Pytorch supporter
Project description
pytorch-supporter
https://pypi.org/project/pytorch-supporter
pip install pytorch-supporter
Supported layers
import pytorch_supporter pytorch_supporter.layers.DictToParameters pytorch_supporter.layers.DotProduct pytorch_supporter.layers.GRULastHiddenState pytorch_supporter.layers.HiddenStateResetGRU pytorch_supporter.layers.HiddenStateResetLSTM pytorch_supporter.layers.HiddenStateResetRNN pytorch_supporter.layers.LazilyInitializedLinear pytorch_supporter.layers.LSTMLastHiddenState pytorch_supporter.layers.Reshape pytorch_supporter.layers.RNNLastHiddenState pytorch_supporter.layers.SelectFromArray
Supported utils
import pytorch_supporter text = '' pytorch_supporter.utils.clean_english(text) pytorch_supporter.utils.clean_korean(text)
Simple time series regression
import pytorch_supporter from sklearn.preprocessing import MinMaxScaler transformer = MinMaxScaler() transformer.fit(train_df[['Close']].to_numpy()) train_np_array = transformer.transform(validation_df[['Close']].to_numpy()) #window_length = sequence_length + 1 train_x, train_label = pytorch_supporter.utils.slice_time_series_data_from_np_array(train_np_array, x_column_indexes=[0], label_column_indexes=[0], sequence_length=7) #print(train_x.shape) #(973, 7, 1) #print(train_labels.shape) #(973, 1) #print(validation_x.shape) #(238, 7, 1) #print(validation_labels.shape) #(238, 1)
Multiple time series regression
import pytorch_supporter from sklearn.preprocessing import MinMaxScaler transformer = MinMaxScaler() transformer.fit(train_df[['Open', 'High', 'Low', 'Close', 'Adj Close', 'Volume']].to_numpy()) train_np_array = transformer.transform(validation_df[['Open', 'High', 'Low', 'Close', 'Adj Close', 'Volume']].to_numpy()) #window_length = sequence_length + 1 train_x, train_label = pytorch_supporter.utils.slice_time_series_data_from_np_array(train_np_array, x_column_indexes=[0, 1, 2, 3, 4, 5], label_column_indexes=[3], sequence_length=7) #print(train_x.shape) #(973, 7, 6) #print(train_labels.shape) #(973, 1) #print(validation_x.shape) #(238, 7, 6) #print(validation_labels.shape) #(238, 1)
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Close
Hashes for pytorch_supporter-0.0.20-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 4c1587638d60bc6b90a930538d1f161e3ac1772c4a50bcd17c7bcb6259e96736 |
|
MD5 | 15f7c6f021c85709dc7743c098449356 |
|
BLAKE2b-256 | 4be8d787bddf20d262456559735509475b7cab6d8f46e32135564dd24371f6b8 |