HeshamMeneisi

def series_to_supervised(df, n_in=1, n_out=1, targets=[], dropnan=True):
    """
    Converts a time series Pandas DataFrame into a supervised learning problem
    returns: X(t-n_in+1,...,t), y(t+1,....,t+n_out)
    """
    assert n_in > 0 and n_out > 0
    assert all([t in df.columns for t in targets])
    
    n_vars = len(df.columns)
    

HeshamMeneisi

word_list = [doc.words for doc in docs]

bigram = gensim.models.Phrases(word_list)
bigram_phr = gensim.models.phrases.Phraser(bigram)

bi_docs = []
for doc in docs:
    words = bigram_phr[doc.words]
    tags = doc.tags

HeshamMeneisi

from gensim.models import Doc2Vec
import pickle

with open('./docs_proc.pkl', 'rb') as file:
    docs = pickle.load(file)

model = Doc2Vec(docs, vector_size = 500, window = 9, min_count = 20, workers=8, dm=0)
with open('./d2v_model.pkl', 'wb') as file:
    pickle.dump(model, file)
    

HeshamMeneisi

import numpy as np
import time
import sys

print("## Checking Keras\n\n")
import keras.backend as K
backend = K.backend()

vlen = 10 * 30 * 768
iters = 1000

HeshamMeneisi

import gym
import time
from gym import envs

print(envs.registry.all())

env = gym.make('CartPole-v0')
env.reset()
t = 0