Filtering and sorting data with pandas

Question

I want to get some part of information from API, but I don't know how to filter data (I want to get only chosen values and don't get values if key don't contain "BTC" string) I'm trying to do something like this:

{"BTC_MINT":{"volume":11.00, "high24":0.002, "low24":0.001},
 "BTC_NOTE":{"volume":11.00, "high24":0.002, "low24":0.001}}

I started with pandas, but I don't know if is it proper way.

link = 'https://poloniex.com/public?command=returnTicker'
with urllib.request.urlopen(link) as rawdata:
    data = rawdata.readall().decode()
data = json.loads(data)
print(data.items())
data = pd.DataFrame([[cur, last, volume, high24, low24] 
                     for cur, d in data.items() 
                     for last, x, x, x, volume, x, x, high24, low24 in d.items()])

Unfortunately, this code don't work. I get following error:

[cur, last, volume, high24, low24] for cur, d, x, w, d, q in data.items() for last, x, x, x, volume, x, x, high24, low24 in d.items()
ValueError: need more than 2 values to unpack

Could someone help and tell me how should I do it?

Alexander · Accepted Answer · 2015-12-17 22:23:54Z

df = pd.DataFrame({symbol: {"baseVolume": data[symbol].get("baseVolume"), 
                            "high24hr": data[symbol].get("high24hr"), 
                            "low24hr": data[symbol].get("low24hr")} 
                   for symbol in data}).T
>>> df.head()
          baseVolume    high24hr     low24hr
BTC_1CR   0.00000000  0.00000000  0.00000000
BTC_ABY   0.01968682  0.00000020  0.00000019
BTC_ADN   0.00000000  0.00000000  0.00000000
BTC_ARCH  0.07205024  0.00004813  0.00004693
BTC_BBR   0.19846259  0.00002123  0.00002115

To just get the names in the index starting with BTC, do the following:

>>> df[df.index.str.startswith('BTC')].head()

          baseVolume    high24hr     low24hr
BTC_1CR   0.00000000  0.00000000  0.00000000
BTC_ABY   0.01968682  0.00000020  0.00000019
BTC_ADN   0.00000000  0.00000000  0.00000000
BTC_ARCH  0.07205024  0.00004813  0.00004693
BTC_BBR   0.19846259  0.00002123  0.00002115

tornesi · Accepted Answer · 2015-12-17 21:46:30Z

1

You can just pass your dictionary (data) to pd.Dataframe to create a pandas dataframe. If you want to subset it to only contain columns with the string BTC in it, you can do:

df = pd.DataFrame(data)
new_cols = [x for x in df.columns if x.find('BTC') > -1]
new_df = df[new_cols]

answered Dec 17, 2015 at 21:46

tornesi

2692 silver badges8 bronze badges

Collectives™ on Stack Overflow

Filtering and sorting data with pandas

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related