Getting Started Code for TIMSER Challenge on AIcrowd¶
Author : Shubhamai¶
Downloading Data 📚¶
In [ ]:
!pip install git+https://gitlab.aicrowd.com/yoogottamk/aicrowd-cli.git
API_KEY = "b42ff570ebfc36a1973801a5b2a3daf7"
!aicrowd login --api-key $API_KEY
In [ ]:
!aicrowd dataset download -c timser >/dev/null
Importing Libraries¶
In [ ]:
import pandas as pd
from fbprophet import Prophet
In [ ]:
train_df = pd.read_csv('train.csv')
# val_df= pd.read_csv("val.csv")
# sample_submission = pd.read_csv("submission.csv")
train_df.head()
In [ ]:
df = pd.concat([train_df, val_df])
df.columns = ['ds', 'y']
print(df.tail())
In [ ]:
m = Prophet()
m.fit(df)
Making Predictions¶
Now we are going to make future predictions
In [ ]:
future = m.make_future_dataframe(periods=2763)
forecast = m.predict(future)
print(forecast[['ds', 'yhat', 'yhat_lower', 'yhat_upper']].tail())
In [ ]:
fig = m.plot(forecast)
fig.savefig('prophetplot.svg')
In [ ]:
sample_submission['value'] = forecast['yhat']
sample_submission.to_csv("submission.csv", index=False)
To download the generated csv in colab run the below command¶
In [ ]:
try:
from google.colab import files
files.download('submission.csv')
except:
print("Option Only avilable in Google Colab")