| /external/tensorflow/tensorflow/python/keras/layers/ |
| lstm_test.py | 39 embedding_dim = 4
45 input_shape=(num_samples, timesteps, embedding_dim))
50 embedding_dim = 4
54 inputs = keras.layers.Dense(embedding_dim,
55 input_shape=(timesteps, embedding_dim))
65 embedding_dim = 4
67 layer = keras.layers.LSTM(units, input_shape=(None, embedding_dim))
73 x = np.random.random((num_samples, timesteps, embedding_dim))
80 embedding_dim = 4
87 input_shape=(num_samples, timesteps, embedding_dim))
[all...] |
| gru_test.py | 38 embedding_dim = 4
44 input_shape=(num_samples, timesteps, embedding_dim))
49 embedding_dim = 4
51 layer = keras.layers.GRU(units, input_shape=(None, embedding_dim))
56 x = np.random.random((num_samples, timesteps, embedding_dim))
63 embedding_dim = 4
70 input_shape=(num_samples, timesteps, embedding_dim))
76 embedding_dim = 4
82 input_shape=(num_samples, timesteps, embedding_dim))
87 embedding_dim =
[all...] |
| simplernn_test.py | 37 embedding_dim = 4
43 input_shape=(num_samples, timesteps, embedding_dim))
48 embedding_dim = 4
50 layer = keras.layers.SimpleRNN(units, input_shape=(None, embedding_dim))
54 x = np.random.random((num_samples, timesteps, embedding_dim))
61 embedding_dim = 4
68 input_shape=(num_samples, timesteps, embedding_dim))
73 embedding_dim = 4
80 input_shape=(num_samples, timesteps, embedding_dim))
83 embedding_dim =
[all...] |
| lstm_v2_test.py | 83 embedding_dim = 4
88 embedding_dim, input_shape=(timesteps, embedding_dim))
98 embedding_dim = 4
100 layer = rnn.LSTM(units, input_shape=(None, embedding_dim))
104 x = np.random.random((num_samples, timesteps, embedding_dim))
130 embedding_dim = 4
135 inputs = keras.Input((timesteps, embedding_dim))
149 inputs = np.random.random((num_samples, timesteps, embedding_dim))
159 embedding_dim =
[all...] |
| recurrent_v2_test.py | 44 embedding_dim = 10
55 keras.layers.Embedding(vocab_size, embedding_dim,
|
| gru_v2_test.py | 111 embedding_dim = 4
113 layer = rnn.GRU(units, input_shape=(None, embedding_dim))
117 x = np.random.random((num_samples, timesteps, embedding_dim))
332 embedding_dim = 4
338 input_shape=(num_samples, timesteps, embedding_dim))
358 embedding_dim = 4
365 input_shape=(num_samples, timesteps, embedding_dim))
368 embedding_dim = 4
377 input_shape=(None, embedding_dim),
381 layer.build((None, None, embedding_dim))
[all...] |
| recurrent_test.py | 284 embedding_dim = 4
288 x = keras.Input((time_step, embedding_dim))
295 embedding_dim)).as_list(),
308 np.zeros((batch, time_step, embedding_dim)),
312 x = keras.Input((time_step, embedding_dim))
328 np.zeros((batch, time_step, embedding_dim)),
332 x = keras.Input((time_step, embedding_dim))
345 np.zeros((batch, time_step, embedding_dim)),
349 x = keras.Input((time_step, embedding_dim))
359 np.zeros((batch, time_step, embedding_dim)),
[all...] |
| /external/libtextclassifier/lang_id/common/ |
| embedding-network.cc | 159 const int embedding_dim = embedding_matrix.cols; local
167 int feature_offset = concat_offset + feature_type->base() * embedding_dim;
168 SAFTM_CHECK_LE(feature_offset + embedding_dim, concat->size());
201 for (int i = 0; i < embedding_dim; ++i, ++weights, ++concat_ptr) {
210 for (int i = 0; i < embedding_dim;
222 for (int i = 0; i < embedding_dim / 2; ++i, ++quant_weights) {
|
| /external/tensorflow/tensorflow/contrib/eager/python/examples/rnn_ptb/ |
| rnn_ptb.py | 81 def __init__(self, vocab_size, embedding_dim, **kwargs):
84 self.embedding_dim = embedding_dim
89 shape=[self.vocab_size, self.embedding_dim],
112 embedding_dim,
121 self.embedding = Embedding(vocab_size, embedding_dim)
269 embedding_dim=200,
280 embedding_dim=650,
291 embedding_dim=20,
314 model = PTBModel(corpus.vocab_size(), FLAGS.embedding_dim,
[all...] |
| /external/tensorflow/tensorflow/python/training/ |
| checkpoint_ops.py | 422 embedding_dim,
442 embedding_dim: `int` specifying the dimension of the embedding vectors from
471 stddev=1.0 / math.sqrt(embedding_dim))
477 new_col_vocab_size=embedding_dim,
|
| checkpoint_ops_test.py | 275 embedding_dim=16,
320 embedding_dim=16,
359 embedding_dim=16,
|
| /external/tensorflow/tensorflow/contrib/seq2seq/python/kernel_tests/ |
| beam_search_decoder_test.py | 490 embedding_dim = 50
497 embedding = np.random.randn(vocab_size, embedding_dim).astype(np.float32)
604 embedding_dim = 50
611 embedding = np.random.randn(vocab_size, embedding_dim).astype(np.float32)
|
| attention_wrapper_v2_test.py | 133 embedding_dim = 6
136 vocab, embedding_dim, mask_zero=True)(
|
| /external/tensorflow/tensorflow/python/keras/ |
| model_subclassing_test.py | 252 def __init__(self, vocab_size, embedding_dim, **kwargs):
255 self.embedding_dim = embedding_dim
260 shape=[self.vocab_size, self.embedding_dim],
[all...] |
| /external/tensorflow/tensorflow/python/grappler/ |
| hierarchical_controller.py | 555 embedding_dim = array_ops.shape(input_layer)[2]
558 [batch_size * self.num_ops, embedding_dim])
[all...] |
