB}(\vec{a})$} node[below, align=center] {generator\\ ($A\rightarrow B$)} (x);

Affordable Private ML Tuitions

ML private tuitions by Teacher: Sayantan Das. An affordable no-nonsense initiative started to help students who are new to this field and professionals who want to change their careers. Course is designed to be math-intensive.Participants will be involving themselves with the basics in math before any state of the art ML project can be tutored. Every class will be around 2 hours with theoretical explanation and programmatic session given equal weightage. Doubt clearing sessions will be separate. Students will get the chance to sit back and study before doubt clearing sessions are held for every corresponding class session.

Curriculum (84 hours)

The course is a bit different.In addition to clearing the foundations of the participant,the instructor wishes to fast-track each of his students into becoming early researchers of Machine Learning engineers. These steps will definitely involve learning to take part in online data science con