botiverse.bots.WhizBot package#

Submodules#

botiverse.bots.WhizBot.WhizBot module#

class botiverse.bots.WhizBot.WhizBot.WhizBot(repr='BERT')[source]#

Bases: object

A class that provides an interface for the WhizBot-BERT and WhizBot-GRU models.

Initializes WhizBot and sets its representation type. :param repr: The representation type of the WhizBot model. Either “BERT” or “GRU”. :type repr: str

read_data(file_path)[source]#

Reads and pre-processes the data, sets up the model based on the data and prepares the train-validation split.

Parameters:

file_path (str) – The path to the file that contains the dataset.

Returns:

None

train(epochs=10, batch_size=32)[source]#

Trains the model using the training dataset.

Parameters:

  • epochs (int) – The number of training epochs.

  • batch_size (int) – The number of training examples utilized used to make one paramenters updat.

Returns:

None

validation(batch_size=32)[source]#

Tests the model performance using the validation dataset and calculates the accuracy.

Parameters:

batch_size (int) – The number of training examples utilized used to make one paramenters updat.

Returns:

None

infer(string)[source]#

Performs inference using the model.

Parameters:

string (str) – The input string to perform inference on.

Returns:

A random response from the response list of the predicted label.

save(path)[source]#

Saves the model parameters to the given path.

Parameters:

path (str) – The path where the model parameters will be saved.

Returns:

None

load(path)[source]#

Loads the model parameters from the given path.

Parameters:

path (str) – The path from where the model parameters will be loaded.

Returns:

None

botiverse.bots.WhizBot.WhizBot_BERT module#

class botiverse.bots.WhizBot.WhizBot_BERT.WhizBot_BERT[source]#

Bases: object

An interface for the WhizBot_BERT model which is a BERT model with a Feed Forward layes at the end.

Initializes WhizBot_BERT, and will prepare the GPU device based on CUDA availability.

read_data(file_path)[source]#

Reads and pre-processes the data, sets up the model based on the data and prepares the train-validation split.

Parameters:

file_path (str) – The path to the file that contains the dataset.

Returns:

None

train(epochs=10, batch_size=32)[source]#

Trains the model using the training dataset.

Parameters:

  • epochs (int) – The number of training epochs.

  • batch_size (int) – The number of training examples utilized used to make one paramenters updat.

Returns:

None

validation(batch_size=32)[source]#

Tests the model performance using the validation dataset and calculates the accuracy.

Parameters:

batch_size (int) – The number of training examples utilized used to make one paramenters updat.

Returns:

None

infer(string)[source]#

Performs inference using the model.

Parameters:

string (str) – The input string to perform inference on.

Returns:

A random response from the response list of the predicted label.

save(path)[source]#

Saves the model parameters to the given path.

Parameters:

path (str) – The path where the model parameters will be saved.

Returns:

None

load(path)[source]#

Loads the model parameters from the given path.

Parameters:

path (str) – The path from where the model parameters will be loaded.

Returns:

None

botiverse.bots.WhizBot.WhizBot_GRU module#

class botiverse.bots.WhizBot.WhizBot_GRU.WhizBot_GRU[source]#

Bases: object

An interface for the WhizBot_GRU model which is based on simple GRU model

Initializes WhizBot_GRU, and will prepare the GPU device based on CUDA availability.

read_data(file_path)[source]#

Reads and pre-processes the data, sets up the model based on the data and prepares the train-validation split.

Parameters:

file_path (str) – The path to the file that contains the dataset.

Returns:

None

train(epochs=50, batch_size=32)[source]#

Trains the model using the training dataset.

Parameters:

  • epochs (int) – The number of training epochs.

  • batch_size (int) – The number of training examples utilized used to make one paramenters updat.

Returns:

None

validation(batch_size=32)[source]#

Tests the model performance using the validation dataset and calculates the accuracy.

Parameters:

batch_size (int) – The number of training examples utilized used to make one paramenters updat.

Returns:

None

infer(string)[source]#

Performs inference using the model.

Parameters:

string (str) – The input string to perform inference on.

Returns:

A random response from the response list of the predicted label.

save(path)[source]#

Saves the model parameters to the given path.

Parameters:

path (str) – The path where the model parameters will be saved.

Returns:

None

load(path)[source]#

Loads the model parameters from the given path.

Parameters:

path (str) – The path from where the model parameters will be loaded.

Returns:

None

Module contents#