|
activation_gelu()
|
Gelu |
|
activation_hardshrink()
|
Hardshrink |
|
activation_lisht()
|
Lisht |
|
activation_mish()
|
Mish |
|
activation_rrelu()
|
Rrelu |
|
activation_softshrink()
|
Softshrink |
|
activation_sparsemax()
|
Sparsemax |
|
activation_tanhshrink()
|
Tanhshrink |
|
attention_bahdanau()
|
Bahdanau Attention |
|
attention_bahdanau_monotonic()
|
Bahdanau Monotonic Attention |
|
attention_luong()
|
Implements Luong-style (multiplicative) attention scoring. |
|
attention_luong_monotonic()
|
Monotonic attention mechanism with Luong-style energy function. |
|
attention_monotonic()
|
Monotonic attention |
|
attention_wrapper()
|
Attention Wrapper |
|
attention_wrapper_state()
|
Attention Wrapper State |
|
callback_average_model_checkpoint()
|
Average Model Checkpoint |
|
callback_time_stopping()
|
Time Stopping |
|
callback_tqdm_progress_bar()
|
TQDM Progress Bar |
|
crf_binary_score()
|
CRF binary score |
|
crf_decode()
|
CRF decode |
|
crf_decode_backward()
|
CRF decode backward |
|
crf_decode_forward()
|
CRF decode forward |
|
crf_forward()
|
CRF forward |
|
crf_log_likelihood()
|
CRF log likelihood |
|
crf_log_norm()
|
CRF log norm |
|
crf_multitag_sequence_score()
|
CRF multitag sequence score |
|
crf_sequence_score()
|
CRF sequence score |
|
crf_unary_score()
|
CRF unary score |
|
decode_dynamic()
|
Dynamic decode |
|
decoder()
|
An RNN Decoder abstract interface object. |
|
decoder_base()
|
Base Decoder |
|
decoder_basic()
|
Basic Decoder |
|
decoder_basic_output()
|
Basic decoder output |
|
decoder_beam_search()
|
BeamSearch sampling decoder |
|
decoder_beam_search_output()
|
Beam Search Decoder Output |
|
decoder_beam_search_state()
|
Beam Search Decoder State |
|
decoder_final_beam_search_output()
|
Final Beam Search Decoder Output |
|
extend_with_decoupled_weight_decay()
|
Factory function returning an optimizer class with decoupled weight decay |
|
gather_tree()
|
Gather tree |
|
gather_tree_from_array()
|
Gather tree from array |
|
hardmax()
|
Hardmax |
|
img_adjust_hsv_in_yiq()
|
Adjust hsv in yiq |
|
img_angles_to_projective_transforms()
|
Angles to projective transforms |
|
img_blend()
|
Blend |
|
img_compose_transforms()
|
Compose transforms |
|
img_connected_components()
|
Connected components |
|
img_cutout()
|
Cutout |
|
img_dense_image_warp()
|
Dense image warp |
|
img_equalize()
|
Equalize |
|
img_euclidean_dist_transform()
|
Euclidean dist transform |
|
img_flat_transforms_to_matrices()
|
Flat transforms to matrices |
|
img_from_4D()
|
From 4D image |
|
img_get_ndims()
|
Get ndims |
|
img_interpolate_bilinear()
|
Interpolate bilinear |
|
img_interpolate_spline()
|
Interpolate spline |
|
img_matrices_to_flat_transforms()
|
Matrices to flat transforms |
|
img_mean_filter2d()
|
Mean filter2d |
|
img_median_filter2d()
|
Median filter2d |
|
img_random_cutout()
|
Random cutout |
|
img_random_hsv_in_yiq()
|
Random hsv in yiq |
|
img_resampler()
|
Resampler |
|
img_rotate()
|
Rotate |
|
img_sharpness()
|
Sharpness |
|
img_shear_x()
|
Shear x-axis |
|
img_shear_y()
|
Shear y-axis |
|
img_sparse_image_warp()
|
Sparse image warp |
|
img_to_4D()
|
To 4D image |
|
img_transform()
|
Transform |
|
img_translate()
|
Translate |
|
img_translate_xy()
|
Translate xy dims |
|
img_translations_to_projective_transforms()
|
Translations to projective transforms |
|
img_unwrap()
|
Uwrap |
|
img_wrap()
|
Wrap |
|
install_tfaddons()
|
Install TensorFlow SIG Addons |
|
layer_activation_gelu()
|
Gaussian Error Linear Unit |
|
layer_correlation_cost()
|
Correlation Cost Layer. |
|
