ontolearn.clip_architectures
Classes
LSTM architecture |
|
GRU architecture |
|
CNN architecture |
|
SetTransformer architecture. |
Module Contents
- class ontolearn.clip_architectures.LengthLearner_LSTM(input_size, output_size, proj_dim=256, rnn_n_layers=2, drop_prob=0.2)[source]
Bases:
ontolearn.nces_modules.nn.Module
LSTM architecture
- name = 'LSTM'
- loss
- lstm
- dropout
- fc1
- fc2
- fc3
- class ontolearn.clip_architectures.LengthLearner_GRU(input_size, output_size, proj_dim=256, rnn_n_layers=2, drop_prob=0.2)[source]
Bases:
ontolearn.nces_modules.nn.Module
GRU architecture
- name = 'GRU'
- loss
- gru
- dropout
- fc1
- fc2
- fc3
- class ontolearn.clip_architectures.LengthLearner_CNN(input_size, output_size, num_examples, proj_dim=256, kernel_size: list = [[5, 7], [5, 7]], stride: list = [[3, 3], [3, 3]], drop_prob=0.2)[source]
Bases:
ontolearn.nces_modules.nn.Module
CNN architecture
- name = 'CNN'
- loss
- conv1
- conv2
- dropout1d
- dropout2d
- fc1
- fc2
- fc3