{ "_id": "6a12a797acfb0bcc41d14bf8", "Package": "keras3", "Type": "Package", "Title": "R Interface to 'Keras'", "Version": "1.5.1.9000", "Authors@R": "c(\nperson(\"Tomasz\", \"Kalinowski\", role = c(\"aut\", \"cph\", \"cre\"),\nemail = \"tomasz@posit.co\"),\nperson(\"Daniel\", \"Falbel\", role = c(\"ctb\", \"cph\"), email = \"daniel@posit.co\"),\nperson(\"JJ\", \"Allaire\", role = c(\"aut\", \"cph\")),\nperson(\"François\", \"Chollet\", role = c(\"aut\", \"cph\")),\nperson(\"Posit Software, PBC\", role = c(\"cph\", \"fnd\")),\nperson(\"Google\", role = c(\"cph\", \"fnd\")),\nperson(\"Yuan\", \"Tang\", role = c(\"ctb\", \"cph\"),\nemail = \"terrytangyuan@gmail.com\",\ncomment = c(ORCID = \"0000-0001-5243-233X\")),\nperson(\"Wouter\", \"Van Der Bijl\", role = c(\"ctb\", \"cph\")),\nperson(\"Martin\", \"Studer\", role = c(\"ctb\", \"cph\")),\nperson(\"Sigrid\", \"Keydana\", role = c(\"ctb\"))\n)", "Description": "Interface to 'Keras' , a high-level\nneural networks API. 'Keras' was developed with a focus on\nenabling fast experimentation, supports both convolution based\nnetworks and recurrent networks (as well as combinations of the\ntwo), and runs seamlessly on both CPU and GPU devices.", "Encoding": "UTF-8", "License": "MIT + file LICENSE", "URL": "https://keras3.posit.co/, https://github.com/rstudio/keras3", "BugReports": "https://github.com/rstudio/keras3/issues", "RoxygenNote": "7.3.3", "VignetteBuilder": "knitr", "Config/pak/sysreqs": "libpng-dev", "Repository": "https://rstudio.r-universe.dev", "Date/Publication": "2026-04-13 18:15:38 UTC", "RemoteUrl": "https://github.com/rstudio/keras3", "RemoteRef": "HEAD", "RemoteSha": "4571bff1befc425ee812faf247e3732ffc510607", "NeedsCompilation": "no", "Packaged": { "Date": "2026-05-24 07:10:41 UTC", "User": "root" }, "Author": "Tomasz Kalinowski [aut, cph, cre],\nDaniel Falbel [ctb, cph],\nJJ Allaire [aut, cph],\nFrançois Chollet [aut, cph],\nPosit Software, PBC [cph, fnd],\nGoogle [cph, fnd],\nYuan Tang [ctb, cph] (ORCID: ),\nWouter Van Der Bijl [ctb, cph],\nMartin Studer [ctb, cph],\nSigrid Keydana [ctb]", "Maintainer": "Tomasz Kalinowski ", "MD5sum": "2169dccc0e2e80715c7173ff78a65c2f", "_user": "rstudio", "_type": "src", "_file": "keras3_1.5.1.9000.tar.gz", "_fileid": "0cbbce7edf742d7f94b9669d3f2ae30989bc37d72f229ffe382aa28564b12760", "_filesize": 15588212, "_sha256": "0cbbce7edf742d7f94b9669d3f2ae30989bc37d72f229ffe382aa28564b12760", "_created": "2026-05-24T07:10:41.000Z", "_published": "2026-05-24T07:24:07.798Z", "_distro": "noble", "_jobs": [ { "job": 77579841153, "time": 282, "config": "linux-devel-x86_64", "r": "4.7.0", "check": "OK", "artifact": "7183218535" }, { "job": 77579841149, "time": 267, "config": "linux-release-x86_64", "r": "4.6.0", "check": "OK", "artifact": "7183216599" }, { "job": 77579841148, "time": 170, "config": "macos-oldrel-arm64", "r": "4.5.3", "check": "OK", "artifact": "7183232604" }, { "job": 77579841152, "time": 212, "config": "macos-release-arm64", "r": "4.6.0", "check": "OK", "artifact": "7183232642" }, { "job": 77579296092, "time": 592, "config": "source", "r": "4.6.0", "check": "OK", "artifact": "7183183693" }, { "job": 77579841150, "time": 145, "config": "wasm-release", "r": "4.6.0", "check": "OK", "artifact": "7183202043" }, { "job": 77579841156, "time": 249, "config": "windows-devel", "r": "4.7.0", "check": "OK", "artifact": "7183214741" }, { "job": 77579841158, "time": 360, "config": "windows-oldrel", "r": "4.5.3", "check": "OK", "artifact": "7183227297" }, { "job": 77579841160, "time": 293, "config": "windows-release", "r": "4.6.0", "check": "OK", "artifact": "7183219834" } ], "_buildurl": "https://github.com/r-universe/rstudio/actions/runs/26354737481", "_status": "success", "_host": "GitHub-Actions", "_upstream": "https://github.com/rstudio/keras3", "_commit": { "id": "4571bff1befc425ee812faf247e3732ffc510607", "author": "Tomasz Kalinowski ", "committer": "Tomasz Kalinowski ", "message": "Increment version number to 1.5.1.9000\n", "time": 1776104138 }, "_maintainer": { "name": "Tomasz Kalinowski", "email": "tomasz@posit.co", "login": "t-kalinowski", "mastodon": "@t_kalinowski@fosstodon.org", "bluesky": "@t-kalinowski.bsky.social", "description": "", "uuid": 8462255 }, "_registered": true, "_dependencies": [ { "package": "R", "version": ">= 4.0", "role": "Depends" }, { "package": "generics", "version": ">= 0.0.1", "role": "Imports" }, { "package": "reticulate", "version": ">= 1.42.0", "role": "Imports" }, { "package": "tensorflow", "version": ">= 2.16.0", "role": "Imports" }, { "package": "tfruns", "version": ">= 1.5.2", "role": "Imports" }, { "package": "magrittr", "role": "Imports" }, { "package": "zeallot", "role": "Imports" }, { "package": "fastmap", "role": "Imports" }, { "package": "glue", "role": "Imports" }, { "package": "cli", "role": "Imports" }, { "package": "rlang", "role": "Imports" }, { "package": "dotty", "role": "Imports" }, { "package": "ggplot2", "role": "Suggests" }, { "package": "testthat", "version": ">= 2.1.0", "role": "Suggests" }, { "package": "knitr", "role": "Suggests" }, { "package": "rmarkdown", "role": "Suggests" }, { "package": "callr", "role": "Suggests" }, { "package": "tfdatasets", "role": "Suggests" }, { "package": "withr", "role": "Suggests" }, { "package": "png", "role": "Suggests" }, { "package": "jsonlite", "role": "Suggests" }, { "package": "purrr", "role": "Suggests" }, { "package": "rstudioapi", "role": "Suggests" }, { "package": "R6", "role": "Suggests" }, { "package": "jpeg", "role": "Suggests" } ], "_owner": "rstudio", "_selfowned": true, "_usedby": 6, "_updates": [ { "week": "2025-24", "n": 3 }, { "week": "2025-29", "n": 1 }, { "week": "2025-40", "n": 1 }, { "week": "2025-41", "n": 6 }, { "week": "2025-42", "n": 1 }, { "week": "2025-50", "n": 2 }, { "week": "2026-05", "n": 1 }, { "week": "2026-07", "n": 2 }, { "week": "2026-11", "n": 1 }, { "week": "2026-16", "n": 2 } ], "_tags": [ { "name": "v2.16.0", "date": "2025-08-18" }, { "name": "v1.5.0", "date": "2025-12-21" }, { "name": "v1.5.1", "date": "2026-02-13" }, { "name": "v1.5.0.9000", "date": "2026-03-10" }, { "name": "v2.16.1", "date": "2026-04-13" } ], "_stars": 855, "_contributors": [ { "user": "t-kalinowski", "count": 2536, "uuid": 8462255 }, { "user": "jjallaire", "count": 1440, "uuid": 104391 }, { "user": "dfalbel", "count": 639, "uuid": 4706822 }, { "user": "javierluraschi", "count": 90, "uuid": 3478847 }, { "user": "christophersparling", "count": 19, "uuid": 17165276 }, { "user": "turgut090", "count": 16, "uuid": 38523699 }, { "user": "ax3man", "count": 7, "uuid": 15309336 }, { "user": "kevinykuo", "count": 7, "uuid": 5582151 }, { "user": "atroiano", "count": 6, "uuid": 15947353 }, { "user": "topepo", "count": 5, "uuid": 5731043 }, { "user": "statist-bhfz", "count": 5, "uuid": 7324186 }, { "user": "ivokwee", "count": 4, "uuid": 19613146 }, { "user": "rajshah4", "count": 4, "uuid": 6808012 }, { "user": "gsteinbu", "count": 3, "uuid": 15330732 }, { "user": "terrytangyuan", "count": 3, "uuid": 4269898 }, { "user": "16eagle", "count": 3, "uuid": 23257860 }, { "user": "jspiewak", "count": 3, "uuid": 399516 }, { "user": "brianrice2", "count": 2, "uuid": 51218458 }, { "user": "edoffagne", "count": 2, "uuid": 4750263 }, { "user": "cdk29", "count": 2, "uuid": 22153296 }, { "user": "osorensen", "count": 1, "uuid": 21175639 }, { "user": "tomazweiss", "count": 1, "uuid": 38002502 }, { "user": "rsangole", "count": 1, "uuid": 6344050 }, { "user": "muschellij2", "count": 1, "uuid": 1075118 }, { "user": "fangzhou-xie", "count": 1, "uuid": 45321577 }, { "user": "edward22", "count": 1, "uuid": 16159694 }, { "user": "pitakakariki", "count": 1, "uuid": 1023705 }, { "user": "michaelchirico", "count": 1, "uuid": 7606389 }, { "user": "martinstuder", "count": 1, "uuid": 1091843 }, { "user": "jemus42", "count": 1, "uuid": 1613346 }, { "user": "yulijia", "count": 1, "uuid": 895125 }, { "user": "janlauge", "count": 1, "uuid": 14994776 }, { "user": "jrosell", "count": 1, "uuid": 1544992 }, { "user": "ifrit98", "count": 1, "uuid": 31426574 }, { "user": "jamesallingham", "count": 1, "uuid": 4865998 }, { "user": "gdequeiroz", "count": 1, "uuid": 1934546 }, { "user": "eddelbuettel", "count": 1, "uuid": 673121 }, { "user": "davidarenburg", "count": 1, "uuid": 7597389 }, { "user": "davechilders", "count": 1, "uuid": 7317563 }, { "user": "benwiseman", "count": 1, "uuid": 11344964 }, { "user": "andrie", "count": 1, "uuid": 479998 }, { "user": "ax42", "count": 1, "uuid": 1254869 } ], "_userbio": { "uuid": 513560, "type": "organization", "name": "RStudio" }, "_downloads": { "count": 10753, "source": "https://cranlogs.r-pkg.org/downloads/total/last-month/keras3" }, "_devurl": "https://github.com/rstudio/keras3", "_pkgdown": "https://keras3.posit.co/", "_searchresults": 1036, "_rbuild": "4.6.0", "_assets": [ "extra/citation.cff", "extra/citation.html", "extra/citation.json", "extra/citation.txt", "extra/contents.json", "extra/keras3.html", "extra/NEWS.html", "extra/NEWS.txt", "extra/readme.html", "extra/readme.md", "manual.pdf" ], "_homeurl": "https://github.com/rstudio/keras3", "_realowner": "rstudio", "_cranurl": true, "_releases": [ { "version": "0.1.0", "date": "2024-02-17" }, { "version": "0.2.0", "date": "2024-04-18" }, { "version": "1.0.0", "date": "2024-05-21" }, { "version": "1.1.0", "date": "2024-07-17" }, { "version": "1.2.0", "date": "2024-09-05" }, { "version": "1.3.0", "date": "2025-03-03" }, { "version": "1.4.0", "date": "2025-05-04" }, { "version": "1.5.0", "date": "2025-12-22" }, { "version": "1.5.1", "date": "2026-02-16" } ], "_exports": [ ".", "%<-%", "%<-active%", "%<>%", "%>%", "%as%", "%py_class%", "activation_celu", "activation_elu", "activation_exponential", "activation_gelu", "activation_glu", "activation_hard_shrink", "activation_hard_sigmoid", "activation_hard_silu", "activation_hard_swish", "activation_hard_tanh", "activation_leaky_relu", "activation_linear", "activation_log_sigmoid", "activation_log_softmax", "activation_mish", "activation_relu", "activation_relu6", "activation_selu", "activation_sigmoid", "activation_silu", "activation_soft_shrink", "activation_softmax", "activation_softplus", "activation_softsign", "activation_sparse_plus", "activation_sparse_sigmoid", "activation_sparsemax", "activation_squareplus", "activation_tanh", "activation_tanh_shrink", "activation_threshold", "active_property", "adapt", "all_dims", "application_convnext_base", "application_convnext_large", "application_convnext_small", "application_convnext_tiny", "application_convnext_xlarge", "application_decode_predictions", "application_densenet121", "application_densenet169", "application_densenet201", "application_efficientnet_b0", "application_efficientnet_b1", "application_efficientnet_b2", "application_efficientnet_b3", "application_efficientnet_b4", "application_efficientnet_b5", "application_efficientnet_b6", "application_efficientnet_b7", "application_efficientnet_v2b0", "application_efficientnet_v2b1", "application_efficientnet_v2b2", "application_efficientnet_v2b3", "application_efficientnet_v2l", "application_efficientnet_v2m", "application_efficientnet_v2s", "application_inception_resnet_v2", "application_inception_v3", "application_mobilenet", "application_mobilenet_v2", "application_mobilenet_v3_large", "application_mobilenet_v3_small", "application_nasnet_large", "application_nasnet_mobile", "application_nasnetlarge", "application_nasnetmobile", "application_preprocess_inputs", "application_resnet101", "application_resnet101_v2", "application_resnet152", "application_resnet152_v2", "application_resnet50", "application_resnet50_v2", "application_vgg16", "application_vgg19", "application_xception", "array_reshape", "as_iterator", "as_tensor", "audio_dataset_from_directory", "bidirectional", "Callback", "callback_backup_and_restore", "callback_csv_logger", "callback_early_stopping", "callback_lambda", "callback_learning_rate_scheduler", "callback_model_checkpoint", "callback_reduce_lr_on_plateau", "callback_remote_monitor", "callback_swap_ema_weights", "callback_tensorboard", "callback_terminate_on_nan", "clear_session", "clone_model", "compile", "config_backend", "config_disable_flash_attention", "config_disable_interactive_logging", "config_disable_traceback_filtering", "config_dtype_policy", "config_enable_flash_attention", "config_enable_interactive_logging", "config_enable_traceback_filtering", "config_enable_unsafe_deserialization", "config_epsilon", "config_floatx", "config_image_data_format", "config_is_flash_attention_enabled", "config_is_interactive_logging_enabled", "config_is_nnx_enabled", "config_is_traceback_filtering_enabled", "config_max_epochs", "config_max_steps_per_epoch", "config_set_backend", "config_set_dtype_policy", "config_set_epsilon", "config_set_floatx", "config_set_image_data_format", "config_set_max_epochs", "config_set_max_steps_per_epoch", "Constraint", "constraint_maxnorm", "constraint_minmaxnorm", "constraint_nonneg", "constraint_unitnorm", "count_params", "custom_metric", "dataset_boston_housing", "dataset_california_housing", "dataset_cifar10", "dataset_cifar100", "dataset_fashion_mnist", "dataset_imdb", "dataset_imdb_word_index", "dataset_mnist", "dataset_reuters", "dataset_reuters_word_index", "deserialize_keras_object", "evaluate", "export_savedmodel", "feature_cross", "feature_custom", "feature_float", "feature_float_discretized", "feature_float_normalized", "feature_float_rescaled", "feature_integer_categorical", "feature_integer_hashed", "feature_string_categorical", "feature_string_hashed", "fit", "flag_boolean", "flag_integer", "flag_numeric", "flag_string", "flags", "freeze_weights", "from_config", "generate", "get_config", "get_custom_objects", "get_file", "get_layer", "get_registered_name", "get_registered_object", "get_source_inputs", "get_state_tree", "get_vocabulary", "get_weights", "image_array_save", "image_dataset_from_directory", "image_from_array", "image_load", "image_smart_resize", "image_to_array", "imagenet_decode_predictions", "imagenet_preprocess_input", "import", "initializer_constant", "initializer_glorot_normal", "initializer_glorot_uniform", "initializer_he_normal", "initializer_he_uniform", "initializer_identity", "initializer_lecun_normal", "initializer_lecun_uniform", "initializer_ones", "initializer_orthogonal", "initializer_random_normal", "initializer_random_uniform", "initializer_stft", "initializer_truncated_normal", "initializer_variance_scaling", "initializer_zeros", "install_keras", "iter_next", "iterate", "keras", "keras_input", "keras_model", "keras_model_sequential", "keras_variable", "Layer", "layer_activation", "layer_activation_elu", "layer_activation_leaky_relu", "layer_activation_parametric_relu", "layer_activation_relu", "layer_activation_softmax", "layer_activity_regularization", "layer_add", "layer_additive_attention", "layer_alpha_dropout", "layer_attention", "layer_aug_mix", "layer_auto_contrast", "layer_average", "layer_average_pooling_1d", "layer_average_pooling_2d", "layer_average_pooling_3d", "layer_batch_normalization", "layer_bidirectional", "layer_category_encoding", "layer_center_crop", "layer_concatenate", "layer_conv_1d", "layer_conv_1d_transpose", "layer_conv_2d", "layer_conv_2d_transpose", "layer_conv_3d", "layer_conv_3d_transpose", "layer_conv_lstm_1d", "layer_conv_lstm_2d", "layer_conv_lstm_3d", "layer_cropping_1d", "layer_cropping_2d", "layer_cropping_3d", "layer_cut_mix", "layer_dense", "layer_depthwise_conv_1d", "layer_depthwise_conv_2d", "layer_discretization", "layer_dot", "layer_dropout", "layer_einsum_dense", "layer_embedding", "layer_equalization", "layer_feature_space", "layer_flatten", "layer_flax_module_wrapper", "layer_gaussian_dropout", "layer_gaussian_noise", "layer_global_average_pooling_1d", "layer_global_average_pooling_2d", "layer_global_average_pooling_3d", "layer_global_max_pooling_1d", "layer_global_max_pooling_2d", "layer_global_max_pooling_3d", "layer_group_normalization", "layer_group_query_attention", "layer_gru", "layer_hashed_crossing", "layer_hashing", "layer_identity", "layer_input", "layer_integer_lookup", "layer_jax_model_wrapper", "layer_lambda", "layer_layer_normalization", "layer_lstm", "layer_masking", "layer_max_num_bounding_boxes", "layer_max_pooling_1d", "layer_max_pooling_2d", "layer_max_pooling_3d", "layer_maximum", "layer_mel_spectrogram", "layer_minimum", "layer_mix_up", "layer_multi_head_attention", "layer_multiply", "layer_normalization", "layer_permute", "layer_pipeline", "layer_rand_augment", "layer_random_brightness", "layer_random_color_degeneration", "layer_random_color_jitter", "layer_random_contrast", "layer_random_crop", "layer_random_elastic_transform", "layer_random_erasing", "layer_random_flip", "layer_random_gaussian_blur", "layer_random_grayscale", "layer_random_hue", "layer_random_invert", "layer_random_perspective", "layer_random_posterization", "layer_random_rotation", "layer_random_saturation", "layer_random_sharpness", "layer_random_shear", "layer_random_translation", "layer_random_zoom", "layer_repeat_vector", "layer_rescaling", "layer_reshape", "layer_resizing", "layer_rms_normalization", "layer_rnn", "layer_separable_conv_1d", "layer_separable_conv_2d", "layer_simple_rnn", "layer_solarization", "layer_spatial_dropout_1d", "layer_spatial_dropout_2d", "layer_spatial_dropout_3d", "layer_spectral_normalization", "layer_stft_spectrogram", "layer_string_lookup", "layer_subtract", "layer_text_vectorization", "layer_tfsm", "layer_time_distributed", "layer_torch_module_wrapper", "layer_unit_normalization", "layer_upsampling_1d", "layer_upsampling_2d", "layer_upsampling_3d", "layer_zero_padding_1d", "layer_zero_padding_2d", "layer_zero_padding_3d", "learning_rate_schedule_cosine_decay", "learning_rate_schedule_cosine_decay_restarts", "learning_rate_schedule_exponential_decay", "learning_rate_schedule_inverse_time_decay", "learning_rate_schedule_piecewise_constant_decay", "learning_rate_schedule_polynomial_decay", "load_model", "load_model_config", "load_model_weights", "Loss", "loss_binary_crossentropy", "loss_binary_focal_crossentropy", "loss_categorical_crossentropy", "loss_categorical_focal_crossentropy", "loss_categorical_generalized_cross_entropy", "loss_categorical_hinge", "loss_circle", "loss_cosine_similarity", "loss_ctc", "loss_dice", "loss_hinge", "loss_huber", "loss_kl_divergence", "loss_log_cosh", "loss_mean_absolute_error", "loss_mean_absolute_percentage_error", "loss_mean_squared_error", "loss_mean_squared_logarithmic_error", "loss_poisson", "loss_sparse_categorical_crossentropy", "loss_squared_hinge", "loss_tversky", "mark_active", "Metric", "metric_auc", "metric_binary_accuracy", "metric_binary_crossentropy", "metric_binary_focal_crossentropy", "metric_binary_iou", "metric_categorical_accuracy", "metric_categorical_crossentropy", "metric_categorical_focal_crossentropy", "metric_categorical_hinge", "metric_concordance_correlation", "metric_cosine_similarity", "metric_f1_score", "metric_false_negatives", "metric_false_positives", "metric_fbeta_score", "metric_hinge", "metric_huber", "metric_iou", "metric_kl_divergence", "metric_log_cosh", "metric_log_cosh_error", "metric_mean", "metric_mean_absolute_error", "metric_mean_absolute_percentage_error", "metric_mean_iou", "metric_mean_squared_error", "metric_mean_squared_logarithmic_error", "metric_mean_wrapper", "metric_one_hot_iou", "metric_one_hot_mean_iou", "metric_pearson_correlation", "metric_poisson", "metric_precision", "metric_precision_at_recall", "metric_r2_score", "metric_recall", "metric_recall_at_precision", "metric_root_mean_squared_error", "metric_sensitivity_at_specificity", "metric_sparse_categorical_accuracy", "metric_sparse_categorical_crossentropy", "metric_sparse_top_k_categorical_accuracy", "metric_specificity_at_sensitivity", "metric_squared_hinge", "metric_sum", "metric_top_k_categorical_accuracy", "metric_true_negatives", "metric_true_positives", "Model", "named_list", "new_callback_class", "new_layer_class", "new_learning_rate_schedule_class", "new_loss_class", "new_metric_class", "new_model_class", "newaxis", "normalize", "np_array", "op_abs", "op_add", "op_all", "op_angle", "op_any", "op_append", "op_arange", "op_arccos", "op_arccosh", "op_arcsin", "op_arcsinh", "op_arctan", "op_arctan2", "op_arctanh", "op_argmax", "op_argmin", "op_argpartition", "op_argsort", "op_array", "op_associative_scan", "op_average", "op_average_pool", "op_bartlett", "op_batch_normalization", "op_binary_crossentropy", "op_bincount", "op_bitwise_and", "op_bitwise_invert", "op_bitwise_left_shift", "op_bitwise_not", "op_bitwise_or", "op_bitwise_right_shift", "op_bitwise_xor", "op_blackman", "op_broadcast_to", "op_cast", "op_categorical_crossentropy", "op_cbrt", "op_ceil", "op_celu", "op_cholesky", "op_clip", "op_concatenate", "op_cond", "op_conj", "op_conv", "op_conv_transpose", "op_convert_to_array", "op_convert_to_numpy", "op_convert_to_tensor", "op_copy", "op_corrcoef", "op_correlate", "op_cos", "op_cosh", "op_count_nonzero", "op_cross", "op_ctc_decode", "op_ctc_loss", "op_cumprod", "op_cumsum", "op_custom_gradient", "op_deg2rad", "op_depthwise_conv", "op_det", "op_diag", "op_diagflat", "op_diagonal", "op_diff", "op_digitize", "op_divide", "op_divide_no_nan", "op_dot", "op_dot_product_attention", "op_dtype", "op_eig", "op_eigh", "op_einsum", "op_elu", "op_empty", "op_equal", "op_erf", "op_erfinv", "op_exp", "op_exp2", "op_expand_dims", "op_expm1", "op_extract_sequences", "op_eye", "op_fft", "op_fft2", "op_flip", "op_floor", "op_floor_divide", "op_fori_loop", "op_full", "op_full_like", "op_gelu", "op_get_item", "op_glu", "op_greater", "op_greater_equal", "op_hamming", "op_hanning", "op_hard_shrink", "op_hard_sigmoid", "op_hard_silu", "op_hard_swish", "op_hard_tanh", "op_heaviside", "op_histogram", "op_hstack", "op_identity", "op_ifft2", "op_imag", "op_image_affine_transform", "op_image_crop", "op_image_elastic_transform", "op_image_extract_patches", "op_image_gaussian_blur", "op_image_hsv_to_rgb", "op_image_map_coordinates", "op_image_pad", "op_image_perspective_transform", "op_image_resize", "op_image_rgb_to_grayscale", "op_image_rgb_to_hsv", "op_in_top_k", "op_inner", "op_inv", "op_irfft", "op_is_tensor", "op_isclose", "op_isfinite", "op_isinf", "op_isnan", "op_istft", "op_kaiser", "op_layer_normalization", "op_leaky_relu", "op_left_shift", "op_less", "op_less_equal", "op_linspace", "op_log", "op_log_sigmoid", "op_log_softmax", "op_log10", "op_log1p", "op_log2", "op_logaddexp", "op_logdet", "op_logical_and", "op_logical_not", "op_logical_or", "op_logical_xor", "op_logspace", "op_logsumexp", "op_lstsq", "op_lu_factor", "op_map", "op_matmul", "op_max", "op_max_pool", "op_maximum", "op_mean", "op_median", "op_meshgrid", "op_min", "op_minimum", "op_mod", "op_moments", "op_moveaxis", "op_multi_hot", "op_multiply", "op_nan_to_num", "op_ndim", "op_negative", "op_nonzero", "op_norm", "op_normalize", "op_not_equal", "op_one_hot", "op_ones", "op_ones_like", "op_outer", "op_pad", "op_pmax", "op_pmin", "op_polar", "op_power", "op_prod", "op_psnr", "op_qr", "op_quantile", "op_ravel", "op_real", "op_rearrange", "op_reciprocal", "op_relu", "op_relu6", "op_repeat", "op_reshape", "op_rfft", "op_right_shift", "op_rms_normalization", "op_roll", "op_rot90", "op_round", "op_rsqrt", "op_saturate_cast", "op_scan", "op_scatter", "op_scatter_update", "op_searchsorted", "op_segment_max", "op_segment_sum", "op_select", "op_selu", "op_separable_conv", "op_shape", "op_sigmoid", "op_sign", "op_signbit", "op_silu", "op_sin", "op_sinh", "op_size", "op_slice", "op_slice_update", "op_slogdet", "op_soft_shrink", "op_softmax", "op_softplus", "op_softsign", "op_solve", "op_solve_triangular", "op_sort", "op_sparse_categorical_crossentropy", "op_sparse_plus", "op_sparse_sigmoid", "op_sparsemax", "op_split", "op_sqrt", "op_square", "op_squareplus", "op_squeeze", "op_stack", "op_std", "op_stft", "op_stop_gradient", "op_subset", "op_subset_set", "op_subset<-", "op_subtract", "op_sum", "op_svd", "op_swapaxes", "op_switch", "op_take", "op_take_along_axis", "op_tan", "op_tanh", "op_tanh_shrink", "op_tensordot", "op_threshold", "op_tile", "op_top_k", "op_trace", "op_transpose", "op_tri", "op_tril", "op_triu", "op_trunc", "op_unravel_index", "op_unstack", "op_var", "op_vdot", "op_vectorize", "op_vectorized_map", "op_view_as_complex", "op_view_as_real", "op_vstack", "op_where", "op_while_loop", "op_zeros", "op_zeros_like", "optimizer_adadelta", "optimizer_adafactor", "optimizer_adagrad", "optimizer_adam", "optimizer_adam_w", "optimizer_adamax", "optimizer_ftrl", "optimizer_lamb", "optimizer_lion", "optimizer_loss_scale", "optimizer_muon", "optimizer_nadam", "optimizer_rmsprop", "optimizer_sgd", "pad_sequences", "pop_layer", "predict_on_batch", "py_help", "py_require", "py_to_r", "quantize_weights", "r_to_py", "random_beta", "random_binomial", "random_categorical", "random_dropout", "random_gamma", "random_integer", "random_normal", "random_seed_generator", "random_shuffle", "random_truncated_normal", "random_uniform", "register_keras_serializable", "regularizer_l1", "regularizer_l1_l2", "regularizer_l2", "regularizer_orthogonal", "reset_state", "rnn_cell_gru", "rnn_cell_lstm", "rnn_cell_simple", "rnn_cells_stack", "run_dir", "save_model", "save_model_config", "save_model_weights", "serialize_keras_object", "set_custom_objects", "set_random_seed", "set_state_tree", "set_vocabulary", "set_weights", "shape", "split_dataset", "tensorboard", "test_on_batch", "text_dataset_from_directory", "time_distributed", "timeseries_dataset_from_array", "to_categorical", "train_on_batch", "tuple", "unfreeze_weights", "use_backend", "use_python", "use_virtualenv", "with_custom_object_scope", "zip_lists" ], "_help": [ { "page": "activation_celu", "title": "Continuously Differentiable Exponential Linear Unit.", "concept": [ "activations" ], "topics": [ "activation_celu" ] }, { "page": "activation_elu", "title": "Exponential Linear Unit.", "concept": [ "activations" ], "topics": [ "activation_elu" ] }, { "page": "activation_exponential", "title": "Exponential activation function.", "concept": [ "activations" ], "topics": [ "activation_exponential" ] }, { "page": "activation_gelu", "title": "Gaussian error linear unit (GELU) activation function.", "concept": [ "activations" ], "topics": [ "activation_gelu" ] }, { "page": "activation_glu", "title": "Gated Linear Unit (GLU) activation function.", "concept": [ "activations" ], "topics": [ "activation_glu" ] }, { "page": "activation_hard_shrink", "title": "Hard Shrink activation function.", "concept": [ "activations" ], "topics": [ "activation_hard_shrink" ] }, { "page": "activation_hard_sigmoid", "title": "Hard sigmoid activation function.", "concept": [ "activations" ], "topics": [ "activation_hard_sigmoid" ] }, { "page": "activation_hard_silu", "title": "Hard SiLU activation function, also known as Hard Swish.", "topics": [ "activation_hard_silu", "activation_hard_swish" ] }, { "page": "activation_hard_tanh", "title": "HardTanh activation function.", "concept": [ "activations" ], "topics": [ "activation_hard_tanh" ] }, { "page": "activation_leaky_relu", "title": "Leaky relu activation function.", "concept": [ "activations" ], "topics": [ "activation_leaky_relu" ] }, { "page": "activation_linear", "title": "Linear activation function (pass-through).", "concept": [ "activations" ], "topics": [ "activation_linear" ] }, { "page": "activation_log_sigmoid", "title": "Logarithm of the sigmoid activation function.", "concept": [ "activations" ], "topics": [ "activation_log_sigmoid" ] }, { "page": "activation_log_softmax", "title": "Log-Softmax activation function.", "concept": [ "activations" ], "topics": [ "activation_log_softmax" ] }, { "page": "activation_mish", "title": "Mish activation function.", "concept": [ "activations" ], "topics": [ "activation_mish" ] }, { "page": "activation_relu", "title": "Applies the rectified linear unit activation function.", "concept": [ "activations" ], "topics": [ "activation_relu" ] }, { "page": "activation_relu6", "title": "Relu6 activation function.", "concept": [ "activations" ], "topics": [ "activation_relu6" ] }, { "page": "activation_selu", "title": "Scaled Exponential Linear Unit (SELU).", "concept": [ "activations" ], "topics": [ "activation_selu" ] }, { "page": "activation_sigmoid", "title": "Sigmoid activation function.", "concept": [ "activations" ], "topics": [ "activation_sigmoid" ] }, { "page": "activation_silu", "title": "Swish (or Silu) activation function.", "concept": [ "activations" ], "topics": [ "activation_silu" ] }, { "page": "activation_soft_shrink", "title": "Soft Shrink activation function.", "concept": [ "activations" ], "topics": [ "activation_soft_shrink" ] }, { "page": "activation_softmax", "title": "Softmax converts a vector of values to a probability distribution.", "concept": [ "activations" ], "topics": [ "activation_softmax" ] }, { "page": "activation_softplus", "title": "Softplus activation function.", "concept": [ "activations" ], "topics": [ "activation_softplus" ] }, { "page": "activation_softsign", "title": "Softsign activation function.", "concept": [ "activations" ], "topics": [ "activation_softsign" ] }, { "page": "activation_sparse_plus", "title": "SparsePlus activation function.", "concept": [ "activations" ], "topics": [ "activation_sparse_plus" ] }, { "page": "activation_sparse_sigmoid", "title": "Sparse sigmoid activation function.", "concept": [ "activations" ], "topics": [ "activation_sparse_sigmoid" ] }, { "page": "activation_sparsemax", "title": "Sparsemax activation function.", "concept": [ "activations" ], "topics": [ "activation_sparsemax" ] }, { "page": "activation_squareplus", "title": "Squareplus activation function.", "concept": [ "activations" ], "topics": [ "activation_squareplus" ] }, { "page": "activation_tanh", "title": "Hyperbolic tangent activation function.", "concept": [ "activations" ], "topics": [ "activation_tanh" ] }, { "page": "activation_tanh_shrink", "title": "Tanh shrink activation function.", "concept": [ "activations" ], "topics": [ "activation_tanh_shrink" ] }, { "page": "activation_threshold", "title": "Threshold activation function.", "concept": [ "activations" ], "topics": [ "activation_threshold" ] }, { "page": "active_property", "title": "Create an active property class method", "topics": [ "active_property" ] }, { "page": "adapt", "title": "Fits the state of the preprocessing layer to the data being passed", "concept": [ "preprocessing layer methods" ], "topics": [ "adapt" ] }, { "page": "application_convnext_base", "title": "Instantiates the ConvNeXtBase architecture.", "topics": [ "application_convnext_base" ] }, { "page": "application_convnext_large", "title": "Instantiates the ConvNeXtLarge architecture.", "topics": [ "application_convnext_large" ] }, { "page": "application_convnext_small", "title": "Instantiates the ConvNeXtSmall architecture.", "topics": [ "application_convnext_small" ] }, { "page": "application_convnext_tiny", "title": "Instantiates the ConvNeXtTiny architecture.", "topics": [ "application_convnext_tiny" ] }, { "page": "application_convnext_xlarge", "title": "Instantiates the ConvNeXtXLarge architecture.", "topics": [ "application_convnext_xlarge" ] }, { "page": "application_densenet121", "title": "Instantiates the Densenet121 architecture.", "topics": [ "application_densenet121" ] }, { "page": "application_densenet169", "title": "Instantiates the Densenet169 architecture.", "topics": [ "application_densenet169" ] }, { "page": "application_densenet201", "title": "Instantiates the Densenet201 architecture.", "topics": [ "application_densenet201" ] }, { "page": "application_efficientnet_b0", "title": "Instantiates the EfficientNetB0 architecture.", "topics": [ "application_efficientnet_b0" ] }, { "page": "application_efficientnet_b1", "title": "Instantiates the EfficientNetB1 architecture.", "topics": [ "application_efficientnet_b1" ] }, { "page": "application_efficientnet_b2", "title": "Instantiates the EfficientNetB2 architecture.", "topics": [ "application_efficientnet_b2" ] }, { "page": "application_efficientnet_b3", "title": "Instantiates the EfficientNetB3 architecture.", "topics": [ "application_efficientnet_b3" ] }, { "page": "application_efficientnet_b4", "title": "Instantiates the EfficientNetB4 architecture.", "topics": [ "application_efficientnet_b4" ] }, { "page": "application_efficientnet_b5", "title": "Instantiates the EfficientNetB5 architecture.", "topics": [ "application_efficientnet_b5" ] }, { "page": "application_efficientnet_b6", "title": "Instantiates the EfficientNetB6 architecture.", "topics": [ "application_efficientnet_b6" ] }, { "page": "application_efficientnet_b7", "title": "Instantiates the EfficientNetB7 architecture.", "topics": [ "application_efficientnet_b7" ] }, { "page": "application_efficientnet_v2b0", "title": "Instantiates the EfficientNetV2B0 architecture.", "topics": [ "application_efficientnet_v2b0" ] }, { "page": "application_efficientnet_v2b1", "title": "Instantiates the EfficientNetV2B1 architecture.", "topics": [ "application_efficientnet_v2b1" ] }, { "page": "application_efficientnet_v2b2", "title": "Instantiates the EfficientNetV2B2 architecture.", "topics": [ "application_efficientnet_v2b2" ] }, { "page": "application_efficientnet_v2b3", "title": "Instantiates the EfficientNetV2B3 architecture.", "topics": [ "application_efficientnet_v2b3" ] }, { "page": "application_efficientnet_v2l", "title": "Instantiates the EfficientNetV2L architecture.", "topics": [ "application_efficientnet_v2l" ] }, { "page": "application_efficientnet_v2m", "title": "Instantiates the EfficientNetV2M architecture.", "topics": [ "application_efficientnet_v2m" ] }, { "page": "application_efficientnet_v2s", "title": "Instantiates the EfficientNetV2S architecture.", "topics": [ "application_efficientnet_v2s" ] }, { "page": "application_inception_resnet_v2", "title": "Instantiates the Inception-ResNet v2 architecture.", "topics": [ "application_inception_resnet_v2" ] }, { "page": "application_inception_v3", "title": "Instantiates the Inception v3 architecture.", "topics": [ "application_inception_v3" ] }, { "page": "application_mobilenet", "title": "Instantiates the MobileNet architecture.", "topics": [ "application_mobilenet" ] }, { "page": "application_mobilenet_v2", "title": "Instantiates the MobileNetV2 architecture.", "topics": [ "application_mobilenet_v2" ] }, { "page": "application_mobilenet_v3_large", "title": "Instantiates the MobileNetV3Large architecture.", "topics": [ "application_mobilenet_v3_large" ] }, { "page": "application_mobilenet_v3_small", "title": "Instantiates the MobileNetV3Small architecture.", "topics": [ "application_mobilenet_v3_small" ] }, { "page": "application_nasnet_large", "title": "Instantiates a NASNet model in ImageNet mode.", "topics": [ "application_nasnet_large" ] }, { "page": "application_nasnet_mobile", "title": "Instantiates a Mobile NASNet model in ImageNet mode.", "topics": [ "application_nasnet_mobile" ] }, { "page": "application_resnet101", "title": "Instantiates the ResNet101 architecture.", "topics": [ "application_resnet101" ] }, { "page": "application_resnet101_v2", "title": "Instantiates the ResNet101V2 architecture.", "topics": [ "application_resnet101_v2" ] }, { "page": "application_resnet152", "title": "Instantiates the ResNet152 architecture.", "topics": [ "application_resnet152" ] }, { "page": "application_resnet152_v2", "title": "Instantiates the ResNet152V2 architecture.", "topics": [ "application_resnet152_v2" ] }, { "page": "application_resnet50", "title": "Instantiates the ResNet50 architecture.", "topics": [ "application_resnet50" ] }, { "page": "application_resnet50_v2", "title": "Instantiates the ResNet50V2 architecture.", "topics": [ "application_resnet50_v2" ] }, { "page": "application_vgg16", "title": "Instantiates the VGG16 model.", "topics": [ "application_vgg16" ] }, { "page": "application_vgg19", "title": "Instantiates the VGG19 model.", "topics": [ "application_vgg19" ] }, { "page": "application_xception", "title": "Instantiates the Xception architecture.", "topics": [ "application_xception" ] }, { "page": "audio_dataset_from_directory", "title": "Generates a 'tf.data.Dataset' from audio files in a directory.", "concept": [ "dataset utils", "utils" ], "topics": [ "audio_dataset_from_directory" ] }, { "page": "Callback", "title": "Define a custom 'Callback' class", "concept": [ "callbacks" ], "topics": [ "Callback" ] }, { "page": "callback_backup_and_restore", "title": "Callback to back up and restore the training state.", "concept": [ "callbacks" ], "topics": [ "callback_backup_and_restore" ] }, { "page": "callback_csv_logger", "title": "Callback that streams epoch results to a CSV file.", "concept": [ "callbacks" ], "topics": [ "callback_csv_logger" ] }, { "page": "callback_early_stopping", "title": "Stop training when a monitored metric has stopped improving.", "concept": [ "callbacks" ], "topics": [ "callback_early_stopping" ] }, { "page": "callback_lambda", "title": "Callback for creating simple, custom callbacks on-the-fly.", "concept": [ "callbacks" ], "topics": [ "callback_lambda" ] }, { "page": "callback_learning_rate_scheduler", "title": "Learning rate scheduler.", "concept": [ "callbacks" ], "topics": [ "callback_learning_rate_scheduler" ] }, { "page": "callback_model_checkpoint", "title": "Callback to save the Keras model or model weights at some frequency.", "concept": [ "callbacks" ], "topics": [ "callback_model_checkpoint" ] }, { "page": "callback_reduce_lr_on_plateau", "title": "Reduce learning rate when a metric has stopped improving.", "concept": [ "callbacks" ], "topics": [ "callback_reduce_lr_on_plateau" ] }, { "page": "callback_remote_monitor", "title": "Callback used to stream events to a server.", "concept": [ "callbacks" ], "topics": [ "callback_remote_monitor" ] }, { "page": "callback_swap_ema_weights", "title": "Swaps model weights and EMA weights before and after evaluation.", "concept": [ "callbacks" ], "topics": [ "callback_swap_ema_weights" ] }, { "page": "callback_tensorboard", "title": "Enable visualizations for TensorBoard.", "concept": [ "callbacks" ], "topics": [ "callback_tensorboard" ] }, { "page": "callback_terminate_on_nan", "title": "Callback that terminates training when a NaN loss is encountered.", "concept": [ "callbacks" ], "topics": [ "callback_terminate_on_nan" ] }, { "page": "clear_session", "title": "Resets all state generated by Keras.", "concept": [ "backend", "utils" ], "topics": [ "clear_session" ] }, { "page": "clone_model", "title": "Clone a Functional or Sequential 'Model' instance.", "topics": [ "clone_model" ] }, { "page": "compile.keras.src.models.model.Model", "title": "Configure a model for training.", "concept": [ "model training" ], "topics": [ "compile.keras.src.models.model.Model" ] }, { "page": "config_backend", "title": "Publicly accessible method for determining the current backend.", "concept": [ "backend", "config", "config backend" ], "topics": [ "config_backend" ] }, { "page": "config_disable_flash_attention", "title": "Disable flash attention.", "concept": [ "config" ], "topics": [ "config_disable_flash_attention" ] }, { "page": "config_disable_interactive_logging", "title": "Turn off interactive logging.", "concept": [ "config", "io utils", "utils" ], "topics": [ "config_disable_interactive_logging" ] }, { "page": "config_disable_traceback_filtering", "title": "Turn off traceback filtering.", "concept": [ "config", "traceback utils", "utils" ], "topics": [ "config_disable_traceback_filtering" ] }, { "page": "config_dtype_policy", "title": "Returns the current default dtype policy object.", "concept": [ "config" ], "topics": [ "config_dtype_policy" ] }, { "page": "config_enable_flash_attention", "title": "Enable flash attention.", "concept": [ "config" ], "topics": [ "config_enable_flash_attention" ] }, { "page": "config_enable_interactive_logging", "title": "Turn on interactive logging.", "concept": [ "config", "io utils", "utils" ], "topics": [ "config_enable_interactive_logging" ] }, { "page": "config_enable_traceback_filtering", "title": "Turn on traceback filtering.", "concept": [ "config", "traceback utils", "utils" ], "topics": [ "config_enable_traceback_filtering" ] }, { "page": "config_enable_unsafe_deserialization", "title": "Disables safe mode globally, allowing deserialization of lambdas.", "concept": [ "config", "saving" ], "topics": [ "config_enable_unsafe_deserialization" ] }, { "page": "config_epsilon", "title": "Return the value of the fuzz factor used in numeric expressions.", "concept": [ "backend", "config", "config backend" ], "topics": [ "config_epsilon" ] }, { "page": "config_floatx", "title": "Return the default float type, as a string.", "concept": [ "backend", "config", "config backend" ], "topics": [ "config_floatx" ] }, { "page": "config_image_data_format", "title": "Return the default image data format convention.", "concept": [ "backend", "config", "config backend" ], "topics": [ "config_image_data_format" ] }, { "page": "config_is_flash_attention_enabled", "title": "Checks whether flash attention is globally enabled in Keras.", "topics": [ "config_is_flash_attention_enabled" ] }, { "page": "config_is_interactive_logging_enabled", "title": "Check if interactive logging is enabled.", "concept": [ "config", "io utils", "utils" ], "topics": [ "config_is_interactive_logging_enabled" ] }, { "page": "config_is_nnx_enabled", "title": "Check whether NNX-specific features are enabled on the JAX backend.", "concept": [ "config" ], "topics": [ "config_is_nnx_enabled" ] }, { "page": "config_is_traceback_filtering_enabled", "title": "Check if traceback filtering is enabled.", "concept": [ "config", "traceback utils", "utils" ], "topics": [ "config_is_traceback_filtering_enabled" ] }, { "page": "config_max_epochs", "title": "Configure the default training loop limits.", "concept": [ "config" ], "topics": [ "config_max_epochs", "config_max_steps_per_epoch", "config_set_max_epochs", "config_set_max_steps_per_epoch" ] }, { "page": "config_set_backend", "title": "Reload the backend (and the Keras package).", "concept": [ "config" ], "topics": [ "config_set_backend" ] }, { "page": "config_set_dtype_policy", "title": "Sets the default dtype policy globally.", "concept": [ "config" ], "topics": [ "config_set_dtype_policy" ] }, { "page": "config_set_epsilon", "title": "Set the value of the fuzz factor used in numeric expressions.", "concept": [ "backend", "config", "config backend" ], "topics": [ "config_set_epsilon" ] }, { "page": "config_set_floatx", "title": "Set the default float dtype.", "concept": [ "backend", "config", "config backend" ], "topics": [ "config_set_floatx" ] }, { "page": "config_set_image_data_format", "title": "Set the value of the image data format convention.", "concept": [ "backend", "config", "config backend" ], "topics": [ "config_set_image_data_format" ] }, { "page": "Constraint", "title": "Define a custom 'Constraint' class", "concept": [ "constraints" ], "topics": [ "Constraint" ] }, { "page": "constraint_maxnorm", "title": "MaxNorm weight constraint.", "concept": [ "constraints" ], "topics": [ "constraint_maxnorm" ] }, { "page": "constraint_minmaxnorm", "title": "MinMaxNorm weight constraint.", "concept": [ "constraints" ], "topics": [ "constraint_minmaxnorm" ] }, { "page": "constraint_nonneg", "title": "Constrains the weights to be non-negative.", "concept": [ "constraints" ], "topics": [ "constraint_nonneg" ] }, { "page": "constraint_unitnorm", "title": "Constrains the weights incident to each hidden unit to have unit norm.", "concept": [ "constraints" ], "topics": [ "constraint_unitnorm" ] }, { "page": "count_params", "title": "Count the total number of scalars composing the weights.", "concept": [ "layer methods" ], "topics": [ "count_params" ] }, { "page": "custom_metric", "title": "Custom metric function", "concept": [ "metrics" ], "topics": [ "custom_metric" ] }, { "page": "dataset_boston_housing", "title": "Boston housing price regression dataset", "concept": [ "datasets" ], "topics": [ "dataset_boston_housing" ] }, { "page": "dataset_california_housing", "title": "Loads the California Housing dataset.", "concept": [ "datasets" ], "topics": [ "dataset_california_housing" ] }, { "page": "dataset_cifar10", "title": "CIFAR10 small image classification", "concept": [ "datasets" ], "topics": [ "dataset_cifar10" ] }, { "page": "dataset_cifar100", "title": "CIFAR100 small image classification", "concept": [ "datasets" ], "topics": [ "dataset_cifar100" ] }, { "page": "dataset_fashion_mnist", "title": "Fashion-MNIST database of fashion articles", "concept": [ "datasets" ], "topics": [ "dataset_fashion_mnist" ] }, { "page": "dataset_imdb", "title": "IMDB Movie reviews sentiment classification", "concept": [ "datasets" ], "topics": [ "dataset_imdb", "dataset_imdb_word_index" ] }, { "page": "dataset_mnist", "title": "MNIST database of handwritten digits", "concept": [ "datasets" ], "topics": [ "dataset_mnist" ] }, { "page": "dataset_reuters", "title": "Reuters newswire topics classification", "concept": [ "datasets" ], "topics": [ "dataset_reuters", "dataset_reuters_word_index" ] }, { "page": "deserialize_keras_object", "title": "Retrieve the object by deserializing the config dict.", "concept": [ "serialization utilities" ], "topics": [ "deserialize_keras_object" ] }, { "page": "evaluate.keras.src.models.model.Model", "title": "Evaluate a Keras Model", "concept": [ "model training" ], "topics": [ "evaluate.keras.src.models.model.Model" ] }, { "page": "export_savedmodel.keras.src.models.model.Model", "title": "Export the model as an artifact for inference.", "concept": [ "saving and loading functions" ], "topics": [ "export_savedmodel.keras.src.models.model.Model" ] }, { "page": "fit.keras.src.models.model.Model", "title": "Train a model for a fixed number of epochs (dataset iterations).", "topics": [ "fit.keras.src.models.model.Model" ] }, { "page": "freeze_weights", "title": "Freeze and unfreeze weights", "topics": [ "freeze_weights", "unfreeze_weights" ] }, { "page": "get_config", "title": "Layer/Model configuration", "concept": [ "layer methods", "model functions" ], "topics": [ "from_config", "get_config" ] }, { "page": "get_custom_objects", "title": "Get/set the currently registered custom objects.", "concept": [ "serialization utilities" ], "topics": [ "get_custom_objects", "set_custom_objects" ] }, { "page": "get_file", "title": "Downloads a file from a URL if it not already in the cache.", "concept": [ "utils" ], "topics": [ "get_file" ] }, { "page": "get_layer", "title": "Retrieves a layer based on either its name (unique) or index.", "concept": [ "model functions" ], "topics": [ "get_layer" ] }, { "page": "get_registered_name", "title": "Returns the name registered to an object within the Keras framework.", "concept": [ "serialization utilities" ], "topics": [ "get_registered_name" ] }, { "page": "get_registered_object", "title": "Returns the class associated with 'name' if it is registered with Keras.", "concept": [ "serialization utilities" ], "topics": [ "get_registered_object" ] }, { "page": "get_source_inputs", "title": "Returns the list of input tensors necessary to compute 'tensor'.", "concept": [ "utils" ], "topics": [ "get_source_inputs" ] }, { "page": "get_state_tree", "title": "Retrieves tree-like structure of model variables.", "concept": [ "model functions" ], "topics": [ "get_state_tree" ] }, { "page": "get_weights", "title": "Layer/Model weights as R arrays", "concept": [ "layer methods", "model persistence" ], "topics": [ "get_weights", "set_weights" ] }, { "page": "image_array_save", "title": "Saves an image stored as an array to a path or file object.", "concept": [ "image utils", "utils" ], "topics": [ "image_array_save" ] }, { "page": "image_dataset_from_directory", "title": "Generates a 'tf.data.Dataset' from image files in a directory.", "concept": [ "dataset utils", "image dataset utils", "preprocessing", "utils" ], "topics": [ "image_dataset_from_directory" ] }, { "page": "image_from_array", "title": "Converts a 3D array to a PIL Image instance.", "concept": [ "image utils", "utils" ], "topics": [ "image_from_array" ] }, { "page": "image_load", "title": "Loads an image into PIL format.", "concept": [ "image utils", "utils" ], "topics": [ "image_load" ] }, { "page": "image_smart_resize", "title": "Resize images to a target size without aspect ratio distortion.", "concept": [ "image utils", "preprocessing", "utils" ], "topics": [ "image_smart_resize" ] }, { "page": "image_to_array", "title": "Converts a PIL Image instance to a matrix.", "concept": [ "image utils", "utils" ], "topics": [ "image_to_array" ] }, { "page": "initializer_constant", "title": "Initializer that generates tensors with constant values.", "concept": [ "constant initializers", "initializers" ], "topics": [ "initializer_constant" ] }, { "page": "initializer_glorot_normal", "title": "The Glorot normal initializer, also called Xavier normal initializer.", "concept": [ "initializers", "random initializers" ], "topics": [ "initializer_glorot_normal" ] }, { "page": "initializer_glorot_uniform", "title": "The Glorot uniform initializer, also called Xavier uniform initializer.", "concept": [ "initializers", "random initializers" ], "topics": [ "initializer_glorot_uniform" ] }, { "page": "initializer_he_normal", "title": "He normal initializer.", "concept": [ "initializers", "random initializers" ], "topics": [ "initializer_he_normal" ] }, { "page": "initializer_he_uniform", "title": "He uniform variance scaling initializer.", "concept": [ "initializers", "random initializers" ], "topics": [ "initializer_he_uniform" ] }, { "page": "initializer_identity", "title": "Initializer that generates the identity matrix.", "concept": [ "constant initializers", "initializers" ], "topics": [ "initializer_identity" ] }, { "page": "initializer_lecun_normal", "title": "Lecun normal initializer.", "concept": [ "initializers", "random initializers" ], "topics": [ "initializer_lecun_normal" ] }, { "page": "initializer_lecun_uniform", "title": "Lecun uniform initializer.", "concept": [ "initializers", "random initializers" ], "topics": [ "initializer_lecun_uniform" ] }, { "page": "initializer_ones", "title": "Initializer that generates tensors initialized to 1.", "concept": [ "constant initializers", "initializers" ], "topics": [ "initializer_ones" ] }, { "page": "initializer_orthogonal", "title": "Initializer that generates an orthogonal matrix.", "concept": [ "initializers", "random initializers" ], "topics": [ "initializer_orthogonal" ] }, { "page": "initializer_random_normal", "title": "Random normal initializer.", "concept": [ "initializers", "random initializers" ], "topics": [ "initializer_random_normal" ] }, { "page": "initializer_random_uniform", "title": "Random uniform initializer.", "concept": [ "initializers", "random initializers" ], "topics": [ "initializer_random_uniform" ] }, { "page": "initializer_stft", "title": "Initializer of Conv kernels for Short-term Fourier Transformation (STFT).", "concept": [ "constant initializers", "initializers" ], "topics": [ "initializer_stft" ] }, { "page": "initializer_truncated_normal", "title": "Initializer that generates a truncated normal distribution.", "concept": [ "initializers", "random initializers" ], "topics": [ "initializer_truncated_normal" ] }, { "page": "initializer_variance_scaling", "title": "Initializer that adapts its scale to the shape of its input tensors.", "concept": [ "initializers", "random initializers" ], "topics": [ "initializer_variance_scaling" ] }, { "page": "initializer_zeros", "title": "Initializer that generates tensors initialized to 0.", "concept": [ "constant initializers", "initializers" ], "topics": [ "initializer_zeros" ] }, { "page": "install_keras", "title": "Install Keras", "topics": [ "install_keras" ] }, { "page": "keras", "title": "Main Keras module", "topics": [ "keras" ] }, { "page": "keras_input", "title": "Create a Keras tensor (Functional API input).", "concept": [ "model creation" ], "topics": [ "keras_input" ] }, { "page": "keras_model", "title": "Keras Model (Functional API)", "concept": [ "model creation", "model functions" ], "topics": [ "keras_model" ] }, { "page": "keras_model_sequential", "title": "Keras Model composed of a linear stack of layers", "concept": [ "model creation", "model functions" ], "topics": [ "keras_model_sequential" ] }, { "page": "keras_variable", "title": "Represents a backend-agnostic variable in Keras.", "topics": [ "keras_variable" ] }, { "page": "Layer", "title": "Define a custom 'Layer' class.", "concept": [ "layers" ], "topics": [ "Layer" ] }, { "page": "layer_activation", "title": "Applies an activation function to an output.", "concept": [ "activation layers", "layers" ], "topics": [ "layer_activation" ] }, { "page": "layer_activation_elu", "title": "Applies an Exponential Linear Unit function to an output.", "concept": [ "activation layers", "layers" ], "topics": [ "layer_activation_elu" ] }, { "page": "layer_activation_leaky_relu", "title": "Leaky version of a Rectified Linear Unit activation layer.", "concept": [ "activation layers", "layers" ], "topics": [ "layer_activation_leaky_relu" ] }, { "page": "layer_activation_parametric_relu", "title": "Parametric Rectified Linear Unit activation layer.", "concept": [ "activation layers", "layers" ], "topics": [ "layer_activation_parametric_relu" ] }, { "page": "layer_activation_relu", "title": "Rectified Linear Unit activation function layer.", "concept": [ "activation layers", "layers" ], "topics": [ "layer_activation_relu" ] }, { "page": "layer_activation_softmax", "title": "Softmax activation layer.", "concept": [ "activation layers", "layers" ], "topics": [ "layer_activation_softmax" ] }, { "page": "layer_activity_regularization", "title": "Layer that applies an update to the cost function based input activity.", "concept": [ "layers", "regularization layers" ], "topics": [ "layer_activity_regularization" ] }, { "page": "layer_add", "title": "Performs elementwise addition operation.", "concept": [ "add merging layers", "layers", "merging layers" ], "topics": [ "layer_add" ] }, { "page": "layer_additive_attention", "title": "Additive attention layer, a.k.a. Bahdanau-style attention.", "concept": [ "attention layers", "layers" ], "topics": [ "layer_additive_attention" ] }, { "page": "layer_alpha_dropout", "title": "Applies Alpha Dropout to the input.", "concept": [ "layers", "regularization layers" ], "topics": [ "layer_alpha_dropout" ] }, { "page": "layer_attention", "title": "Dot-product attention layer, a.k.a. Luong-style attention.", "concept": [ "attention layers", "layers" ], "topics": [ "layer_attention" ] }, { "page": "layer_aug_mix", "title": "Performs the AugMix data augmentation technique.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_aug_mix" ] }, { "page": "layer_auto_contrast", "title": "Performs the auto-contrast operation on an image.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_auto_contrast" ] }, { "page": "layer_average", "title": "Averages a list of inputs element-wise..", "concept": [ "average merging layers", "layers", "merging layers" ], "topics": [ "layer_average" ] }, { "page": "layer_average_pooling_1d", "title": "Average pooling for temporal data.", "concept": [ "layers", "pooling layers" ], "topics": [ "layer_average_pooling_1d" ] }, { "page": "layer_average_pooling_2d", "title": "Average pooling operation for 2D spatial data.", "concept": [ "layers", "pooling layers" ], "topics": [ "layer_average_pooling_2d" ] }, { "page": "layer_average_pooling_3d", "title": "Average pooling operation for 3D data (spatial or spatio-temporal).", "concept": [ "layers", "pooling layers" ], "topics": [ "layer_average_pooling_3d" ] }, { "page": "layer_batch_normalization", "title": "Layer that normalizes its inputs.", "concept": [ "layers", "normalization layers" ], "topics": [ "layer_batch_normalization" ] }, { "page": "layer_bidirectional", "title": "Bidirectional wrapper for RNNs.", "concept": [ "layers", "rnn layers" ], "topics": [ "layer_bidirectional" ] }, { "page": "layer_category_encoding", "title": "A preprocessing layer which encodes integer features.", "concept": [ "categorical features preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_category_encoding" ] }, { "page": "layer_center_crop", "title": "A preprocessing layer which crops images.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_center_crop" ] }, { "page": "layer_concatenate", "title": "Concatenates a list of inputs.", "concept": [ "concatenate merging layers", "layers", "merging layers" ], "topics": [ "layer_concatenate" ] }, { "page": "layer_conv_1d", "title": "1D convolution layer (e.g. temporal convolution).", "concept": [ "convolutional layers", "layers" ], "topics": [ "layer_conv_1d" ] }, { "page": "layer_conv_1d_transpose", "title": "1D transposed convolution layer.", "concept": [ "convolutional layers", "layers" ], "topics": [ "layer_conv_1d_transpose" ] }, { "page": "layer_conv_2d", "title": "2D convolution layer.", "concept": [ "convolutional layers", "layers" ], "topics": [ "layer_conv_2d" ] }, { "page": "layer_conv_2d_transpose", "title": "2D transposed convolution layer.", "concept": [ "convolutional layers", "layers" ], "topics": [ "layer_conv_2d_transpose" ] }, { "page": "layer_conv_3d", "title": "3D convolution layer.", "concept": [ "convolutional layers", "layers" ], "topics": [ "layer_conv_3d" ] }, { "page": "layer_conv_3d_transpose", "title": "3D transposed convolution layer.", "concept": [ "convolutional layers", "layers" ], "topics": [ "layer_conv_3d_transpose" ] }, { "page": "layer_conv_lstm_1d", "title": "1D Convolutional LSTM.", "concept": [ "layers", "rnn layers" ], "topics": [ "layer_conv_lstm_1d" ] }, { "page": "layer_conv_lstm_2d", "title": "2D Convolutional LSTM.", "concept": [ "layers", "rnn layers" ], "topics": [ "layer_conv_lstm_2d" ] }, { "page": "layer_conv_lstm_3d", "title": "3D Convolutional LSTM.", "concept": [ "layers", "rnn layers" ], "topics": [ "layer_conv_lstm_3d" ] }, { "page": "layer_cropping_1d", "title": "Cropping layer for 1D input (e.g. temporal sequence).", "concept": [ "layers", "reshaping layers" ], "topics": [ "layer_cropping_1d" ] }, { "page": "layer_cropping_2d", "title": "Cropping layer for 2D input (e.g. picture).", "concept": [ "layers", "reshaping layers" ], "topics": [ "layer_cropping_2d" ] }, { "page": "layer_cropping_3d", "title": "Cropping layer for 3D data (e.g. spatial or spatio-temporal).", "concept": [ "layers", "reshaping layers" ], "topics": [ "layer_cropping_3d" ] }, { "page": "layer_cut_mix", "title": "CutMix data augmentation technique.