layer_filter_response_normalization()
|
FilterResponseNormalization |
|
layer_group_normalization()
|
Group normalization layer |
|
layer_instance_normalization()
|
Instance normalization layer |
|
layer_maxout()
|
Maxout layer |
|
layer_multi_head_attention()
|
Keras-based multi head attention layer |
|
layer_nas_cell()
|
Neural Architecture Search (NAS) recurrent network cell. |
|
layer_norm_lstm_cell()
|
LSTM cell with layer normalization and recurrent dropout. |
|
layer_poincare_normalize()
|
Project into the Poincare ball with norm <= 1.0 - epsilon |
|
layer_sparsemax()
|
Sparsemax activation function |
|
layer_weight_normalization()
|
Weight Normalization layer |
|
lookahead_mechanism()
|
Lookahead mechanism |
|
loss_contrastive()
|
Contrastive loss |
|
loss_giou()
|
Implements the GIoU loss function. |
|
loss_hamming()
|
Hamming loss |
|
loss_lifted_struct()
|
Lifted structured loss |
|
loss_npairs()
|
Npairs loss |
|
loss_npairs_multilabel()
|
Npairs multilabel loss |
|
loss_pinball()
|
Pinball loss |
|
loss_sequence()
|
Weighted cross-entropy loss for a sequence of logits. |
|
loss_sigmoid_focal_crossentropy()
|
Sigmoid focal crossentropy loss |
|
loss_sparsemax()
|
Sparsemax loss |
|
loss_triplet_hard()
|
Triplet hard loss |
|
loss_triplet_semihard()
|
Triplet semihard loss |
|
metric_cohen_kappa()
|
Computes Kappa score between two raters |
|
metric_fbetascore()
|
FBetaScore |
|
metric_hamming_distance()
|
Hamming distance |
|
metric_mcc()
|
MatthewsCorrelationCoefficient |
|
metric_multilabel_confusion_matrix()
|
MultiLabelConfusionMatrix |
|
metric_rsquare()
|
RSquare
This is also called as coefficient of determination. It tells how close
are data to the fitted regression line. Highest score can be 1.0 and it
indicates that the predictors perfectly accounts for variation in the target.
Score 0.0 indicates that the predictors do not account for variation in the
target. It can also be negative if the model is worse. |
|
metrics_f1score()
|
F1Score |
|
optimizer_conditional_gradient()
|
Conditional Gradient |
|
optimizer_decay_adamw()
|
Optimizer that implements the Adam algorithm with weight decay |
|
optimizer_decay_sgdw()
|
Optimizer that implements the Momentum algorithm with weight_decay |
|
optimizer_lamb()
|
Layer-wise Adaptive Moments |
|
optimizer_lazy_adam()
|
Lazy Adam |
|
optimizer_moving_average()
|
Moving Average |
|
optimizer_novograd()
|
NovoGrad |
|
optimizer_radam()
|
Rectified Adam (a.k.a. RAdam) |
|
optimizer_swa()
|
Stochastic Weight Averaging |
|
optimizer_yogi()
|
Yogi |
|
parse_time()
|
Parse time |
|
register_all()
|
Register all |
|
register_custom_kernels()
|
Register custom kernels |
|
register_keras_objects()
|
Register keras objects |
|
safe_cumprod()
|
Safe cumprod |
|
sample_bernoulli()
|
Bernoulli sample |
|
sample_categorical()
|
Categorical sample |
|
sampler()
|
Sampler |
|
sampler_custom()
|
Base abstract class that allows the user to customize sampling. |
|
sampler_greedy_embedding()
|
Greedy Embedding Sampler |
|
sampler_inference()
|
Inference Sampler |
|
sampler_sample_embedding()
|
Sample Embedding Sampler |
|
sampler_scheduled_embedding_training()
|
A training sampler that adds scheduled sampling |
|
sampler_scheduled_output_training()
|
Scheduled Output Training Sampler |
|
sampler_training()
|
A Sampler for use during training. |
|
skip_gram_sample()
|
Skip gram sample |
|
skip_gram_sample_with_text_vocab()
|
Skip gram sample with text vocab |
|
tfaddons_version()
|
Version of TensorFlow SIG Addons |
|
tile_batch()
|
Tile batch |
|
viterbi_decode()
|
Viterbi decode |