1 - How do you use the API?
- Make sure to read this page after going through the /about section.
- To use the API, you'll need a valid temporary token (valid for 60 minutes).
You can get one by signing up, logging in, and navigating to /token.
- Once you have your token, you can use it to access the API endpoints. How do you do that?
The simplest way to interact with the
API is by using
curl at the
command line.
More importantly,
you can also use
any programming language that supports sending HTTP requests.
Below are a few
curl commands to get you started. You can copy and paste them into the
curlconverter to generate the equivalent commands for your preferred programming language.
Feel free to explore the
other API endpoints, and check out the datasets available at
Techtonique/datasets.
It's also worth noting that
nnetsauce.MTS
is used for Forecasting,
nnetsauce.DeepClassifier for Machine Learning Classification, and
nnetsauce.DeepRegressor for Machine Learning Regression.
- Univariate forecasting using RidgeCV, 5 hidden features, 25 lags,
KDE sampling, 4 replications, 3 steps ahead:
curl -X POST \
-H "Authorization: Bearer YOUR_TOKEN" \
-F "file=@/Users/t/Documents/datasets/time_series/univariate/a10.csv;type=text/csv" \
"https://www.techtonique.net/forecasting?base_model=RidgeCV&n_hidden_features=5&lags=25&type_pi=kde&replications=4&h=3"
- Univariate forecasting using ElasticNet, 5 hidden features, 25 lags, 3 steps ahead:
curl -X POST \
-H "Authorization: Bearer YOUR_TOKEN" \
-F "file=@/Users/t/Documents/datasets/time_series/univariate/a10.csv;type=text/csv" \
"https://www.techtonique.net/forecasting?base_model=RidgeCV&n_hidden_features=5&lags=25&h=3"
- Univariate forecasting using ElasticNet, 5 hidden features, 25 lags, 3 steps ahead:
curl -X POST \
-H "Authorization: Bearer YOUR_TOKEN" \
-F "file=@/Users/t/Documents/datasets/time_series/univariate/AirPassengers.csv;type=text/csv" \
"https://www.techtonique.net/forecasting?base_model=ElasticNet&n_hidden_features=5&lags=25&h=3"
- Multivariate forecasting with RidgeCV, 5 hidden features, 25 lags,
3 steps ahead, KDE sampling, 4 replications:
curl -X POST \
-H "Authorization: Bearer YOUR_TOKEN" \
-F "file=@/Users/t/Documents/datasets/time_series/univariate/AirPassengers.csv;type=text/csv" \
"https://www.techtonique.net/forecasting?base_model=ElasticNet&n_hidden_features=5&lags=25&h=3&type_pi=kde&replications=4"
- Multivariate forecasting using ElasticNet, 5 hidden features, 25 lags, 3 steps ahead:
curl -X POST \
-H "Authorization: Bearer YOUR_TOKEN" \
-F "file=@/Users/t/Documents/datasets/time_series/multivariate/ice_cream_vs_heater.csv;type=text/csv" \
"https://www.techtonique.net/forecasting?base_model=RidgeCV&n_hidden_features=5&lags=25&h=3"
- Multivariate forecasting using ElasticNet, 5 hidden features, 25 lags, 3 steps ahead:
curl -X POST \
-H "Authorization: Bearer YOUR_TOKEN" \
-F "file=@/Users/t/Documents/datasets/time_series/multivariate/ice_cream_vs_heater.csv;type=text/csv" \
"https://www.techtonique.net/forecasting?base_model=ElasticNet&n_hidden_features=5&lags=25&h=3"
- Multivariate forecasting with ElasticNet, 5 hidden features, 25 lags,
3 steps ahead, KDE sampling, 4 replications, 5 steps ahead:
curl -X POST \
-H "Authorization: Bearer YOUR_TOKEN" \
-F "file=@/Users/t/Documents/datasets/time_series/multivariate/ice_cream_vs_heater.csv;type=text/csv" \
"https://www.techtonique.net/forecasting?base_model=ElasticNet&n_hidden_features=5&lags=25&h=3&type_pi=kde&replications=4&h=5"
2 - More curl examples
3 - Microsoft Excel examples
See: