rohit-gupta

begin;
lock my_table in exclusive mode;
delete from my_table where key='key';
insert into my_table (key, value) values ('key', 'value');
commit;

rohit-gupta

[alias]
    history = log --pretty=format:'%C(red) %h %C(yellow) %ad %C(green) %an %C(white) %s' --date=relative
    newbranch = checkout -b

rohit-gupta

python helloworld.py > helloworld.out & echo $! > helloworld.pid
cat helloworld.pid | xargs kill -9

rohit-gupta

sudo -H pip install drawille
wget https://raw.githubusercontent.com/asciimoo/drawille/master/examples/xkcd.py
mkdir -p ~/.terminalscripts/
mv xkcd.py ~/.terminalscripts/xkcd.py
alias xkcd="python ~/.terminalscripts/xkcd.py"

rohit-gupta

library(survey)

load("~/CCES/CCES2016_Common_FirstRelease.RData")
survObj <- svydesign(ids = ~1, data = x, weights = x$commonweight_post)
prop.table(svytable(~CC16_410a, design = survObj))

rohit-gupta

from nltk.tokenize import sent_tokenize


text = """All of the MSM is fake news at this point. Their blind hatred of Trump has caused them to throw away what credibility they once had.

#DeathToTheMSM"""
sent_tokenize_list = sent_tokenize(text)
for sentence in sent_tokenize_list:
  print sentence

rohit-gupta

# clip videos and delete original
ffmpeg -hide_banner -y -i clipped.avi -vf "trim=start=0:end=2,setpts=PTS-STARTPTS" -an tiny.avi && rm -rf clipped.avi

# Extract Frames
ffmpeg -i vid4.avi frames/vid4/%04d.jpg -hide_banner

# Download videos from Youtube
youtube-dl --get-filename -o '%(id)s_%(fps)s.%(ext)s' w4JM08PDEng -f 'bestvideo[height<=240]'

rohit-gupta

def generator(features, labels, batch_size):

 # Create empty arrays to contain batch of features and labels#

 batch_features = np.zeros((batch_size, 64, 64, 3))
 batch_labels = np.zeros((batch_size,1))

 while True:
   for i in range(batch_size):
     # choose random index in features

rohit-gupta

# Create generator
train_generator = MSRVTTSequence(train_captions, video_folder=videos_folder, fps_dict=video_fps, tag_dict=tags, batch_size=16)
validation_generator = MSRVTTSequence(validation_captions, video_folder=videos_folder, fps_dict=video_fps, tag_dict=tags, batch_size=16)


from keras.applications.resnet50 import ResNet50
from keras.layers import TimeDistributed, Bidirectional
from keras.layers import Input, LSTM, Dense
from keras.models import Model
from keras.callbacks import CSVLogger, ModelCheckpoint, ReduceLROnPlateau

rohit-gupta

from keras.models import Model
from keras.layers import Input, LSTM, Dense, TimeDistributed

inputs = Input(batch_shape=(8,16,1024))
layer = LSTM(256, return_state=True, return_sequences=True)
outputs = layer(inputs)

# TODO Add Teacher Forcing
output, state = outputs[0], outputs[1:]
output = LSTM(256)(output, initial_state=state)