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_cut_mix" ] }, { "page": "layer_dense", "title": "Just your regular densely-connected NN layer.", "concept": [ "core layers", "layers" ], "topics": [ "layer_dense" ] }, { "page": "layer_depthwise_conv_1d", "title": "1D depthwise convolution layer.", "concept": [ "convolutional layers", "layers" ], "topics": [ "layer_depthwise_conv_1d" ] }, { "page": "layer_depthwise_conv_2d", "title": "2D depthwise convolution layer.", "concept": [ "convolutional layers", "layers" ], "topics": [ "layer_depthwise_conv_2d" ] }, { "page": "layer_discretization", "title": "A preprocessing layer which buckets continuous features by ranges.", "concept": [ "layers", "numerical features preprocessing layers", "preprocessing layers" ], "topics": [ "layer_discretization" ] }, { "page": "layer_dot", "title": "Computes element-wise dot product of two tensors.", "concept": [ "dot merging layers", "layers", "merging layers" ], "topics": [ "layer_dot" ] }, { "page": "layer_dropout", "title": "Applies dropout to the input.", "concept": [ "layers", "regularization layers" ], "topics": [ "layer_dropout" ] }, { "page": "layer_einsum_dense", "title": "A layer that uses 'einsum' as the backing computation.", "concept": [ "core layers", "layers" ], "topics": [ "layer_einsum_dense" ] }, { "page": "layer_embedding", "title": "Turns nonnegative integers (indexes) into dense vectors of fixed size.", "concept": [ "core layers", "layers" ], "topics": [ "layer_embedding" ] }, { "page": "layer_equalization", "title": "Preprocessing layer for histogram equalization on image channels.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_equalization" ] }, { "page": "layer_feature_space", "title": "One-stop utility for preprocessing and encoding structured data.", "concept": [ "layers", "preprocessing layers", "utils" ], "topics": [ "feature_cross", "feature_custom", "feature_float", "feature_float_discretized", "feature_float_normalized", "feature_float_rescaled", "feature_integer_categorical", "feature_integer_hashed", "feature_string_categorical", "feature_string_hashed", "layer_feature_space" ] }, { "page": "layer_flatten", "title": "Flattens the input. Does not affect the batch size.", "concept": [ "layers", "reshaping layers" ], "topics": [ "layer_flatten" ] }, { "page": "layer_flax_module_wrapper", "title": "Keras Layer that wraps a Flax module.", "concept": [ "layers", "wrapping layers" ], "topics": [ "layer_flax_module_wrapper" ] }, { "page": "layer_gaussian_dropout", "title": "Apply multiplicative 1-centered Gaussian noise.", "concept": [ "layers", "regularization layers" ], "topics": [ "layer_gaussian_dropout" ] }, { "page": "layer_gaussian_noise", "title": "Apply additive zero-centered Gaussian noise.", "concept": [ "layers", "regularization layers" ], "topics": [ "layer_gaussian_noise" ] }, { "page": "layer_global_average_pooling_1d", "title": "Global average pooling operation for temporal data.", "concept": [ "layers", "pooling layers" ], "topics": [ "layer_global_average_pooling_1d" ] }, { "page": "layer_global_average_pooling_2d", "title": "Global average pooling operation for 2D data.", "concept": [ "layers", "pooling layers" ], "topics": [ "layer_global_average_pooling_2d" ] }, { "page": "layer_global_average_pooling_3d", "title": "Global average pooling operation for 3D data.", "concept": [ "layers", "pooling layers" ], "topics": [ "layer_global_average_pooling_3d" ] }, { "page": "layer_global_max_pooling_1d", "title": "Global max pooling operation for temporal data.", "concept": [ "layers", "pooling layers" ], "topics": [ "layer_global_max_pooling_1d" ] }, { "page": "layer_global_max_pooling_2d", "title": "Global max pooling operation for 2D data.", "concept": [ "layers", "pooling layers" ], "topics": [ "layer_global_max_pooling_2d" ] }, { "page": "layer_global_max_pooling_3d", "title": "Global max pooling operation for 3D data.", "concept": [ "layers", "pooling layers" ], "topics": [ "layer_global_max_pooling_3d" ] }, { "page": "layer_group_normalization", "title": "Group normalization layer.", "concept": [ "layers", "normalization layers" ], "topics": [ "layer_group_normalization" ] }, { "page": "layer_group_query_attention", "title": "Grouped Query Attention layer.", "concept": [ "attention layers", "layers" ], "topics": [ "layer_group_query_attention" ] }, { "page": "layer_gru", "title": "Gated Recurrent Unit - Cho et al. 2014.", "concept": [ "gru rnn layers", "layers", "rnn layers" ], "topics": [ "layer_gru" ] }, { "page": "layer_hashed_crossing", "title": "A preprocessing layer which crosses features using the \"hashing trick\".", "concept": [ "categorical features preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_hashed_crossing" ] }, { "page": "layer_hashing", "title": "A preprocessing layer which hashes and bins categorical features.", "concept": [ "categorical features preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_hashing" ] }, { "page": "layer_identity", "title": "Identity layer.", "concept": [ "core layers", "layers" ], "topics": [ "layer_identity" ] }, { "page": "layer_integer_lookup", "title": "A preprocessing layer that maps integers to (possibly encoded) indices.", "concept": [ "categorical features preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_integer_lookup" ] }, { "page": "layer_jax_model_wrapper", "title": "Keras Layer that wraps a JAX model.", "concept": [ "layers", "wrapping layers" ], "topics": [ "layer_jax_model_wrapper" ] }, { "page": "layer_lambda", "title": "Wraps arbitrary expressions as a 'Layer' object.", "concept": [ "core layers", "layers" ], "topics": [ "layer_lambda" ] }, { "page": "layer_layer_normalization", "title": "Layer normalization layer (Ba et al., 2016).", "concept": [ "layers", "normalization layers" ], "topics": [ "layer_layer_normalization" ] }, { "page": "layer_lstm", "title": "Long Short-Term Memory layer - Hochreiter 1997.", "concept": [ "layers", "lstm rnn layers", "rnn layers" ], "topics": [ "layer_lstm" ] }, { "page": "layer_masking", "title": "Masks a sequence by using a mask value to skip timesteps.", "concept": [ "core layers", "layers" ], "topics": [ "layer_masking" ] }, { "page": "layer_max_num_bounding_boxes", "title": "Ensure the maximum number of bounding boxes.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_max_num_bounding_boxes" ] }, { "page": "layer_max_pooling_1d", "title": "Max pooling operation for 1D temporal data.", "concept": [ "layers", "pooling layers" ], "topics": [ "layer_max_pooling_1d" ] }, { "page": "layer_max_pooling_2d", "title": "Max pooling operation for 2D spatial data.", "concept": [ "layers", "pooling layers" ], "topics": [ "layer_max_pooling_2d" ] }, { "page": "layer_max_pooling_3d", "title": "Max pooling operation for 3D data (spatial or spatio-temporal).", "concept": [ "layers", "pooling layers" ], "topics": [ "layer_max_pooling_3d" ] }, { "page": "layer_maximum", "title": "Computes element-wise maximum on a list of inputs.", "concept": [ "layers", "maximum merging layers", "merging layers" ], "topics": [ "layer_maximum" ] }, { "page": "layer_mel_spectrogram", "title": "A preprocessing layer to convert raw audio signals to Mel spectrograms.", "concept": [ "audio preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_mel_spectrogram" ] }, { "page": "layer_minimum", "title": "Computes elementwise minimum on a list of inputs.", "concept": [ "layers", "merging layers", "minimum merging layers" ], "topics": [ "layer_minimum" ] }, { "page": "layer_mix_up", "title": "MixUp implements the MixUp data augmentation technique.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_mix_up" ] }, { "page": "layer_multi_head_attention", "title": "Multi Head Attention layer.", "concept": [ "attention layers", "layers" ], "topics": [ "layer_multi_head_attention" ] }, { "page": "layer_multiply", "title": "Performs elementwise multiplication.", "concept": [ "layers", "merging layers", "multiply merging layers" ], "topics": [ "layer_multiply" ] }, { "page": "layer_normalization", "title": "A preprocessing layer that normalizes continuous features.", "concept": [ "layers", "numerical features preprocessing layers", "preprocessing layers" ], "topics": [ "layer_normalization" ] }, { "page": "layer_permute", "title": "Permutes the dimensions of the input according to a given pattern.", "concept": [ "layers", "reshaping layers" ], "topics": [ "layer_permute" ] }, { "page": "layer_pipeline", "title": "Applies a series of layers to an input.", "topics": [ "layer_pipeline" ] }, { "page": "layer_rand_augment", "title": "RandAugment performs the Rand Augment operation on input images.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_rand_augment" ] }, { "page": "layer_random_brightness", "title": "A preprocessing layer which randomly adjusts brightness during training.", "concept": [ "image augmentation layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_brightness" ] }, { "page": "layer_random_color_degeneration", "title": "Randomly performs the color degeneration operation on given images.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_color_degeneration" ] }, { "page": "layer_random_color_jitter", "title": "Randomly apply brightness, contrast, saturation", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_color_jitter" ] }, { "page": "layer_random_contrast", "title": "A preprocessing layer which randomly adjusts contrast during training.", "concept": [ "image augmentation layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_contrast" ] }, { "page": "layer_random_crop", "title": "A preprocessing layer which randomly crops images during training.", "concept": [ "image augmentation layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_crop" ] }, { "page": "layer_random_elastic_transform", "title": "A preprocessing layer that applies random elastic transformations.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_elastic_transform" ] }, { "page": "layer_random_erasing", "title": "Random Erasing data augmentation technique.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_erasing" ] }, { "page": "layer_random_flip", "title": "A preprocessing layer which randomly flips images during training.", "concept": [ "image augmentation layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_flip" ] }, { "page": "layer_random_gaussian_blur", "title": "Applies random Gaussian blur to images for data augmentation.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_gaussian_blur" ] }, { "page": "layer_random_grayscale", "title": "Preprocessing layer for random conversion of RGB images to grayscale.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_grayscale" ] }, { "page": "layer_random_hue", "title": "Randomly adjusts the hue on given images.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_hue" ] }, { "page": "layer_random_invert", "title": "Preprocessing layer for random inversion of image colors.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_invert" ] }, { "page": "layer_random_perspective", "title": "A preprocessing layer that applies random perspective transformations.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_perspective" ] }, { "page": "layer_random_posterization", "title": "Reduces the number of bits for each color channel.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_posterization" ] }, { "page": "layer_random_rotation", "title": "A preprocessing layer which randomly rotates images during training.", "concept": [ "image augmentation layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_rotation" ] }, { "page": "layer_random_saturation", "title": "Randomly adjusts the saturation on given images.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_saturation" ] }, { "page": "layer_random_sharpness", "title": "Randomly performs the sharpness operation on given images.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_sharpness" ] }, { "page": "layer_random_shear", "title": "A preprocessing layer that randomly applies shear transformations", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_shear" ] }, { "page": "layer_random_translation", "title": "A preprocessing layer which randomly translates images during training.", "concept": [ "image augmentation layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_translation" ] }, { "page": "layer_random_zoom", "title": "A preprocessing layer which randomly zooms images during training.", "concept": [ "image augmentation layers", "layers", "preprocessing layers" ], "topics": [ "layer_random_zoom" ] }, { "page": "layer_repeat_vector", "title": "Repeats the input n times.", "concept": [ "layers", "reshaping layers" ], "topics": [ "layer_repeat_vector" ] }, { "page": "layer_rescaling", "title": "A preprocessing layer which rescales input values to a new range.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_rescaling" ] }, { "page": "layer_reshape", "title": "Layer that reshapes inputs into the given shape.", "concept": [ "layers", "reshaping layers" ], "topics": [ "layer_reshape" ] }, { "page": "layer_resizing", "title": "A preprocessing layer which resizes images.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_resizing" ] }, { "page": "layer_rms_normalization", "title": "Root Mean Square (RMS) Normalization layer.", "concept": [ "layers", "normalization layers" ], "topics": [ "layer_rms_normalization" ] }, { "page": "layer_rnn", "title": "Base class for recurrent layers", "concept": [ "layers", "rnn cells", "rnn layers" ], "topics": [ "layer_rnn" ] }, { "page": "layer_separable_conv_1d", "title": "1D separable convolution layer.", "concept": [ "convolutional layers", "layers" ], "topics": [ "layer_separable_conv_1d" ] }, { "page": "layer_separable_conv_2d", "title": "2D separable convolution layer.", "concept": [ "convolutional layers", "layers" ], "topics": [ "layer_separable_conv_2d" ] }, { "page": "layer_simple_rnn", "title": "Fully-connected RNN where the output is to be fed back as the new input.", "concept": [ "layers", "rnn layers", "simple rnn layers" ], "topics": [ "layer_simple_rnn" ] }, { "page": "layer_solarization", "title": "Applies '(max_value - pixel + min_value)' for each pixel in the image.", "concept": [ "image preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_solarization" ] }, { "page": "layer_spatial_dropout_1d", "title": "Spatial 1D version of Dropout.", "concept": [ "layers", "regularization layers", "spatial dropout regularization layers" ], "topics": [ "layer_spatial_dropout_1d" ] }, { "page": "layer_spatial_dropout_2d", "title": "Spatial 2D version of Dropout.", "concept": [ "layers", "regularization layers", "spatial dropout regularization layers" ], "topics": [ "layer_spatial_dropout_2d" ] }, { "page": "layer_spatial_dropout_3d", "title": "Spatial 3D version of Dropout.", "concept": [ "layers", "regularization layers", "spatial dropout regularization layers" ], "topics": [ "layer_spatial_dropout_3d" ] }, { "page": "layer_spectral_normalization", "title": "Performs spectral normalization on the weights of a target layer.", "concept": [ "layers", "normalization layers" ], "topics": [ "layer_spectral_normalization" ] }, { "page": "layer_stft_spectrogram", "title": "Layer to compute the Short-Time Fourier Transform (STFT) on a 1D signal.", "concept": [ "audio preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_stft_spectrogram" ] }, { "page": "layer_string_lookup", "title": "A preprocessing layer that maps strings to (possibly encoded) indices.", "concept": [ "categorical features preprocessing layers", "layers", "preprocessing layers" ], "topics": [ "layer_string_lookup" ] }, { "page": "layer_subtract", "title": "Performs elementwise subtraction.", "concept": [ "layers", "merging layers", "subtract merging layers" ], "topics": [ "layer_subtract" ] }, { "page": "layer_text_vectorization", "title": "A preprocessing layer which maps text features to integer sequences.", "concept": [ "layers", "preprocessing layers" ], "topics": [ "get_vocabulary", "layer_text_vectorization", "set_vocabulary" ] }, { "page": "layer_tfsm", "title": "Reload a Keras model/layer that was saved via 'export_savedmodel()'.", "concept": [ "layers", "saving and loading functions" ], "topics": [ "layer_tfsm" ] }, { "page": "layer_time_distributed", "title": "This wrapper allows to apply a layer to every temporal slice of an input.", "concept": [ "layers", "rnn layers" ], "topics": [ "layer_time_distributed" ] }, { "page": "layer_torch_module_wrapper", "title": "Torch module wrapper layer.", "concept": [ "layers", "wrapping layers" ], "topics": [ "layer_torch_module_wrapper" ] }, { "page": "layer_unit_normalization", "title": "Unit normalization layer.", "concept": [ "layers", "normalization layers" ], "topics": [ "layer_unit_normalization" ] }, { "page": "layer_upsampling_1d", "title": "Upsampling layer for 1D inputs.", "concept": [ "layers", "reshaping layers" ], "topics": [ "layer_upsampling_1d" ] }, { "page": "layer_upsampling_2d", "title": "Upsampling layer for 2D inputs.", "concept": [ "layers", "reshaping layers" ], "topics": [ "layer_upsampling_2d" ] }, { "page": "layer_upsampling_3d", "title": "Upsampling layer for 3D inputs.", "concept": [ "layers", "reshaping layers" ], "topics": [ "layer_upsampling_3d" ] }, { "page": "layer_zero_padding_1d", "title": "Zero-padding layer for 1D input (e.g. temporal sequence).", "concept": [ "layers", "reshaping layers" ], "topics": [ "layer_zero_padding_1d" ] }, { "page": "layer_zero_padding_2d", "title": "Zero-padding layer for 2D input (e.g. picture).", "concept": [ "layers", "reshaping layers" ], "topics": [ "layer_zero_padding_2d" ] }, { "page": "layer_zero_padding_3d", "title": "Zero-padding layer for 3D data (spatial or spatio-temporal).", "concept": [ "layers", "reshaping layers" ], "topics": [ "layer_zero_padding_3d" ] }, { "page": "learning_rate_schedule_cosine_decay", "title": "A 'LearningRateSchedule' that uses a cosine decay with optional warmup.", "concept": [ "optimizer learning rate schedules" ], "topics": [ "learning_rate_schedule_cosine_decay" ] }, { "page": "learning_rate_schedule_cosine_decay_restarts", "title": "A 'LearningRateSchedule' that uses a cosine decay schedule with restarts.", "concept": [ "optimizer learning rate schedules" ], "topics": [ "learning_rate_schedule_cosine_decay_restarts" ] }, { "page": "learning_rate_schedule_exponential_decay", "title": "A 'LearningRateSchedule' that uses an exponential decay schedule.", "concept": [ "optimizer learning rate schedules" ], "topics": [ "learning_rate_schedule_exponential_decay" ] }, { "page": "learning_rate_schedule_inverse_time_decay", "title": "A 'LearningRateSchedule' that uses an inverse time decay schedule.", "concept": [ "optimizer learning rate schedules" ], "topics": [ "learning_rate_schedule_inverse_time_decay" ] }, { "page": "learning_rate_schedule_piecewise_constant_decay", "title": "A 'LearningRateSchedule' that uses a piecewise constant decay schedule.", "concept": [ "optimizer learning rate schedules" ], "topics": [ "learning_rate_schedule_piecewise_constant_decay" ] }, { "page": "learning_rate_schedule_polynomial_decay", "title": "A 'LearningRateSchedule' that uses a polynomial decay schedule.", "concept": [ "optimizer learning rate schedules" ], "topics": [ "learning_rate_schedule_polynomial_decay" ] }, { "page": "LearningRateSchedule", "title": "Define a custom 'LearningRateSchedule' class", "concept": [ "optimizer learning rate schedules" ], "topics": [ "LearningRateSchedule" ] }, { "page": "load_model", "title": "Loads a model saved via 'save_model()'.", "concept": [ "saving and loading functions" ], "topics": [ "load_model" ] }, { "page": "load_model_weights", "title": "Load the weights from a single file or sharded files.", "concept": [ "saving and loading functions" ], "topics": [ "load_model_weights" ] }, { "page": "Loss", "title": "Subclass the base 'Loss' class", "concept": [ "losses" ], "topics": [ "Loss" ] }, { "page": "loss_binary_crossentropy", "title": "Computes the cross-entropy loss between true labels and predicted labels.", "concept": [ "losses" ], "topics": [ "loss_binary_crossentropy" ] }, { "page": "loss_binary_focal_crossentropy", "title": "Computes focal cross-entropy loss between true labels and predictions.", "concept": [ "losses" ], "topics": [ "loss_binary_focal_crossentropy" ] }, { "page": "loss_categorical_crossentropy", "title": "Computes