def test_normal_like(): """Tests just running the normal likelihood.""" n_train_sequences = 200 n_test_sequences = 50 n_frames = 20 n_channels = 3 train_sequences, test_seqences = prototype.generate_zero_dataset( n_train_sequences, n_test_sequences, n_frames, n_channels) params = prototype.normal_init(n_channels) prototype.normal_likelihood_sequences(train_sequences, params)
def test_simple_train(): """Tests a simple training function.""" n_train_sequences = 200 n_test_sequences = 50 n_frames = 20 n_channels = 3 train_sequences, test_seqences = prototype.generate_zero_dataset( n_train_sequences, n_test_sequences, n_frames, n_channels) n_nodes = 32 rnn = prototype.rnn_init(n_channels, n_nodes) test_likelihood = prototype.rnn_test_likelihood(test_sequences, rnn)
def test_simple_train(): """Tests a simple training function.""" n_train_sequences = 200 n_test_sequences = 50 n_frames = 20 n_channels = 3 train_sequences, test_seqences = prototype.generate_zero_dataset( n_train_sequences, n_test_sequences, n_frames, n_channels) n_nodes = 32 rnn = prototype.rnn_init(n_channels, n_nodes) test_likelihood = prototype.rnn_test_likelihood( test_sequences, rnn)