将 CSV 转置为 DF Pandas

Question

I have a JSON that looks like this:

{ID:{"ATT1": "VAL", "ATT2": "VAL","ATT3": "VAL",...}}

This JSON has around 3000+ ID's inside of it, all with its own dict of ATT/VAL pairs.

After I read that I used this:

with open("result.json", "r") as data_file:
    data = json.load(data_file)

df = pd.DataFrame(dict([ (k,pd.Series(v)) for k,v in data.items() ]))

df.to_csv("pandarised.csv", index=False)

Which then gives me a csv like this:

ID
{"ATT": "some att", "VAL": "some val"}
{"ATT": "some att", "VAL": "some val"}
{"ATT": "some att", "VAL": "some val"}
{"ATT": "some att", "VAL": "some val"}
{"ATT": "some att", "VAL": "some val"}
{"ATT": "some att", "VAL": "some val"}

There are about 3000 columns, where each column has a new ID and its own set of ATT/VAL dicts. I was trying to read this from DF into Pandas as a new DF, and was trying to make it look like this:

   ATT ATT2 ATT3 etc..
ID VAL VAL  VAL  etc..
ID VAL VAL  VAL  etc..
ID VAL VAL  VAL  etc..
ID VAL VAL  VAL  etc..
ID VAL VAL  VAL  etc..
etc..

I was thinking to try using either concat with some combination of transpose but I cant seem get a grip on where to start with this.

Thank you

Answer 1

I think this is what you are after:

data = [{"ID":{"ATT": "VAL", "ATT2": "VAL"}},
        {"ID":{"ATT": "VAL", "ATT2": "VAL"}}]
df = pd.DataFrame()

for row in data:
    df = df.append(row[list(row.keys())[0]], ignore_index=True)