Minimize repeating colors matplotlib line chart, when plotting pandas dataframe

Question

I have seen that there is several questions about line colors for matplotlib, however have not been able to localize an example with dataframe.

A subset of my raw data is as follows (full set 8400r x 14c)

[
    {
        "datapoints": [
            [
                12151151,
                1647944400
            ],
            [
                5452151,
                1647944700
            ],
            [
                null,
                1647945000
            ]
        ],
        "targets": "port1.good_rx",
        "tags": {
            "nonNegativeDerivative": 1,
            "name": "good_rx"
        }
    },
    {
        "datapoints": [
            [
                12354545,
                1647944400
            ],
            [
                25454524,
                1647944700
            ],
            [
                17222,
                1647945000
            ]
        ],
        "targets": "port2.good_rx",
        "tags": {
            "nonNegativeDerivative": 1,
            "name": "good_rx"
        }
    }
]

This data is the normalized using

# https://github.com/esc/graphite2pandas/blob/cf914d0d8d5ef3f4e6d6f837366423c5674882ef/graphite2pandas.py
def _localize(index, tz):
    index = index.tz_localize('UTC')
    index = index.tz_convert('CET')
    return index

def g2p(url, localize='CET'):
    resp = requests.get(url)
    decoded_json = json.loads(resp.content)
    times, values, labels =  [], [], []
    for element in decoded_json:
        labels.append(element['target'])
        datapoints = list(zip(*element['datapoints']))
        values.append(datapoints[0])
        times.append(datapoints[1])
    index = pandas.DatetimeIndex((numpy.array(times[0],dtype='datetime64[s]')))
    if localize:
        index = _localize(index, localize)
    return pandas.DataFrame(
            dict(((labels[i], values[i]) for i in range(len(values)))),
            index=index)

Following these steps I then do some cleanup work on the columns and plot the chart. This is when I realized that a get a lot of repetetive colors, so I tried to set the color using (np.random.random(), np.random.random(), np.random.random()) this resulted in all columns having the the same colors. How could I plot one column at the time in order to randomly generate color per column (line)

df = g2p(URL) #return sa pd.DataFrame()  8400 rows x 14 col - Time(index), bytes(row), port(column)

plot = df.plot(c=(np.random.random(), np.random.random(), np.random.random()))     
plot.set_title("test")
fig = plot.get_figure()

fig.savefig("fileout.png", transparent=True, bbox_inches='tight')

Default

When using random