hub / github.com/benoitsteiner/tensorflow-opencl / LSTMStateTuple

Class LSTMStateTuple

tensorflow/python/ops/rnn_cell_impl.py:397–413 · view source on GitHub ↗

Tuple used by LSTM Cells for `state_size`, `zero_state`, and output state. Stores two elements: `(c, h)`, in that order. Where `c` is the hidden state and `h` is the output. Only used when `state_is_tuple=True`.

Source from the content-addressed store, hash-verified

source not stored for this graph (policy: none)

Callers 15

testBahdanauNotNormalizedMethod · 0.85

testBahdanauNormalizedMethod · 0.85

testLuongNotNormalizedMethod · 0.85

testLuongScaledMethod · 0.85

testNotUseAttentionLayerMethod · 0.85

testBahdanauMonotonicNotNormalizedMethod · 0.85

testBahdanauMonotonicNormalizedMethod · 0.85

testLuongMonotonicNotNormalizedMethod · 0.85

testLuongMonotonicScaledMethod · 0.85

testMultiAttentionMethod · 0.85

testMultiAttentionNoAttentionLayerMethod · 0.85

testSingleAttentionAsListMethod · 0.85

Calls

no outgoing calls

Tested by 12

testBahdanauNotNormalizedMethod · 0.68

testBahdanauNormalizedMethod · 0.68

testLuongNotNormalizedMethod · 0.68

testLuongScaledMethod · 0.68

testNotUseAttentionLayerMethod · 0.68

testBahdanauMonotonicNotNormalizedMethod · 0.68

testBahdanauMonotonicNormalizedMethod · 0.68

testLuongMonotonicNotNormalizedMethod · 0.68

testLuongMonotonicScaledMethod · 0.68

testMultiAttentionMethod · 0.68

testMultiAttentionNoAttentionLayerMethod · 0.68

testSingleAttentionAsListMethod · 0.68