the crossentropy loss between the labels and predictions.", "concept": [ "losses" ], "topics": [ "loss_categorical_crossentropy" ] }, { "page": "loss_categorical_focal_crossentropy", "title": "Computes the alpha balanced focal crossentropy loss.", "concept": [ "losses" ], "topics": [ "loss_categorical_focal_crossentropy" ] }, { "page": "loss_categorical_generalized_cross_entropy", "title": "Computes the generalized cross entropy loss.", "concept": [ "losses" ], "topics": [ "loss_categorical_generalized_cross_entropy" ] }, { "page": "loss_categorical_hinge", "title": "Computes the categorical hinge loss between 'y_true' & 'y_pred'.", "concept": [ "losses" ], "topics": [ "loss_categorical_hinge" ] }, { "page": "loss_circle", "title": "Computes Circle Loss between integer labels and L2-normalized embeddings.", "concept": [ "losses" ], "topics": [ "loss_circle" ] }, { "page": "loss_cosine_similarity", "title": "Computes the cosine similarity between 'y_true' & 'y_pred'.", "concept": [ "losses" ], "topics": [ "loss_cosine_similarity" ] }, { "page": "loss_ctc", "title": "CTC (Connectionist Temporal Classification) loss.", "concept": [ "losses" ], "topics": [ "loss_ctc" ] }, { "page": "loss_dice", "title": "Computes the Dice loss value between 'y_true' and 'y_pred'.", "concept": [ "losses" ], "topics": [ "loss_dice" ] }, { "page": "loss_hinge", "title": "Computes the hinge loss between 'y_true' & 'y_pred'.", "concept": [ "losses" ], "topics": [ "loss_hinge" ] }, { "page": "loss_huber", "title": "Computes the Huber loss between 'y_true' & 'y_pred'.", "concept": [ "losses" ], "topics": [ "loss_huber" ] }, { "page": "loss_kl_divergence", "title": "Computes Kullback-Leibler divergence loss between 'y_true' & 'y_pred'.", "concept": [ "losses" ], "topics": [ "loss_kl_divergence" ] }, { "page": "loss_log_cosh", "title": "Computes the logarithm of the hyperbolic cosine of the prediction error.", "concept": [ "losses" ], "topics": [ "loss_log_cosh" ] }, { "page": "loss_mean_absolute_error", "title": "Computes the mean of absolute difference between labels and predictions.", "concept": [ "losses" ], "topics": [ "loss_mean_absolute_error" ] }, { "page": "loss_mean_absolute_percentage_error", "title": "Computes the mean absolute percentage error between 'y_true' and 'y_pred'.", "concept": [ "losses" ], "topics": [ "loss_mean_absolute_percentage_error" ] }, { "page": "loss_mean_squared_error", "title": "Computes the mean of squares of errors between labels and predictions.", "concept": [ "losses" ], "topics": [ "loss_mean_squared_error" ] }, { "page": "loss_mean_squared_logarithmic_error", "title": "Computes the mean squared logarithmic error between 'y_true' and 'y_pred'.", "concept": [ "losses" ], "topics": [ "loss_mean_squared_logarithmic_error" ] }, { "page": "loss_poisson", "title": "Computes the Poisson loss between 'y_true' & 'y_pred'.", "concept": [ "losses" ], "topics": [ "loss_poisson" ] }, { "page": "loss_sparse_categorical_crossentropy", "title": "Computes the crossentropy loss between the labels and predictions.", "concept": [ "losses" ], "topics": [ "loss_sparse_categorical_crossentropy" ] }, { "page": "loss_squared_hinge", "title": "Computes the squared hinge loss between 'y_true' & 'y_pred'.", "concept": [ "losses" ], "topics": [ "loss_squared_hinge" ] }, { "page": "loss_tversky", "title": "Computes the Tversky loss value between 'y_true' and 'y_pred'.", "concept": [ "losses" ], "topics": [ "loss_tversky" ] }, { "page": "Metric", "title": "Subclass the base 'Metric' class", "concept": [ "metrics" ], "topics": [ "Metric" ] }, { "page": "metric_auc", "title": "Approximates the AUC (Area under the curve) of the ROC or PR curves.", "concept": [ "confusion metrics", "metrics" ], "topics": [ "metric_auc" ] }, { "page": "metric_binary_accuracy", "title": "Calculates how often predictions match binary labels.", "concept": [ "accuracy metrics", "metrics" ], "topics": [ "metric_binary_accuracy" ] }, { "page": "metric_binary_crossentropy", "title": "Computes the crossentropy metric between the labels and predictions.", "concept": [ "losses", "metrics", "probabilistic metrics" ], "topics": [ "metric_binary_crossentropy" ] }, { "page": "metric_binary_focal_crossentropy", "title": "Computes the binary focal crossentropy loss.", "concept": [ "losses", "metrics" ], "topics": [ "metric_binary_focal_crossentropy" ] }, { "page": "metric_binary_iou", "title": "Computes the Intersection-Over-Union metric for class 0 and/or 1.", "concept": [ "iou metrics", "metrics" ], "topics": [ "metric_binary_iou" ] }, { "page": "metric_categorical_accuracy", "title": "Calculates how often predictions match one-hot labels.", "concept": [ "accuracy metrics", "metrics" ], "topics": [ "metric_categorical_accuracy" ] }, { "page": "metric_categorical_crossentropy", "title": "Computes the crossentropy metric between the labels and predictions.", "concept": [ "losses", "metrics", "probabilistic metrics" ], "topics": [ "metric_categorical_crossentropy" ] }, { "page": "metric_categorical_focal_crossentropy", "title": "Computes the categorical focal crossentropy loss.", "concept": [ "losses", "metrics" ], "topics": [ "metric_categorical_focal_crossentropy" ] }, { "page": "metric_categorical_hinge", "title": "Computes the categorical hinge metric between 'y_true' and 'y_pred'.", "concept": [ "hinge metrics", "losses", "metrics" ], "topics": [ "metric_categorical_hinge" ] }, { "page": "metric_concordance_correlation", "title": "Calculates the Concordance Correlation Coefficient (CCC).", "concept": [ "metrics", "regression metrics" ], "topics": [ "metric_concordance_correlation" ] }, { "page": "metric_cosine_similarity", "title": "Computes the cosine similarity between the labels and predictions.", "concept": [ "metrics", "regression metrics" ], "topics": [ "metric_cosine_similarity" ] }, { "page": "metric_f1_score", "title": "Computes F-1 Score.", "concept": [ "f score metrics", "metrics" ], "topics": [ "metric_f1_score" ] }, { "page": "metric_false_negatives", "title": "Calculates the number of false negatives.", "concept": [ "confusion metrics", "metrics" ], "topics": [ "metric_false_negatives" ] }, { "page": "metric_false_positives", "title": "Calculates the number of false positives.", "concept": [ "confusion metrics", "metrics" ], "topics": [ "metric_false_positives" ] }, { "page": "metric_fbeta_score", "title": "Computes F-Beta score.", "concept": [ "f score metrics", "metrics" ], "topics": [ "metric_fbeta_score" ] }, { "page": "metric_hinge", "title": "Computes the hinge metric between 'y_true' and 'y_pred'.", "concept": [ "hinge metrics", "losses", "metrics" ], "topics": [ "metric_hinge" ] }, { "page": "metric_huber", "title": "Computes Huber loss value.", "concept": [ "losses", "metrics" ], "topics": [ "metric_huber" ] }, { "page": "metric_iou", "title": "Computes the Intersection-Over-Union metric for specific target classes.", "concept": [ "iou metrics", "metrics" ], "topics": [ "metric_iou" ] }, { "page": "metric_kl_divergence", "title": "Computes Kullback-Leibler divergence metric between 'y_true' and", "concept": [ "losses", "metrics", "probabilistic metrics" ], "topics": [ "metric_kl_divergence" ] }, { "page": "metric_log_cosh", "title": "Logarithm of the hyperbolic cosine of the prediction error.", "concept": [ "losses", "metrics" ], "topics": [ "metric_log_cosh" ] }, { "page": "metric_log_cosh_error", "title": "Computes the logarithm of the hyperbolic cosine of the prediction error.", "concept": [ "metrics", "regression metrics" ], "topics": [ "metric_log_cosh_error" ] }, { "page": "metric_mean", "title": "Compute the (weighted) mean of the given values.", "concept": [ "metrics", "reduction metrics" ], "topics": [ "metric_mean" ] }, { "page": "metric_mean_absolute_error", "title": "Computes the mean absolute error between the labels and predictions.", "concept": [ "losses", "metrics", "regression metrics" ], "topics": [ "metric_mean_absolute_error" ] }, { "page": "metric_mean_absolute_percentage_error", "title": "Computes mean absolute percentage error between 'y_true' and 'y_pred'.", "concept": [ "losses", "metrics", "regression metrics" ], "topics": [ "metric_mean_absolute_percentage_error" ] }, { "page": "metric_mean_iou", "title": "Computes the mean Intersection-Over-Union metric.", "concept": [ "iou metrics", "metrics" ], "topics": [ "metric_mean_iou" ] }, { "page": "metric_mean_squared_error", "title": "Computes the mean squared error between 'y_true' and 'y_pred'.", "concept": [ "losses", "metrics", "regression metrics" ], "topics": [ "metric_mean_squared_error" ] }, { "page": "metric_mean_squared_logarithmic_error", "title": "Computes mean squared logarithmic error between 'y_true' and 'y_pred'.", "concept": [ "losses", "metrics", "regression metrics" ], "topics": [ "metric_mean_squared_logarithmic_error" ] }, { "page": "metric_mean_wrapper", "title": "Wrap a stateless metric function with the 'Mean' metric.", "concept": [ "metrics", "reduction metrics" ], "topics": [ "metric_mean_wrapper" ] }, { "page": "metric_one_hot_iou", "title": "Computes the Intersection-Over-Union metric for one-hot encoded labels.", "concept": [ "iou metrics", "metrics" ], "topics": [ "metric_one_hot_iou" ] }, { "page": "metric_one_hot_mean_iou", "title": "Computes mean Intersection-Over-Union metric for one-hot encoded labels.", "concept": [ "iou metrics", "metrics" ], "topics": [ "metric_one_hot_mean_iou" ] }, { "page": "metric_pearson_correlation", "title": "Calculates the Pearson Correlation Coefficient (PCC).", "concept": [ "metrics", "regression metrics" ], "topics": [ "metric_pearson_correlation" ] }, { "page": "metric_poisson", "title": "Computes the Poisson metric between 'y_true' and 'y_pred'.", "concept": [ "losses", "metrics", "probabilistic metrics" ], "topics": [ "metric_poisson" ] }, { "page": "metric_precision", "title": "Computes the precision of the predictions with respect to the labels.", "concept": [ "confusion metrics", "metrics" ], "topics": [ "metric_precision" ] }, { "page": "metric_precision_at_recall", "title": "Computes best precision where recall is >= specified value.", "concept": [ "confusion metrics", "metrics" ], "topics": [ "metric_precision_at_recall" ] }, { "page": "metric_r2_score", "title": "Computes R2 score.", "concept": [ "metrics", "regression metrics" ], "topics": [ "metric_r2_score" ] }, { "page": "metric_recall", "title": "Computes the recall of the predictions with respect to the labels.", "concept": [ "confusion metrics", "metrics" ], "topics": [ "metric_recall" ] }, { "page": "metric_recall_at_precision", "title": "Computes best recall where precision is >= specified value.", "concept": [ "confusion metrics", "metrics" ], "topics": [ "metric_recall_at_precision" ] }, { "page": "metric_root_mean_squared_error", "title": "Computes root mean squared error metric between 'y_true' and 'y_pred'.", "concept": [ "metrics", "regression metrics" ], "topics": [ "metric_root_mean_squared_error" ] }, { "page": "metric_sensitivity_at_specificity", "title": "Computes best sensitivity where specificity is >= specified value.", "concept": [ "confusion metrics", "metrics" ], "topics": [ "metric_sensitivity_at_specificity" ] }, { "page": "metric_sparse_categorical_accuracy", "title": "Calculates how often predictions match integer labels.", "concept": [ "accuracy metrics", "metrics" ], "topics": [ "metric_sparse_categorical_accuracy" ] }, { "page": "metric_sparse_categorical_crossentropy", "title": "Computes the crossentropy metric between the labels and predictions.", "concept": [ "losses", "metrics", "probabilistic metrics" ], "topics": [ "metric_sparse_categorical_crossentropy" ] }, { "page": "metric_sparse_top_k_categorical_accuracy", "title": "Computes how often integer targets are in the top 'K' predictions.", "concept": [ "accuracy metrics", "metrics" ], "topics": [ "metric_sparse_top_k_categorical_accuracy" ] }, { "page": "metric_specificity_at_sensitivity", "title": "Computes best specificity where sensitivity is >= specified value.", "concept": [ "confusion metrics", "metrics" ], "topics": [ "metric_specificity_at_sensitivity" ] }, { "page": "metric_squared_hinge", "title": "Computes the hinge metric between 'y_true' and 'y_pred'.", "concept": [ "hinge metrics", "losses", "metrics" ], "topics": [ "metric_squared_hinge" ] }, { "page": "metric_sum", "title": "Compute the (weighted) sum of the given values.", "concept": [ "metrics", "reduction metrics" ], "topics": [ "metric_sum" ] }, { "page": "metric_top_k_categorical_accuracy", "title": "Computes how often targets are in the top 'K' predictions.", "concept": [ "accuracy metrics", "metrics" ], "topics": [ "metric_top_k_categorical_accuracy" ] }, { "page": "metric_true_negatives", "title": "Calculates the number of true negatives.", "concept": [ "confusion metrics", "metrics" ], "topics": [ "metric_true_negatives" ] }, { "page": "metric_true_positives", "title": "Calculates the number of true positives.", "concept": [ "confusion metrics", "metrics" ], "topics": [ "metric_true_positives" ] }, { "page": "Model", "title": "Subclass the base Keras 'Model' Class", "topics": [ "Model" ] }, { "page": "named_list", "title": "Create a named list from arguments", "topics": [ "named_list" ] }, { "page": "newaxis", "title": "New axis", "topics": [ "newaxis" ] }, { "page": "normalize", "title": "Normalizes an array.", "concept": [ "numerical utils", "utils" ], "topics": [ "normalize" ] }, { "page": "op_abs", "title": "Compute the absolute value element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_abs" ] }, { "page": "op_add", "title": "Add arguments element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_add" ] }, { "page": "op_all", "title": "Test whether all array elements along a given axis evaluate to 'TRUE'.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_all" ] }, { "page": "op_angle", "title": "Element-wise angle of a complex tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_angle" ] }, { "page": "op_any", "title": "Test whether any array element along a given axis evaluates to 'TRUE'.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_any" ] }, { "page": "op_append", "title": "Append tensor 'x2' to the end of tensor 'x1'.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_append" ] }, { "page": "op_arange", "title": "Return evenly spaced values within a given interval.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_arange" ] }, { "page": "op_arccos", "title": "Trigonometric inverse cosine, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_arccos" ] }, { "page": "op_arccosh", "title": "Inverse hyperbolic cosine, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_arccosh" ] }, { "page": "op_arcsin", "title": "Inverse sine, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_arcsin" ] }, { "page": "op_arcsinh", "title": "Inverse hyperbolic sine, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_arcsinh" ] }, { "page": "op_arctan", "title": "Trigonometric inverse tangent, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_arctan" ] }, { "page": "op_arctan2", "title": "Element-wise arc tangent of 'x1/x2' choosing the quadrant correctly.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_arctan2" ] }, { "page": "op_arctanh", "title": "Inverse hyperbolic tangent, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_arctanh" ] }, { "page": "op_argmax", "title": "Returns the indices of the maximum values along an axis.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_argmax" ] }, { "page": "op_argmin", "title": "Returns the indices of the minimum values along an axis.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_argmin" ] }, { "page": "op_argpartition", "title": "Performs an indirect partition along the given axis.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_argpartition" ] }, { "page": "op_argsort", "title": "Returns the indices that would sort a tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_argsort" ] }, { "page": "op_array", "title": "Create a tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_array" ] }, { "page": "op_associative_scan", "title": "Performs a scan with an associative binary operation, in parallel.", "concept": [ "core ops", "ops" ], "topics": [ "op_associative_scan" ] }, { "page": "op_average", "title": "Compute the weighted average along the specified axis.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_average" ] }, { "page": "op_average_pool", "title": "Average pooling operation.", "concept": [ "nn ops", "ops" ], "topics": [ "op_average_pool" ] }, { "page": "op_bartlett", "title": "Bartlett window function.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_bartlett" ] }, { "page": "op_batch_normalization", "title": "Normalizes 'x' by 'mean' and 'variance'.", "concept": [ "nn ops", "ops" ], "topics": [ "op_batch_normalization" ] }, { "page": "op_binary_crossentropy", "title": "Computes binary cross-entropy loss between target and output tensor.", "concept": [ "nn ops", "ops" ], "topics": [ "op_binary_crossentropy" ] }, { "page": "op_bincount", "title": "Count the number of occurrences of each value in a tensor of integers.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_bincount" ] }, { "page": "op_bitwise_and", "title": "Compute the bit-wise AND of two arrays element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_bitwise_and" ] }, { "page": "op_bitwise_invert", "title": "Compute bit-wise inversion, or bit-wise NOT, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_bitwise_invert" ] }, { "page": "op_bitwise_left_shift", "title": "Shift the bits of an integer to the left.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_bitwise_left_shift" ] }, { "page": "op_bitwise_not", "title": "Compute bit-wise inversion, or bit-wise NOT, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_bitwise_not" ] }, { "page": "op_bitwise_or", "title": "Compute the bit-wise OR of two arrays element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_bitwise_or" ] }, { "page": "op_bitwise_right_shift", "title": "Shift the bits of an integer to the right.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_bitwise_right_shift" ] }, { "page": "op_bitwise_xor", "title": "Compute the bit-wise XOR of two arrays element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_bitwise_xor" ] }, { "page": "op_blackman", "title": "Blackman window function.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_blackman" ] }, { "page": "op_broadcast_to", "title": "Broadcast a tensor to a new shape.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_broadcast_to" ] }, { "page": "op_cast", "title": "Cast a tensor to the desired dtype.", "concept": [ "core ops", "ops" ], "topics": [ "op_cast" ] }, { "page": "op_categorical_crossentropy", "title": "Computes categorical cross-entropy loss between target and output tensor.", "concept": [ "nn ops", "ops" ], "topics": [ "op_categorical_crossentropy" ] }, { "page": "op_cbrt", "title": "Computes the cube root of the input tensor, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_cbrt" ] }, { "page": "op_ceil", "title": "Return the ceiling of the input, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_ceil" ] }, { "page": "op_celu", "title": "Continuously-differentiable exponential linear unit.", "concept": [ "nn ops", "ops" ], "topics": [ "op_celu" ] }, { "page": "op_cholesky", "title": "Computes the Cholesky decomposition of a positive semi-definite matrix.", "concept": [ "linear algebra ops", "ops" ], "topics": [ "op_cholesky" ] }, { "page": "op_clip", "title": "Clip (limit) the values in a tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_clip" ] }, { "page": "op_concatenate", "title": "Join a sequence of tensors along an existing axis.