I decided to learn python a little bit more. For me best learning was always on a real task / project, so I decided to create a few CSV files with data like:
- price of oil / metals
- Dow Jones index
- Bitcoin price
- google trends for “buy bitcoin”
- and a few more
Send that data to a friend (Data Analyst) and ask him for a favour of checking if there’s a strong correlation between a pair or a mix of above. Interesting!
With the help of available methods in cryptory package get above data should be piece of cake. Storing in a CSV file seems like a tiny, tiny challange to me at the moment. Where to start ?
Let’s code! Well.. using vim as a code editor for language I completly do not know doesn’t seem fair. Let’s install Eclipse on the Raspberry Pi first followed by a PyDev plugin.
import pandas as pd from cryptory import Cryptory my_cryptory = Cryptory(from_date = "2017-01-01") btc_cmc = my_cryptory.extract_coinmarketcap("bitcoin") my_df = pd.DataFrame(btc_cmc) my_df.to_csv('btc_coinmarket.csv', index=False, header=True) reddit = my_cryptory.extract_reddit_metrics(subreddit="bitcoin", metric="total-subscribers") my_df = pd.DataFrame(reddit) my_df.to_csv('btc_reddit.csv', index=False, header=True) gtrends = my_cryptory.get_google_trends(kw_list=["buy bitcoin"]) my_df = pd.DataFrame(gtrends) my_df.to_csv('btc_gtrends.csv', index=False, header=True) stock = my_cryptory.get_stock_prices(market="%5EDJI") my_df = pd.DataFrame(stock) my_df.to_csv('btc_stock.csv', index=False, header=True) rates = my_cryptory.get_exchange_rates(from_currency="USD", to_currency="EUR") my_df = pd.DataFrame(rates) my_df.to_csv('btc_rates.csv', index=False, header=True) metal = my_cryptory.get_metal_prices() my_df = pd.DataFrame(metal) my_df.to_csv('btc_metal.csv', index=False, header=True) exit()
Below links to generated files:
Thanks for reading,