What came from fucking up a tech interview with Facebook

Having decided to focus on what I already know, I disgarded learning binary search trees, which just so happened to be the coding question asked on the technical interview with Facebook. FAIL ! Oh well, moving on, I was prompted to look into its uses and ultimately how the game Doom came into existence. This led to the concept of Binary Search Partitions. Having experimented with 3d rendering back in the 90's, I instinctively knew the rendering limitations. This YouTube video explains how gave developers overcame them. https://www.youtube.com/watch?v=yTRzfKh4Tg0

Maya - Newly released single "Quicksand"

I would normally not post something like this, but feel compelled as I was at the hospital when she was born. https://mayamuzic.com/

PHP Set File Permissions

Recently, images uploaded to SpiderOnFire via the Widespread Augmented Reality app were suddenly writing to the server with permissions 0600. No es bueno, since these images need to be viewed in a web browser and on Android. Naturally, the server guys had no idea why the sudden deviation from the default upload permissions of 0644. Therefore I had to add "chmod($target_dir, 0644);" in all my upload and image resizing scripts. Seems to have fixed the issue, but I am sure that I have broken something else by going back in to change code that hasn't been touched in 4 years.

How to Be Anti-Social in Augmented Reality

Augmented Reality Heads Up Display

Communicate anonymously through an augmented reality heads up display for Android only. Download app from Google Play.

---

Python Machine Learning on Amazon stock prices

This Python code reads Amazon's historical stock prices from 2014 to 2019. I downloaded the CSV file from Yahoo Finance. The chart below shows how well this algorithm predicts stocks prices when compared to actual stock prices. The code was cobbled together from snippets at Analytics Vidhya and Medium.

# importing libraries

import pandas as pd

import numpy as np

from datetime import date, datetime

import calendar

#importing required libraries

from sklearn.preprocessing import MinMaxScaler

from keras.models import Sequential

from keras.layers import Dense, Dropout, LSTM

# reading the data
df = pd.read_csv('amzn2.csv')


# looking at the first five rows of the data

print('\n Original data:')

print(df.head())

print('\n Shape of original data:')

print(df.shape)


# setting the index as date

df['Date'] = pd.to_datetime(df.Date,format='%Y-%m-%d')

df.index = df['Date']


#creating dataframe

data = df.sort_index(ascending=True, axis=0)

new_data = pd.DataFrame(index=range(0,len(df)),columns=['Date', 'Close'])

#populate new data frame

for i in range(0,len(data)):

    new_data['Date'][i] = data['Date'][i]

    new_data['Close'][i] = data['Close'][i]


#setting index

new_data.index = new_data.Date

new_data.drop('Date', axis=1, inplace=True)


#creating train and test sets

dataset = new_data.values

#the csv file has 1260 records

train = dataset[0:630,:]

valid = dataset[630:,:]


#converting dataset into x_train and y_train

scaler = MinMaxScaler(feature_range=(0, 1))

scaled_data = scaler.fit_transform(dataset)


x_train, y_train = [], []

for i in range(60,len(train)):

    x_train.append(scaled_data[i-60:i,0])

    y_train.append(scaled_data[i,0])

x_train, y_train = np.array(x_train), np.array(y_train)


x_train = np.reshape(x_train, (x_train.shape[0],x_train.shape[1],1))


# create and fit the LSTM network

model = Sequential()

model.add(LSTM(units=50, return_sequences=True, input_shape=(x_train.shape[1],1)))

model.add(LSTM(units=50))

model.add(Dense(1))


model.compile(loss='mean_squared_error', optimizer='adam')

model.fit(x_train, y_train, epochs=1, batch_size=1, verbose=2)


#predicting 246 values, using past 60 from the train data

inputs = new_data[len(new_data) - len(valid) - 60:].values

inputs = inputs.reshape(-1,1)

inputs  = scaler.transform(inputs)


X_test = []

for i in range(60,inputs.shape[0]):

    X_test.append(inputs[i-60:i,0])

X_test = np.array(X_test)


X_test = np.reshape(X_test, (X_test.shape[0],X_test.shape[1],1))

closing_price = model.predict(X_test)

closing_price = scaler.inverse_transform(closing_price)

rms=np.sqrt(np.mean(np.power((valid-closing_price),2)))

print('\n Root Mean Square Deviation:')

print(rms)

#for plotting

#plot

import matplotlib.pyplot as plt

train = new_data[:630]

valid = new_data[630:]

valid['Predictions'] = closing_price

plt.plot(train['Close'])

plt.plot(valid[['Close','Predictions']])

plt.show()

Victory Cross Country 2011 motorcycle stalling

Came to a stop okay, but then accelerated and the bike stalled like it was flooded but it's fuel injection. Rolled to a stop on a downwill slope and reastarted briefly. The engine light was on and it stalled again, never to restart. Battery working as evidenced by lights being on and pump priming. The culprit was a busted and dangling tip over sensor that cuts off the fuel supply when bike is on its side. Since the tip over sensor broke off and was dangling upside down, the bike was going no where fast. Must replace tip over sensor. Found one here.