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_concatenate" ] }, { "page": "op_cond", "title": "Conditionally applies 'true_fn' or 'false_fn'.", "concept": [ "core ops", "ops" ], "topics": [ "op_cond" ] }, { "page": "op_conj", "title": "Returns the complex conjugate, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_conj" ] }, { "page": "op_conv", "title": "General N-D convolution.", "concept": [ "nn ops", "ops" ], "topics": [ "op_conv" ] }, { "page": "op_conv_transpose", "title": "General N-D convolution transpose.", "concept": [ "nn ops", "ops" ], "topics": [ "op_conv_transpose" ] }, { "page": "op_convert_to_numpy", "title": "Convert a tensor to an R or NumPy array.", "concept": [ "core ops", "ops" ], "topics": [ "op_convert_to_array", "op_convert_to_numpy" ] }, { "page": "op_convert_to_tensor", "title": "Convert an array to a tensor.", "concept": [ "core ops", "ops" ], "topics": [ "op_convert_to_tensor" ] }, { "page": "op_copy", "title": "Returns a copy of 'x'.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_copy" ] }, { "page": "op_corrcoef", "title": "Compute the Pearson correlation coefficient matrix.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_corrcoef" ] }, { "page": "op_correlate", "title": "Compute the cross-correlation of two 1-dimensional tensors.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_correlate" ] }, { "page": "op_cos", "title": "Cosine, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_cos" ] }, { "page": "op_cosh", "title": "Hyperbolic cosine, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_cosh" ] }, { "page": "op_count_nonzero", "title": "Counts the number of non-zero values in 'x' along the given 'axis'.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_count_nonzero" ] }, { "page": "op_cross", "title": "Returns the cross product of two (arrays of) vectors.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_cross" ] }, { "page": "op_ctc_decode", "title": "Decodes the output of a CTC model.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_ctc_decode" ] }, { "page": "op_ctc_loss", "title": "CTC (Connectionist Temporal Classification) loss.", "concept": [ "nn ops", "ops" ], "topics": [ "op_ctc_loss" ] }, { "page": "op_cumprod", "title": "Return the cumulative product of elements along a given axis.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_cumprod" ] }, { "page": "op_cumsum", "title": "Returns the cumulative sum of elements along a given axis.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_cumsum" ] }, { "page": "op_custom_gradient", "title": "Decorator to define a function with a custom gradient.", "concept": [ "core ops", "ops" ], "topics": [ "op_custom_gradient" ] }, { "page": "op_deg2rad", "title": "Convert angles from degrees to radians.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_deg2rad" ] }, { "page": "op_depthwise_conv", "title": "General N-D depthwise convolution.", "concept": [ "nn ops", "ops" ], "topics": [ "op_depthwise_conv" ] }, { "page": "op_det", "title": "Computes the determinant of a square tensor.", "concept": [ "linear algebra ops", "ops" ], "topics": [ "op_det" ] }, { "page": "op_diag", "title": "Extract a diagonal or construct a diagonal array.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_diag" ] }, { "page": "op_diagflat", "title": "Create a two-dimensional array with the flattened input diagonal.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_diagflat" ] }, { "page": "op_diagonal", "title": "Return specified diagonals.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_diagonal" ] }, { "page": "op_diff", "title": "Calculate the n-th discrete difference along the given axis.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_diff" ] }, { "page": "op_digitize", "title": "Returns the indices of the bins to which each value in 'x' belongs.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_digitize" ] }, { "page": "op_divide", "title": "Divide arguments element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_divide" ] }, { "page": "op_divide_no_nan", "title": "Safe element-wise division which returns 0 where the denominator is 0.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_divide_no_nan" ] }, { "page": "op_dot", "title": "Dot product of two tensors.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_dot" ] }, { "page": "op_dot_product_attention", "title": "Scaled dot product attention function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_dot_product_attention" ] }, { "page": "op_dtype", "title": "Return the dtype of the tensor input as a standardized string.", "concept": [ "core ops", "ops" ], "topics": [ "op_dtype" ] }, { "page": "op_eig", "title": "Computes the eigenvalues and eigenvectors of a square matrix.", "concept": [ "linear algebra ops", "ops" ], "topics": [ "op_eig" ] }, { "page": "op_eigh", "title": "Computes the eigenvalues and eigenvectors of a complex Hermitian.", "concept": [ "linear algebra ops", "ops" ], "topics": [ "op_eigh" ] }, { "page": "op_einsum", "title": "Evaluates the Einstein summation convention on the operands.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_einsum" ] }, { "page": "op_elu", "title": "Exponential Linear Unit activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_elu" ] }, { "page": "op_empty", "title": "Return a tensor of given shape and type filled with uninitialized data.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_empty" ] }, { "page": "op_equal", "title": "Returns '(x1 == x2)' element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_equal" ] }, { "page": "op_erf", "title": "Computes the error function of 'x', element-wise.", "concept": [ "math ops", "ops" ], "topics": [ "op_erf" ] }, { "page": "op_erfinv", "title": "Computes the inverse error function of 'x', element-wise.", "concept": [ "math ops", "ops" ], "topics": [ "op_erfinv" ] }, { "page": "op_exp", "title": "Calculate the exponential of all elements in the input tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_exp" ] }, { "page": "op_exp2", "title": "Calculate the base-2 exponential of all elements in the input tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_exp2" ] }, { "page": "op_expand_dims", "title": "Expand the shape of a tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_expand_dims" ] }, { "page": "op_expm1", "title": "Calculate 'exp(x) - 1' for all elements in the tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_expm1" ] }, { "page": "op_extract_sequences", "title": "Expands the dimension of last axis into sequences of 'sequence_length'.", "concept": [ "math ops", "ops" ], "topics": [ "op_extract_sequences" ] }, { "page": "op_eye", "title": "Return a 2-D tensor with ones on the diagonal and zeros elsewhere.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_eye" ] }, { "page": "op_fft", "title": "Computes the Fast Fourier Transform along last axis of input.", "concept": [ "math ops", "ops" ], "topics": [ "op_fft" ] }, { "page": "op_fft2", "title": "Computes the 2D Fast Fourier Transform along the last two axes of input.", "concept": [ "math ops", "ops" ], "topics": [ "op_fft2" ] }, { "page": "op_flip", "title": "Reverse the order of elements in the tensor along the given axis.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_flip" ] }, { "page": "op_floor", "title": "Return the floor of the input, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_floor" ] }, { "page": "op_floor_divide", "title": "Returns the largest integer smaller or equal to the division of inputs.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_floor_divide" ] }, { "page": "op_fori_loop", "title": "For loop implementation.", "concept": [ "core ops", "ops" ], "topics": [ "op_fori_loop" ] }, { "page": "op_full", "title": "Return a new tensor of given shape and type, filled with 'fill_value'.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_full" ] }, { "page": "op_full_like", "title": "Return a full tensor with the same shape and type as the given tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_full_like" ] }, { "page": "op_gelu", "title": "Gaussian Error Linear Unit (GELU) activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_gelu" ] }, { "page": "op_get_item", "title": "Return 'x[key]'.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_get_item" ] }, { "page": "op_glu", "title": "Gated Linear Unit (GLU) activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_glu" ] }, { "page": "op_greater", "title": "Return the truth value of 'x1 > x2' element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_greater" ] }, { "page": "op_greater_equal", "title": "Return the truth value of 'x1 >= x2' element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_greater_equal" ] }, { "page": "op_hamming", "title": "Hamming window function.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_hamming" ] }, { "page": "op_hanning", "title": "Hanning window function.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_hanning" ] }, { "page": "op_hard_shrink", "title": "Hard Shrink activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_hard_shrink" ] }, { "page": "op_hard_sigmoid", "title": "Hard sigmoid activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_hard_sigmoid" ] }, { "page": "op_hard_silu", "title": "Hard SiLU activation function, also known as Hard Swish.", "concept": [ "nn ops", "ops" ], "topics": [ "op_hard_silu", "op_hard_swish" ] }, { "page": "op_hard_tanh", "title": "Applies the HardTanh function element-wise.", "concept": [ "nn ops", "ops" ], "topics": [ "op_hard_tanh" ] }, { "page": "op_heaviside", "title": "Heaviside step function.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_heaviside" ] }, { "page": "op_histogram", "title": "Computes a histogram of the data tensor 'x'.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_histogram" ] }, { "page": "op_hstack", "title": "Stack tensors in sequence horizontally (column wise).", "concept": [ "numpy ops", "ops" ], "topics": [ "op_hstack" ] }, { "page": "op_identity", "title": "Return the identity tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_identity" ] }, { "page": "op_ifft2", "title": "Computes the 2D Inverse Fast Fourier Transform along the last two axes of", "concept": [ "math ops", "ops" ], "topics": [ "op_ifft2" ] }, { "page": "op_imag", "title": "Return the imaginary part of the complex argument.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_imag" ] }, { "page": "op_image_affine_transform", "title": "Applies the given transform(s) to the image(s).", "concept": [ "image ops", "image utils", "ops" ], "topics": [ "op_image_affine_transform" ] }, { "page": "op_image_crop", "title": "Crop 'images' to a specified 'height' and 'width'.", "concept": [ "image ops", "image utils", "ops" ], "topics": [ "op_image_crop" ] }, { "page": "op_image_elastic_transform", "title": "Applies elastic deformation to the image(s).", "topics": [ "op_image_elastic_transform" ] }, { "page": "op_image_extract_patches", "title": "Extracts patches from the image(s).", "concept": [ "image ops", "image utils", "ops" ], "topics": [ "op_image_extract_patches" ] }, { "page": "op_image_gaussian_blur", "title": "Applies a Gaussian blur to the image(s).", "concept": [ "image ops", "image utils", "ops" ], "topics": [ "op_image_gaussian_blur" ] }, { "page": "op_image_hsv_to_rgb", "title": "Convert HSV images to RGB.", "concept": [ "image ops", "image utils", "ops" ], "topics": [ "op_image_hsv_to_rgb" ] }, { "page": "op_image_map_coordinates", "title": "Map the input array to new coordinates by interpolation.", "concept": [ "image ops", "image utils", "ops" ], "topics": [ "op_image_map_coordinates" ] }, { "page": "op_image_pad", "title": "Pad 'images' with zeros to the specified 'height' and 'width'.", "concept": [ "image ops", "image utils", "ops" ], "topics": [ "op_image_pad" ] }, { "page": "op_image_perspective_transform", "title": "Applies a perspective transformation to the image(s).", "concept": [ "image ops", "image utils", "ops" ], "topics": [ "op_image_perspective_transform" ] }, { "page": "op_image_resize", "title": "Resize images to size using the specified interpolation method.", "concept": [ "image ops", "image utils", "ops" ], "topics": [ "op_image_resize" ] }, { "page": "op_image_rgb_to_grayscale", "title": "Convert RGB images to grayscale.", "concept": [ "image ops", "image utils", "ops" ], "topics": [ "op_image_rgb_to_grayscale" ] }, { "page": "op_image_rgb_to_hsv", "title": "Convert RGB images to HSV.", "concept": [ "image ops", "image utils", "ops" ], "topics": [ "op_image_rgb_to_hsv" ] }, { "page": "op_in_top_k", "title": "Checks if the targets are in the top-k predictions.", "concept": [ "math ops", "ops" ], "topics": [ "op_in_top_k" ] }, { "page": "op_inner", "title": "Return the inner product of two tensors.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_inner" ] }, { "page": "op_inv", "title": "Computes the inverse of a square tensor.", "concept": [ "linear algebra ops", "ops" ], "topics": [ "op_inv" ] }, { "page": "op_irfft", "title": "Inverse real-valued Fast Fourier transform along the last axis.", "concept": [ "math ops", "ops" ], "topics": [ "op_irfft" ] }, { "page": "op_is_tensor", "title": "Check whether the given object is a tensor.", "concept": [ "core ops", "ops" ], "topics": [ "op_is_tensor" ] }, { "page": "op_isclose", "title": "Return whether two tensors are element-wise almost equal.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_isclose" ] }, { "page": "op_isfinite", "title": "Return whether a tensor is finite, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_isfinite" ] }, { "page": "op_isinf", "title": "Test element-wise for positive or negative infinity.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_isinf" ] }, { "page": "op_isnan", "title": "Test element-wise for NaN and return result as a boolean tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_isnan" ] }, { "page": "op_istft", "title": "Inverse Short-Time Fourier Transform along the last axis of the input.", "concept": [ "math ops", "ops" ], "topics": [ "op_istft" ] }, { "page": "op_kaiser", "title": "Kaiser window function.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_kaiser" ] }, { "page": "op_layer_normalization", "title": "Layer normalization (Ba et al., 2016).", "concept": [ "nn ops", "ops" ], "topics": [ "op_layer_normalization" ] }, { "page": "op_leaky_relu", "title": "Leaky version of a Rectified Linear Unit activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_leaky_relu" ] }, { "page": "op_left_shift", "title": "Shift the bits of an integer to the left.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_left_shift" ] }, { "page": "op_less", "title": "Return the truth value of 'x1 < x2' element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_less" ] }, { "page": "op_less_equal", "title": "Return the truth value of 'x1 <= x2' element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_less_equal" ] }, { "page": "op_linspace", "title": "Return evenly spaced numbers over a specified interval.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_linspace" ] }, { "page": "op_log", "title": "Natural logarithm, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_log" ] }, { "page": "op_log_sigmoid", "title": "Logarithm of the sigmoid activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_log_sigmoid" ] }, { "page": "op_log_softmax", "title": "Log-softmax activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_log_softmax" ] }, { "page": "op_log10", "title": "Return the base 10 logarithm of the input tensor, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_log10" ] }, { "page": "op_log1p", "title": "Returns the natural logarithm of one plus the 'x', element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_log1p" ] }, { "page": "op_log2", "title": "Base-2 logarithm of 'x', element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_log2" ] }, { "page": "op_logaddexp", "title": "Logarithm of the sum of exponentiations of the inputs.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_logaddexp" ] }, { "page": "op_logdet", "title": "Computes log of the determinant of a hermitian positive definite matrix.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_logdet" ] }, { "page": "op_logical_and", "title": "Computes the element-wise logical AND of the given input tensors.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_logical_and" ] }, { "page": "op_logical_not", "title": "Computes the element-wise NOT of the given input tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_logical_not" ] }, { "page": "op_logical_or", "title": "Computes the element-wise logical OR of the given input tensors.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_logical_or" ] }, { "page": "op_logical_xor", "title": "Compute the truth value of x1 XOR x2, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_logical_xor" ] }, { "page": "op_logspace", "title": "Returns numbers spaced evenly on a log scale.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_logspace" ] }, { "page": "op_logsumexp", "title": "Computes the logarithm of sum of exponentials of elements in a tensor.", "concept": [ "math ops", "ops" ], "topics": [ "op_logsumexp" ] }, { "page": "op_lstsq", "title": "Return the least-squares solution to a linear matrix equation.", "concept": [ "linear algebra ops", "numpy ops", "ops" ], "topics": [ "op_lstsq" ] }, { "page": "op_lu_factor", "title": "Computes the lower-upper decomposition of a square matrix.", "concept": [ "linear algebra ops", "ops" ], "topics": [ "op_lu_factor" ] }, { "page": "op_map", "title": "Map a function over leading array axes.", "concept": [ "core ops", "ops" ], "topics": [ "op_map" ] }, { "page": "op_matmul", "title": "Matrix product of two tensors.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_matmul" ] }, { "page": "op_max", "title": "Return the maximum of a tensor or maximum along an axis.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_amax", "op_max" ] }, { "page": "op_max_pool", "title": "Max pooling operation.", "concept": [ "nn ops", "ops" ], "topics": [ "op_max_pool" ] }, { "page": "op_maximum", "title": "Element-wise maximum of 'x1' and 'x2'.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_maximum", "op_pmax" ] }, { "page": "op_mean", "title": "Compute the arithmetic mean along the specified axes.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_mean" ] }, { "page": "op_median", "title": "Compute the median along the specified axis.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_median" ] }, { "page": "op_meshgrid", "title": "Creates grids of coordinates from coordinate vectors.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_meshgrid" ] }, { "page": "op_min", "title": "Return the minimum of a tensor or minimum along an axis.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_amin", "op_min" ] }, { "page": "op_minimum", "title": "Element-wise minimum of 'x1' and 'x2'.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_minimum", "op_pmin" ] }, { "page": "op_mod", "title": "Returns the element-wise remainder of division.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_mod" ] }, { "page": "op_moments", "title": "Calculates the mean and variance of 'x'.", "concept": [ "nn ops", "ops" ], "topics": [ "op_moments" ] }, { "page": "op_moveaxis", "title": "Move axes of a tensor to new positions.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_moveaxis" ] }, { "page": "op_multi_hot", "title": "Encodes integer labels as multi-hot vectors.", "concept": [ "nn ops", "ops" ], "topics": [ "op_multi_hot" ] }, { "page": "op_multiply", "title": "Multiply arguments element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_multiply" ] }, { "page": "op_nan_to_num", "title": "Replace NaN with zero and infinity with large finite numbers.