We use cookies to ensure you have the best browsing experience on our website. Please read our cookie policy for more information about how we use cookies.
- Prepare
- Artificial Intelligence
- Statistics and Machine Learning
- Forecasting passenger traffic
- Discussions
Forecasting passenger traffic
Forecasting passenger traffic
Sort by
recency
|
13 Discussions
|
Please Login in order to post a comment
from statsmodels.tsa.ar_model import AutoReg as AR from scipy import interpolate
N = int(input())
data = [] for i in range(N): data.append(int(input().split()[1]))
leng = list(range(len(data)))
interpolation = interpolate.interp1d(leng, data, fill_value='extrapolate')
y = interpolation(range(len(data), len(data)+12))
for x in y: print(x)
from statsmodels.tsa.ar_model import AutoReg as AR
N = int(input())
data = [] for i in range(N): data.append(int(input().split()[1]))
model = AR(data, lags=1) fitted_model = model.fit() pred = fitted_model.predict()[:12]
for i in pred: print(i, end ='\n')
Don't waste your time on this problem. It is garbage. The editorial suggest that generate independent samples from the estimated distribution. Obviously it can only give something close to the expected solution by chance. Some of the top solutions simple return a list of values depending on first input line (there is only two test cases). Some other solution uses the interpolation function of the scipy module. They even produces negative values. So it is not clear how can they achived max points.
can i use Arima or fbprophet with this? last exercise i tried to use those libraries it output an error
Very interesting to read about new marketing concepts! Specifically about Growth hacking, I recently came across an article https://www.plerdy.com/blog/growth-hacking-and-talk-with-georgiy/ about it and the opinion is interesting!