""" Transform output into dataframe """
agg_out = pd.DataFrame(columns = allsecurities.columns.values.tolist())
for k, v in client_portfolio.portfolio.items():
current_security = allsecurities.loc[k].to_frame(name = allsecurities.loc[k].name).T
agg_out = pd.concat([agg_out, current_security], axis = 0)
""" Visualization """
print ("Check browser for visualization")
visual.visualize(kmean_out, option_list, 'temp-plot.html')
visual.visualize(agg_out, option_list_agg, 'temp-plot2.html')
""" Recommendations """
# TODO: fix data frame index - security - name confusion
option_list['Name'] = option_list.index
option_list_agg['Name'] = option_list_agg.index
#option_list.reset_index(level = 0, inplace = True)
ranked_options = rank_by_preferences([1,3,4,5,6], option_list, None, False).index.values.tolist()
ranked_options_agg = rank_by_preferences([1,3,4,5,6], option_list_agg, None, False).index.values.tolist()
print ("\nRecommendations from KMeans")
ranked_options_kmeans = pd.DataFrame(columns = Y.columns.values.tolist())
for i in range(0, len(ranked_options)):
opt = Y.loc[ranked_options[0]]
ranked_options_kmeans = pd.concat([ranked_options_kmeans, Y.loc[ranked_options[i]]], axis = 0)
print ("\nTop 5 recommendations from AgglomerativeClustering")
print (Y.loc[ranked_options_agg[0]])
print (Y.loc[ranked_options_agg[1]])
print (Y.loc[ranked_options_agg[2]])
print (Y.loc[ranked_options_agg[3]])
print (Y.loc[ranked_options_agg[4]])
''' sharpe portfolio ''' # picks stock with highest sharpe ratio
portfolio = []
for i in range(0, len(groups)):
max_sh, idx = 0, 0
for j in range (0, len(groups[i])):
if stocks.iloc[groups[i][j]].Sharpe_Ratio > max_sh:
max_sh = stocks.iloc[groups[i][j]].Sharpe_Ratio
idx = groups[i][j]
portfolio.append(stocks.iloc[idx].Name)
pd.Series(portfolio, name = 'Stocks').to_hdf("../../evaluation/portfolios_ml/fundamental/ml_fundamental_portfolio_kmeans_1.hdf5", "dataset1/x")
''' random portfolio '''
portfolio = []
for i in range(0, len(groups)):
j = choice(groups[i])
portfolio.append(stocks.iloc[j].Name)
pd.Series(portfolio, name = 'Stocks').to_hdf("../../evaluation/portfolios_ml/fundamental/ml_fundamental_portfolio_kmeans_2.hdf5", "dataset1/x")
''' preferences portfolio '''
portfolio = []
for i in range(0, len(groups)):
grouped_stocks = pd.DataFrame()
for j in range (0, len(groups[i])):
row = stocks.iloc[groups[i][j]]
grouped_stocks = grouped_stocks.append(row)
ranking = rank_by_preferences(preferences, grouped_stocks, economy, False)
portfolio.append(ranking.iloc[0].Name)
pd.Series(portfolio, name = 'Stocks').to_hdf("../../evaluation/portfolios_ml/fundamental/ml_fundamental_portfolio_kmeans_3.hdf5", "dataset1/x")
''' similarity portfolio '''
pyfancy.pyfancy().cyan("Generating random economy...").output()
economy = 'Mid'
#economy = choice(['Mid', 'Late', 'Recession', 'Early'])
if setTimer:
time.sleep(1)
print('We are using cycle {} for economy'.format(economy))
options = [0, 1, 2, 3, 4]
compliments = [
"Briliant!", "Well done!", "Excellent!", "Magnificent!",
"You're investements are growing!"
]
user_in = 'start'
all_preferences = rank_by_preferences([1, 2, 3, 4, 5, 6, 7], allsecurities,
economy, False).index.values.tolist()
preferences_default = True
while (user_in != 'exit'):
print("\n")
print(choice(compliments))
print("What would you like to do next?..\n")
print("1 - Visualize portfolio")
print("2 - Visualize client")
print("3 - Set / reset preferences")
print("4 - Fill portfolio")
print("5 - Save portfolio")
print("0 - Exit")
user_in = [int(x) for x in input('Insert your options.. ').split()]