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_nan_to_num" ] }, { "page": "op_ndim", "title": "Return the number of dimensions of a tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_ndim" ] }, { "page": "op_negative", "title": "Numerical negative, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_negative" ] }, { "page": "op_nonzero", "title": "Return the indices of the elements that are non-zero.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_nonzero" ] }, { "page": "op_norm", "title": "Matrix or vector norm.", "concept": [ "linear algebra ops", "ops" ], "topics": [ "op_norm" ] }, { "page": "op_normalize", "title": "Normalizes 'x' over the specified axis.", "concept": [ "nn ops", "ops" ], "topics": [ "op_normalize" ] }, { "page": "op_not_equal", "title": "Return '(x1 != x2)' element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_not_equal" ] }, { "page": "op_one_hot", "title": "Converts integer tensor 'x' into a one-hot tensor.", "concept": [ "nn ops", "ops" ], "topics": [ "op_one_hot" ] }, { "page": "op_ones", "title": "Return a new tensor of given shape and type, filled with ones.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_ones" ] }, { "page": "op_ones_like", "title": "Return a tensor of ones with the same shape and type of 'x'.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_ones_like" ] }, { "page": "op_outer", "title": "Compute the outer product of two vectors.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_outer" ] }, { "page": "op_pad", "title": "Pad a tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_pad" ] }, { "page": "op_polar", "title": "Constructs a complex tensor whose elements are Cartesian", "concept": [ "nn ops", "ops" ], "topics": [ "op_polar" ] }, { "page": "op_power", "title": "First tensor elements raised to powers from second tensor, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_power" ] }, { "page": "op_prod", "title": "Return the product of tensor elements over a given axis.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_prod" ] }, { "page": "op_psnr", "title": "Peak Signal-to-Noise Ratio (PSNR) function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_psnr" ] }, { "page": "op_qr", "title": "Computes the QR decomposition of a tensor.", "concept": [ "math ops", "ops" ], "topics": [ "op_qr" ] }, { "page": "op_quantile", "title": "Compute the q-th quantile(s) of the data along the specified axis.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_quantile" ] }, { "page": "op_ravel", "title": "Return a contiguous flattened tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_ravel" ] }, { "page": "op_real", "title": "Return the real part of the complex argument.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_real" ] }, { "page": "op_rearrange", "title": "Rearranges the axes of a Keras tensor according to a specified pattern,", "concept": [ "core ops", "ops" ], "topics": [ "op_rearrange" ] }, { "page": "op_reciprocal", "title": "Return the reciprocal of the argument, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_reciprocal" ] }, { "page": "op_relu", "title": "Rectified linear unit activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_relu" ] }, { "page": "op_relu6", "title": "Rectified linear unit activation function with upper bound of 6.", "concept": [ "nn ops", "ops" ], "topics": [ "op_relu6" ] }, { "page": "op_repeat", "title": "Repeat each element of a tensor after themselves.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_repeat" ] }, { "page": "op_reshape", "title": "Gives a new shape to a tensor without changing its data.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_reshape" ] }, { "page": "op_rfft", "title": "Real-valued Fast Fourier Transform along the last axis of the input.", "concept": [ "math ops", "ops" ], "topics": [ "op_rfft" ] }, { "page": "op_right_shift", "title": "Shift the bits of an integer to the right.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_right_shift" ] }, { "page": "op_rms_normalization", "title": "Performs Root Mean Square (RMS) normalization on 'x'.", "concept": [ "nn ops", "ops" ], "topics": [ "op_rms_normalization" ] }, { "page": "op_roll", "title": "Roll tensor elements along a given axis.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_roll" ] }, { "page": "op_rot90", "title": "Rotate an array by 90 degrees in the plane specified by axes.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_rot90" ] }, { "page": "op_round", "title": "Evenly round to the given number of decimals.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_round" ] }, { "page": "op_rsqrt", "title": "Computes reciprocal of square root of x element-wise.", "concept": [ "math ops", "ops" ], "topics": [ "op_rsqrt" ] }, { "page": "op_saturate_cast", "title": "Performs a safe saturating cast to the desired dtype.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_saturate_cast" ] }, { "page": "op_scan", "title": "Scan a function over leading array axes while carrying along state.", "concept": [ "core ops", "ops" ], "topics": [ "op_scan" ] }, { "page": "op_scatter", "title": "Returns a tensor of shape 'shape' where 'indices' are set to 'values'.", "concept": [ "core ops", "ops" ], "topics": [ "op_scatter" ] }, { "page": "op_scatter_update", "title": "Update inputs via updates at scattered (sparse) indices.", "concept": [ "core ops", "ops" ], "topics": [ "op_scatter_update" ] }, { "page": "op_searchsorted", "title": "Perform a binary search.", "concept": [ "core ops", "ops" ], "topics": [ "op_searchsorted" ] }, { "page": "op_segment_max", "title": "Computes the max of segments in a tensor.", "concept": [ "math ops", "ops" ], "topics": [ "op_segment_max" ] }, { "page": "op_segment_sum", "title": "Computes the sum of segments in a tensor.", "concept": [ "math ops", "ops" ], "topics": [ "op_segment_sum" ] }, { "page": "op_select", "title": "Return elements from 'choicelist', based on conditions in 'condlist'.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_select" ] }, { "page": "op_selu", "title": "Scaled Exponential Linear Unit (SELU) activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_selu" ] }, { "page": "op_separable_conv", "title": "General N-D separable convolution.", "concept": [ "nn ops", "ops" ], "topics": [ "op_separable_conv" ] }, { "page": "op_shape", "title": "Gets the shape of the tensor input.", "concept": [ "core ops", "ops" ], "topics": [ "op_shape" ] }, { "page": "op_sigmoid", "title": "Sigmoid activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_sigmoid" ] }, { "page": "op_sign", "title": "Returns a tensor with the signs of the elements of 'x'.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_sign" ] }, { "page": "op_signbit", "title": "Return the sign bit of the elements of 'x'.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_signbit" ] }, { "page": "op_silu", "title": "Sigmoid Linear Unit (SiLU) activation function, also known as Swish.", "concept": [ "nn ops", "ops" ], "topics": [ "op_silu" ] }, { "page": "op_sin", "title": "Trigonometric sine, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_sin" ] }, { "page": "op_sinh", "title": "Hyperbolic sine, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_sinh" ] }, { "page": "op_size", "title": "Return the number of elements in a tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_size" ] }, { "page": "op_slice", "title": "Return a slice of an input tensor.", "concept": [ "core ops", "ops" ], "topics": [ "op_slice" ] }, { "page": "op_slice_update", "title": "Update an input by slicing in a tensor of updated values.", "concept": [ "core ops", "ops" ], "topics": [ "op_slice_update" ] }, { "page": "op_slogdet", "title": "Compute the sign and natural logarithm of the determinant of a matrix.", "concept": [ "linear algebra ops", "ops" ], "topics": [ "op_slogdet" ] }, { "page": "op_soft_shrink", "title": "Soft Shrink activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_soft_shrink" ] }, { "page": "op_softmax", "title": "Softmax activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_softmax" ] }, { "page": "op_softplus", "title": "Softplus activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_softplus" ] }, { "page": "op_softsign", "title": "Softsign activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_softsign" ] }, { "page": "op_solve", "title": "Solves a linear system of equations given by a x = b.", "concept": [ "math ops", "ops" ], "topics": [ "op_solve" ] }, { "page": "op_solve_triangular", "title": "Solves a linear system of equations given by 'a %*% x = b'.", "concept": [ "linear algebra ops", "ops" ], "topics": [ "op_solve_triangular" ] }, { "page": "op_sort", "title": "Sorts the elements of 'x' along a given axis in ascending order.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_sort" ] }, { "page": "op_sparse_categorical_crossentropy", "title": "Computes sparse categorical cross-entropy loss.", "concept": [ "nn ops", "ops" ], "topics": [ "op_sparse_categorical_crossentropy" ] }, { "page": "op_sparse_plus", "title": "SparsePlus activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_sparse_plus" ] }, { "page": "op_sparse_sigmoid", "title": "Sparse sigmoid activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_sparse_sigmoid" ] }, { "page": "op_sparsemax", "title": "Sparsemax activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_sparsemax" ] }, { "page": "op_split", "title": "Split a tensor into chunks.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_split" ] }, { "page": "op_sqrt", "title": "Return the non-negative square root of a tensor, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_sqrt" ] }, { "page": "op_square", "title": "Return the element-wise square of the input.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_square" ] }, { "page": "op_squareplus", "title": "Squareplus activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_squareplus" ] }, { "page": "op_squeeze", "title": "Remove axes of length one from 'x'.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_squeeze" ] }, { "page": "op_stack", "title": "Join a sequence of tensors along a new axis.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_stack" ] }, { "page": "op_std", "title": "Compute the standard deviation along the specified axis.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_std" ] }, { "page": "op_stft", "title": "Short-Time Fourier Transform along the last axis of the input.", "concept": [ "math ops", "ops" ], "topics": [ "op_stft" ] }, { "page": "op_stop_gradient", "title": "Stops gradient computation.", "concept": [ "core ops", "ops" ], "topics": [ "op_stop_gradient" ] }, { "page": "op_subset", "title": "Subset elements from a tensor", "concept": [ "core ops", "ops" ], "topics": [ "op_subset", "op_subset<-", "op_subset_set" ] }, { "page": "op_subtract", "title": "Subtract arguments element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_subtract" ] }, { "page": "op_sum", "title": "Sum of a tensor over the given axes.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_sum" ] }, { "page": "op_svd", "title": "Computes the singular value decomposition of a matrix.", "concept": [ "linear algebra ops", "ops" ], "topics": [ "op_svd" ] }, { "page": "op_swapaxes", "title": "Interchange two axes of a tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_swapaxes" ] }, { "page": "op_switch", "title": "Apply exactly one of the 'branches' given by 'index'.", "concept": [ "core ops", "ops" ], "topics": [ "op_switch" ] }, { "page": "op_take", "title": "Take elements from a tensor along an axis.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_take" ] }, { "page": "op_take_along_axis", "title": "Select values from 'x' at the 1-D 'indices' along the given axis.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_take_along_axis" ] }, { "page": "op_tan", "title": "Compute tangent, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_tan" ] }, { "page": "op_tanh", "title": "Hyperbolic tangent, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_tanh" ] }, { "page": "op_tanh_shrink", "title": "Applies the tanh shrink function element-wise.", "concept": [ "nn ops", "ops" ], "topics": [ "op_tanh_shrink" ] }, { "page": "op_tensordot", "title": "Compute the tensor dot product along specified axes.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_tensordot" ] }, { "page": "op_threshold", "title": "Threshold activation function.", "concept": [ "nn ops", "ops" ], "topics": [ "op_threshold" ] }, { "page": "op_tile", "title": "Repeat 'x' the number of times given by 'repeats'.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_tile" ] }, { "page": "op_top_k", "title": "Finds the top-k values and their indices in a tensor.", "concept": [ "math ops", "ops" ], "topics": [ "op_top_k" ] }, { "page": "op_trace", "title": "Return the sum along diagonals of the tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_trace" ] }, { "page": "op_transpose", "title": "Returns a tensor with 'axes' transposed.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_transpose" ] }, { "page": "op_tri", "title": "Return a tensor with ones at and below a diagonal and zeros elsewhere.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_tri" ] }, { "page": "op_tril", "title": "Return lower triangle of a tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_tril" ] }, { "page": "op_triu", "title": "Return upper triangle of a tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_triu" ] }, { "page": "op_trunc", "title": "Return the truncated value of the input, element-wise.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_trunc" ] }, { "page": "op_unravel_index", "title": "Convert flat indices to coordinate arrays in a given array shape.", "concept": [ "nn ops", "ops" ], "topics": [ "op_unravel_index" ] }, { "page": "op_unstack", "title": "Unpacks the given dimension of a rank-R tensor into rank-(R-1) tensors.", "concept": [ "core ops", "ops" ], "topics": [ "op_unstack" ] }, { "page": "op_var", "title": "Compute the variance along the specified axes.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_var" ] }, { "page": "op_vdot", "title": "Return the dot product of two vectors.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_vdot" ] }, { "page": "op_vectorize", "title": "Turn a function into a vectorized function.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_vectorize" ] }, { "page": "op_vectorized_map", "title": "Parallel map of function 'f' on the first axis of tensor(s) 'elements'.", "concept": [ "core ops", "ops" ], "topics": [ "op_vectorized_map" ] }, { "page": "op_view_as_complex", "title": "Convert a real tensor with two channels into a complex tensor.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_view_as_complex" ] }, { "page": "op_view_as_real", "title": "Convert a complex tensor into a stacked real representation.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_view_as_real" ] }, { "page": "op_vstack", "title": "Stack tensors in sequence vertically (row wise).", "concept": [ "numpy ops", "ops" ], "topics": [ "op_vstack" ] }, { "page": "op_where", "title": "Return elements chosen from 'x1' or 'x2' depending on 'condition'.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_where" ] }, { "page": "op_while_loop", "title": "While loop implementation.", "concept": [ "core ops", "ops" ], "topics": [ "op_while_loop" ] }, { "page": "op_zeros", "title": "Return a new tensor of given shape and type, filled with zeros.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_zeros" ] }, { "page": "op_zeros_like", "title": "Return a tensor of zeros with the same shape and type as 'x'.", "concept": [ "numpy ops", "ops" ], "topics": [ "op_zeros_like" ] }, { "page": "optimizer_adadelta", "title": "Optimizer that implements the Adadelta algorithm.", "concept": [ "optimizers" ], "topics": [ "optimizer_adadelta" ] }, { "page": "optimizer_adafactor", "title": "Optimizer that implements the Adafactor algorithm.", "concept": [ "optimizers" ], "topics": [ "optimizer_adafactor" ] }, { "page": "optimizer_adagrad", "title": "Optimizer that implements the Adagrad algorithm.", "concept": [ "optimizers" ], "topics": [ "optimizer_adagrad" ] }, { "page": "optimizer_adam", "title": "Optimizer that implements the Adam algorithm.", "concept": [ "optimizers" ], "topics": [ "optimizer_adam" ] }, { "page": "optimizer_adam_w", "title": "Optimizer that implements the AdamW algorithm.", "concept": [ "optimizers" ], "topics": [ "optimizer_adam_w" ] }, { "page": "optimizer_adamax", "title": "Optimizer that implements the Adamax algorithm.", "concept": [ "optimizers" ], "topics": [ "optimizer_adamax" ] }, { "page": "optimizer_ftrl", "title": "Optimizer that implements the FTRL algorithm.", "concept": [ "optimizers" ], "topics": [ "optimizer_ftrl" ] }, { "page": "optimizer_lamb", "title": "Optimizer that implements the Lamb algorithm.", "concept": [ "optimizers" ], "topics": [ "optimizer_lamb" ] }, { "page": "optimizer_lion", "title": "Optimizer that implements the Lion algorithm.", "concept": [ "optimizers" ], "topics": [ "optimizer_lion" ] }, { "page": "optimizer_loss_scale", "title": "An optimizer that dynamically scales the loss to prevent underflow.", "concept": [ "optimizers" ], "topics": [ "optimizer_loss_scale" ] }, { "page": "optimizer_muon", "title": "Optimizer that implements the Muon algorithm.", "concept": [ "optimizers" ], "topics": [ "optimizer_muon" ] }, { "page": "optimizer_nadam", "title": "Optimizer that implements the Nadam algorithm.", "concept": [ "optimizers" ], "topics": [ "optimizer_nadam" ] }, { "page": "optimizer_rmsprop", "title": "Optimizer that implements the RMSprop algorithm.", "concept": [ "optimizers" ], "topics": [ "optimizer_rmsprop" ] }, { "page": "optimizer_sgd", "title": "Gradient descent (with momentum) optimizer.", "concept": [ "optimizers" ], "topics": [ "optimizer_sgd" ] }, { "page": "pad_sequences", "title": "Pads sequences to the same length.", "concept": [ "utils" ], "topics": [ "pad_sequences" ] }, { "page": "plot.keras_training_history", "title": "Plot training history", "topics": [ "plot.keras_training_history" ] }, { "page": "plot.keras.src.models.model.Model", "title": "Plot a Keras model", "topics": [ "plot.keras.src.models.model.Model" ] }, { "page": "pop_layer", "title": "Remove the last layer in a Sequential model", "concept": [ "model functions" ], "topics": [ "pop_layer" ] }, { "page": "predict_on_batch", "title": "Returns predictions for a single batch of samples.", "concept": [ "model training" ], "topics": [ "predict_on_batch" ] }, { "page": "predict.keras.src.models.model.Model", "title": "Generates output predictions for the input samples.", "concept": [ "model training" ], "topics": [ "predict.keras.src.models.model.Model" ] }, { "page": "process_utils", "title": "Preprocessing and postprocessing utilities", "topics": [ "application_decode_predictions", "application_preprocess_inputs", "process_utils" ] }, { "page": "quantize_weights", "title": "Quantize the weights of a model.", "concept": [ "layer methods" ], "topics": [ "quantize_weights" ] }, { "page": "random_beta", "title": "Draw samples from a Beta distribution.", "concept": [ "random" ], "topics": [ "random_beta" ] }, { "page": "random_binomial", "title": "Draw samples from a Binomial distribution.", "concept": [ "random" ], "topics": [ "random_binomial" ] }, { "page": "random_categorical", "title": "Draws samples from a categorical distribution.", "concept": [ "random" ], "topics": [ "random_categorical" ] }, { "page": "random_dropout", "title": "Randomly set some values in a tensor to 0.", "concept": [ "random" ], "topics": [ "random_dropout" ] }, { "page": "random_gamma", "title": "Draw random samples from the Gamma distribution.", "concept": [ "random" ], "topics": [ "random_gamma" ] }, { "page": "random_integer", "title": "Draw random integers from a uniform distribution.", "concept": [ "random" ], "topics": [ "random_integer" ] }, { "page": "random_normal", "title": "Draw random samples from a normal (Gaussian) distribution.", "concept": [ "random" ], "topics": [ "random_normal" ] }, { "page": "random_seed_generator", "title": "Generates variable seeds upon each call to a function generating random numbers.", "concept": [ "random" ], "topics": [ "random_seed_generator" ] }, { "page": "random_shuffle", "title": "Shuffle the elements of a tensor uniformly at random along an axis.", "concept": [ "random" ], "topics": [ "random_shuffle" ] }, { "page": "random_truncated_normal", "title": "Draw samples from a truncated normal distribution.", "concept": [ "random" ], "topics": [ "random_truncated_normal" ] }, { "page": "random_uniform", "title": "Draw samples from a uniform distribution.", "concept": [ "random" ], "topics": [ "random_uniform" ] }, { "page": "register_keras_serializable", "title": "Registers a custom object with the Keras serialization framework.", "concept": [ "saving and loading functions", "serialization utilities" ], "topics": [ "register_keras_serializable" ] }, { "page": "regularizer_l1", "title": "A regularizer that applies a L1 regularization penalty.", "concept": [ "regularizers" ], "topics": [ "regularizer_l1" ] }, { "page": "regularizer_l1_l2", "title": "A regularizer that applies both L1 and L2 regularization penalties.", "concept": [ "regularizers" ], "topics": [ "regularizer_l1_l2" ] }, { "page": "regularizer_l2", "title": "A regularizer that applies a L2 regularization penalty.", "concept": [ "regularizers" ], "topics": [ "regularizer_l2" ] }, { "page": "regularizer_orthogonal", "title": "Regularizer that encourages input vectors to be orthogonal to each other.", "concept": [ "regularizers" ], "topics": [ "regularizer_orthogonal" ] }, { "page": "reset_state", "title": "Reset the state for a model, layer or metric.", "concept": [ "layer methods" ], "topics": [ "reset_state" ] }, { "page": "rnn_cell_gru", "title": "Cell class for the GRU layer.", "concept": [ "gru rnn layers", "layers", "rnn cells", "rnn layers" ], "topics": [ "rnn_cell_gru" ] }, { "page": "rnn_cell_lstm", "title": "Cell class for the LSTM layer.", "concept": [ "layers", "lstm rnn layers", "rnn cells", "rnn layers" ], "topics": [ "rnn_cell_lstm" ] }, { "page": "rnn_cell_simple", "title": "Cell class for SimpleRNN.", "concept": [ "layers", "rnn cells", "rnn layers", "simple rnn layers" ], "topics": [ "rnn_cell_simple" ] }, { "page": "rnn_cells_stack", "title": "Wrapper allowing a stack of RNN cells to behave as a single cell.", "concept": [ "layers", "rnn layers" ], "topics": [ "rnn_cells_stack" ] }, { "page": "save_model", "title": "Saves a model as a '.keras' file.", "concept": [ "saving and loading functions" ], "topics": [ "save_model" ] }, { "page": "save_model_config", "title": "Save and load model configuration as JSON", "concept": [ "saving and loading functions" ], "topics": [ "load_model_config", "save_model_config" ] }, { "page": "save_model_weights", "title": "Saves all weights to a single file or sharded files.", "concept": [ "saving and loading functions" ], "topics": [ "save_model_weights" ] }, { "page": "serialize_keras_object", "title": "Retrieve the full config by serializing the Keras object.", "concept": [ "serialization utilities" ], "topics": [ "serialize_keras_object" ] }, { "page": "set_random_seed", "title": "Sets all random seeds (Python, NumPy, and backend framework, e.g. TF).", "concept": [ "utils" ], "topics": [ "set_random_seed" ] }, { "page": "set_state_tree", "title": "Assigns values to variables of the model.", "concept": [ "model functions" ], "topics": [ "set_state_tree" ] }, { "page": "shape", "title": "Tensor shape utility", "topics": [ "!=.keras_shape", "==.keras_shape", "as.integer.keras_shape", "as.list.keras_shape", "format.keras_shape", "print.keras_shape", "shape", "Summary.keras_shape", "[.keras_shape" ] }, { "page": "split_dataset", "title": "Splits a dataset into a left half and a right half (e.g. train / test).", "concept": [ "dataset utils", "utils" ], "topics": [ "split_dataset" ] }, { "page": "summary.keras.src.models.model.Model", "title": "Print a summary of a Keras Model", "concept": [ "model functions" ], "topics": [ "format.keras.src.models.model.Model", "print.keras.src.models.model.Model", "summary.keras.src.models.model.Model" ] }, { "page": "test_on_batch", "title": "Test the model on a single batch of samples.", "concept": [ "model training" ], "topics": [ "test_on_batch" ] }, { "page": "text_dataset_from_directory", "title": "Generates a 'tf.data.Dataset' from text files in a directory.", "concept": [ "dataset utils", "preprocessing", "text dataset utils", "utils" ], "topics": [ "text_dataset_from_directory" ] }, { "page": "timeseries_dataset_from_array", "title": "Creates a dataset of sliding windows over a timeseries provided as array.", "concept": [ "dataset utils", "preprocessing", "timesery dataset utils", "utils" ], "topics": [ "timeseries_dataset_from_array" ] }, { "page": "to_categorical", "title": "Converts a class vector (integers) to binary class matrix.", "concept": [ "numerical utils", "utils" ], "topics": [ "to_categorical" ] }, { "page": "train_on_batch", "title": "Runs a single gradient update on a single batch of data.", "concept": [ "model training" ], "topics": [ "train_on_batch" ] }, { "page": "use_backend", "title": "Configure a Keras backend", "topics": [ "use_backend" ] }, { "page": "with_custom_object_scope", "title": "Provide a scope with mappings of names to custom objects", "concept": [ "saving and loading functions", "serialization utilities" ], "topics": [ "with_custom_object_scope" ] }, { "page": "zip_lists", "title": "Zip lists", "concept": [ "data utils", "utils" ], "topics": [ "zip_lists" ] } ], "_readme": "https://github.com/rstudio/keras3/raw/HEAD/README.md", "_rundeps": [ "backports", "base64enc", "cli", "config", "dotty", "fastmap", "generics", "glue", "here", "jsonlite", "lattice", "lifecycle", "magrittr", "Matrix", "png", "processx", "ps", "R6", "rappdirs", "Rcpp", "RcppTOML", "reticulate", "rlang", "rprojroot", "rstudioapi", "tensorflow", "tfautograph", "tfruns", "tidyselect", "vctrs", "whisker", "withr", "yaml", "zeallot" ], "_vignettes": [ { "source": "custom_train_step_in_tensorflow.Rmd", "filename": "custom_train_step_in_tensorflow.html", "title": "Customizing what happens in fit() with TensorFlow", "engine": "knitr::rmarkdown", "headings": [ "Introduction", "Setup", "A first simple example", "Going lower-level", "Supporting sample_weight & class_weight", "Providing your own evaluation step", "Wrapping up: an end-to-end GAN example" ], "created": "2024-01-18 16:17:09", "modified": "2026-04-13 18:13:23", "commits": 21 }, { "source": "distribution.Rmd", "filename": "distribution.html", "title": "Distributed training with Keras 3", "engine": "knitr::rmarkdown", "headings": [ "Introduction", "How it works", "Setup", "DeviceMesh and TensorLayout", "Distribution", "DataParallel", "ModelParallel and LayoutMap", "Further reading" ], "created": "2024-04-02 12:54:52", "modified": "2026-04-13 18:13:23", "commits": 12 }, { "source": "getting_started.Rmd", "filename": "getting_started.html", "title": "Getting Started with Keras", "engine": "knitr::rmarkdown", "headings": [ "Overview", "Installation", "MNIST Example", "Preparing the Data", "Defining the Model", "Training and Evaluation", "Deep Learning with R Book", "Why this name, Keras?" ], "created": "2017-08-29 14:26:50", "modified": "2026-04-13 18:13:23", "commits": 45 }, { "source": "intro_to_keras_for_engineers.Rmd", "filename": "intro_to_keras_for_engineers.html", "title": "Introduction to Keras for engineers", "engine": "knitr::rmarkdown", "headings": [ "Introduction", "Setup", "A first example: A MNIST convnet", "Writing cross-framework custom components", "Training models on arbitrary data sources", "Further reading", "How to customize what happens in fit()", "How to write custom training loops", "How to distribute training" ], "created": "2024-01-18 16:17:09", "modified": "2026-04-13 18:13:23", "commits": 21 }, { "source": "intro_to_keras_for_researchers.Rmd", "filename": "intro_to_keras_for_researchers.html", "title": "Introduction to Keras for Researchers", "engine": "knitr::rmarkdown", "headings": [ "Setup", "Introduction", "Tensors", "Variables", "Doing math in TensorFlow", "Gradients", "Keras layers", "Layer weight creation in build(input_shape)", "Layer gradients", "Trainable and non-trainable weights", "Layers that own layers", "Tracking losses created by layers", "Keeping track of training metrics", "Compiled functions", "Training mode & inference mode", "The Functional API for model-building", "End-to-end experiment example 1: variational autoencoders.", "End-to-end experiment example 2: hypernetworks." ], "created": "2024-01-18 16:17:09", "modified": "2024-07-17 12:25:26", "commits": 6 }, { "source": "making_new_layers_and_models_via_subclassing.Rmd", "filename": "making_new_layers_and_models_via_subclassing.html", "title": "Making new layers and models via subclassing", "engine": "knitr::rmarkdown", "headings": [ "Introduction", "Setup", "The Layer class: the combination of state (weights) and some computation", "Layers can have non-trainable weights", "Best practice: deferring weight creation until the shape of the inputs is known", "Layers are recursively composable", "Backend-agnostic layers and backend-specific layers", "The add_loss() method", "You can optionally enable serialization on your layers", "Privileged training argument in the call() method", "Privileged mask argument in the call() method", "The Model class", "Putting it all together: an end-to-end example" ], "created": "2024-01-18 16:17:09", "modified": "2026-04-13 18:13:23", "commits": 22 }, { "source": "distributed_training_with_tensorflow.Rmd", "filename": "distributed_training_with_tensorflow.html", "title": "Multi-GPU distributed training with TensorFlow", "engine": "knitr::rmarkdown", "headings": [ "Introduction", "Setup", "Single-host, multi-device synchronous training", "Using callbacks to ensure fault tolerance", "tf$data performance tips" ], "created": "2024-01-18 16:17:09", "modified": "2026-04-13 18:13:23", "commits": 16 }, { "source": "serialization_and_saving.Rmd", "filename": "serialization_and_saving.html", "title": "Save, serialize, and export models", "author": "Neel Kovelamudi, Francois Chollet", "engine": "knitr::rmarkdown", "headings": [ "Introduction", "How to save and load a model", "Setup", "Saving", "APIs", "Custom objects", "Registering custom objects (preferred)", "Passing custom objects to load_model()", "Using a custom object scope", "Model serialization", "In-memory model cloning", "get_config() and from_config()", "save_model_config() and load_model_config()", "Arbitrary object serialization and deserialization", "Model weights saving", "APIs for in-memory weight transfer", "APIs for saving weights to disk & loading them back", "Transfer learning example", "Appendix: Handling custom objects", "Defining the config methods", "How custom objects are serialized" ], "created": "2024-01-18 16:17:09", "modified": "2026-02-01 16:20:25", "commits": 10 }, { "source": "functional_api.Rmd", "filename": "functional_api.html", "title": "The Functional API", "engine": "knitr::rmarkdown", "headings": [ "Setup", "Introduction", "Training, evaluation, and inference", "Save and serialize", "Use the same graph of layers to define multiple models", "All models are callable, just like layers", "Manipulate complex graph topologies", "Models with multiple inputs and outputs", "A toy ResNet model", "Shared layers", "Extract and reuse nodes in the graph of layers", "Extend the API using custom layers", "When to use the functional API", "Functional API strengths:", "Less verbose", "Model validation while defining its connectivity graph", "A functional model is plottable and inspectable", "A functional model can be serialized or cloned", "Functional API weakness:", "It does not support dynamic architectures", "Mix-and-match API styles" ], "created": "2017-04-03 17:28:41", "modified": "2026-04-13 18:13:23", "commits": 50 }, { "source": "sequential_model.Rmd", "filename": "sequential_model.html", "title": "The Sequential model", "engine": "knitr::rmarkdown", "headings": [ "Setup", "When to use a Sequential model", "Creating a Sequential model", "Specifying the input shape in advance", "A common debugging workflow: add layers + summary()", "What to do once you have a model", "Feature extraction with a Sequential model", "Transfer learning with a Sequential model" ], "created": "2017-04-03 17:37:25", "modified": "2025-05-02 17:44:47", "commits": 39 }, { "source": "training_with_built_in_methods.Rmd", "filename": "training_with_built_in_methods.html", "title": "Training & evaluation with the built-in methods", "engine": "knitr::rmarkdown", "headings": [ "Setup", "Introduction", "API overview: a first end-to-end example", "The compile() method: specifying a loss, metrics, and an optimizer", "Many built-in optimizers, losses, and metrics are available", "Custom losses", "Custom metrics", "Handling losses and metrics that don't fit the standard signature", "Automatically setting apart a validation holdout set", "Training & evaluation using TF Dataset objects", "Using sample weighting and class weighting", "Class weights", "Sample weights", "Passing data to multi-input, multi-output models", "Using callbacks", "Many built-in callbacks are available", "Writing your own callback", "Checkpointing models", "Using learning rate schedules", "Passing a schedule to an optimizer", "Using callbacks to implement a dynamic learning rate schedule", "Visualizing loss and metrics during training with TensorBoard", "Using the TensorBoard callback" ], "created": "2024-01-18 16:17:09", "modified": "2026-04-13 18:13:23", "commits": 25 }, { "source": "transfer_learning.Rmd", "filename": "transfer_learning.html", "title": "Transfer learning & fine-tuning", "engine": "knitr::rmarkdown", "headings": [ "Setup", "Introduction", "Freezing layers: understanding the trainable attribute", "Recursive setting of the trainable attribute", "The typical transfer-learning workflow", "Fine-tuning", "An end-to-end example: fine-tuning an image classification model on a cats vs. dogs dataset", "Getting the data", "Standardizing the data", "Using random data augmentation", "Build a model", "Train the top layer", "Do a round of fine-tuning of the entire model" ], "created": "2024-01-18 16:17:09", "modified": "2026-04-13 18:13:23", "commits": 22 }, { "source": "understanding_masking_and_padding.Rmd", "filename": "understanding_masking_and_padding.html", "title": "Understanding masking & padding", "author": "Scott Zhu, Francois Chollet", "engine": "knitr::rmarkdown", "headings": [ "Setup", "Introduction", "Padding sequence data", "Masking", "Mask-generating layers: Embedding and Masking", "Mask propagation in the Functional API and Sequential API", "Passing mask tensors directly to layers", "Supporting masking in your custom layers", "Opting-in to mask propagation on compatible layers", "Writing layers that need mask information", "Summary" ], "created": "2024-01-18 16:17:09", "modified": "2024-09-04 19:43:52", "commits": 6 }, { "source": "writing_a_custom_training_loop_in_tensorflow.Rmd", "filename": "writing_a_custom_training_loop_in_tensorflow.html", "title": "Writing a training loop from scratch in TensorFlow", "engine": "knitr::rmarkdown", "headings": [ "Setup", "Introduction", "Using the GradientTape: a first end-to-end example", "Low-level handling of metrics", "Speeding-up your training step with tf_function()", "Low-level handling of losses tracked by the model", "Summary", "End-to-end example: a GAN training loop from scratch" ], "created": "2024-01-18 16:17:09", "modified": "2026-04-13 18:13:23", "commits": 23 }, { "source": "writing_your_own_callbacks.Rmd", "filename": "writing_your_own_callbacks.html", "title": "Writing your own callbacks", "author": "Rick Chao, Francois Chollet", "engine": "knitr::rmarkdown", "headings": [ "Introduction", "Setup", "Keras callbacks overview", "An overview of callback methods", "Global methods", "on_(train|test|predict)_begin(logs = NULL)", "on_(train|test|predict)_end(logs = NULL)", "Batch-level methods for training/testing/predicting", "on_(train|test|predict)_batch_begin(batch, logs = NULL)", "on_(train|test|predict)_batch_end(batch, logs = NULL)", "Epoch-level methods (training only)", "on_epoch_begin(epoch, logs = NULL)", "on_epoch_end(epoch, logs = NULL)", "A basic example", "Usage of logs list", "Usage of self$model attribute", "Examples of Keras callback applications", "Early stopping at minimum loss", "Learning rate scheduling", "Built-in Keras callbacks" ], "created": "2024-01-18 16:17:09", "modified": "2026-02-01 16:20:25", "commits": 10 } ], "_score": 14.721973141155551, "_indexed": true, "_nocasepkg": "keras3", "_universes": [ "rstudio", "t-kalinowski" ], "_binaries": [ { "r": "4.7.0", "os": "linux", "version": "1.5.1.9000", "date": "2026-05-24T07:18:56.000Z", "distro": "noble", "commit": "4571bff1befc425ee812faf247e3732ffc510607", "fileid": "b9cc6137138c91095a022b55d0936ab1ccf5d1051f1360b49aa1caf745d4ca9a", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/rstudio/actions/runs/26354737481" }, { "r": "4.6.0", "os": "linux", "version": "1.5.1.9000", "date": "2026-05-24T07:19:07.000Z", "distro": "noble", "commit": "4571bff1befc425ee812faf247e3732ffc510607", "fileid": "1be42300771c062036dced7636afc2005af13c7bec8194ba0d25b7a7a3d08ae3", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/rstudio/actions/runs/26354737481" }, { "r": "4.5.3", "os": "mac", "version": "1.5.1.9000", "date": "2026-05-24T07:22:06.000Z", "commit": "4571bff1befc425ee812faf247e3732ffc510607", "fileid": "3348391a20e20b283eee9972b521aebd185b91beaad267d8912eed488b3fd4aa", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/rstudio/actions/runs/26354737481" }, { "r": "4.6.0", "os": "mac", "version": "1.5.1.9000", "date": "2026-05-24T07:21:36.000Z", "commit": "4571bff1befc425ee812faf247e3732ffc510607", "fileid": "e80c1b347907ada1631ca8d18588483686b255bb39401cb7e4163f3c4434e890", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/rstudio/actions/runs/26354737481" }, { "r": "4.6.0", "os": "wasm", "version": "1.5.1.9000", "date": "2026-05-24T07:18:59.000Z", "commit": "4571bff1befc425ee812faf247e3732ffc510607", "fileid": "a20e66698c522afc82228919d7b554d92457b42433e5a6dd2c86be87ab5c325c", "status": "success", "buildurl": "https://github.com/r-universe/rstudio/actions/runs/26354737481" }, { "r": "4.7.0", "os": "win", "version": "1.5.1.9000", "date": "2026-05-24T07:18:05.000Z", "commit": "4571bff1befc425ee812faf247e3732ffc510607", "fileid": "ff150ed0433a80f6750549e47c1866a83d6a9876b759b41ef4810a9fe532d1b6", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/rstudio/actions/runs/26354737481" }, { "r": "4.5.3", "os": "win", "version": "1.5.1.9000", "date": "2026-05-24T07:19:09.000Z", "commit": "4571bff1befc425ee812faf247e3732ffc510607", "fileid": "5779b18484c77c6c5421447f6f8e4054c7008bb0197eca647168ce88ee070410", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/rstudio/actions/runs/26354737481" }, { "r": "4.6.0", "os": "win", "version": "1.5.1.9000", "date": "2026-05-24T07:18:01.000Z", "commit": "4571bff1befc425ee812faf247e3732ffc510607", "fileid": "36227e489addc2155f264acbe2d589d5fd7eb363557ab744ea0ded40d5c79131", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/rstudio/actions/runs/26354737481" } ] }