MCPcopy Create free account

hub / github.com/keras-team/keras / functions

Functions14,520 in github.com/keras-team/keras

⨍ Functions 14,520◇ Types & classes 1,698↳ Endpoints 165
↓ 8 callersMethodfrom_config
(cls, config)
keras/src/ops/core.py:1163
↓ 8 callersMethodfrom_config
(cls, config, custom_objects=None)
keras/src/models/model.py:865
↓ 8 callersFunctionget_block_size_for_layer
Determine the block size for int4 quantization. The block size can be specified either through the `config` argument or through the `dtype_po
keras/src/quantizers/quantization_config.py:303
↓ 8 callersMethodget_config
(self)
keras/src/saving/saving_lib_test.py:951
↓ 8 callersMethodget_dropout_mask
(self, step_input)
keras/src/layers/rnn/dropout_rnn_cell.py:37
↓ 8 callersMethodget_layer
Retrieves a layer based on either its name (unique) or index. If `name` and `index` are both provided, `index` will take precedence.
keras/src/models/model.py:182
↓ 8 callersFunctionget_model
(type="sequential", input_shape=(10,), layer_list=None)
keras/src/export/torch_test.py:53
↓ 8 callersMethodhard_shrink
(x)
keras/src/activations/activations_test.py:721
↓ 8 callersFunctionin_stateless_scope
()
keras/src/backend/common/stateless_scope.py:100
↓ 8 callersFunctionrearrange
Rearranges the axes of a Keras tensor according to a specified pattern, einops-style. Args: tensor: Input Keras tensor. patte
keras/src/ops/einops.py:107
↓ 8 callersMethodreset
(self)
keras/src/utils/backend_utils.py:92
↓ 8 callersMethodset_model
(self, model)
keras/src/callbacks/callback.py:74
↓ 8 callersMethodsetup
(self, delta=1.0)
keras/src/losses/losses_test.py:612
↓ 8 callersFunctionshape_equal
Check if two shapes are equal. Args: shape1: A list or tuple of integers for first shape to be compared. shape2: A list or tuple
keras/src/ops/numpy.py:90
↓ 8 callersMethodsoft_shrink
(x, threshold=0.5)
keras/src/activations/activations_test.py:751
↓ 8 callersMethodtanh_shrink
(x)
keras/src/activations/activations_test.py:703
↓ 8 callersFunctionto_snake_case
(name)
keras/src/utils/naming.py:27
↓ 8 callersMethodtree_flatten
(self)
keras/src/utils/tracking.py:424
↓ 8 callersMethodupdate_state
(_, y, y_pred, sample_weight=None)
keras/src/backend/tensorflow/trainer.py:670
↓ 7 callersFunctionEfficientNetV2
Instantiates the EfficientNetV2 architecture using given scaling coefficients. Args: width_coefficient: float, scaling coefficient fo
keras/src/applications/efficientnet_v2.py:823
↓ 7 callersMethod__init__
( self, class_aggregation="uniform_average", num_regressors=0, name="r2_score"
keras/src/metrics/regression_metrics.py:404
↓ 7 callersMethod__init__
(self)
keras/src/models/model_test.py:126
↓ 7 callersMethod_compute_loss
Backwards compatibility wrapper for `compute_loss`. This should be used instead `compute_loss` within `train_step` and `test_step` to
keras/src/trainers/trainer.py:404
↓ 7 callersFunction_convert_conv_transpose_padding_args_from_keras_to_torch
Convert the padding arguments from Keras to the ones used by Torch. Torch starts with an output shape of `(input-1) * stride + kernel_size`, t
keras/src/backend/common/backend_utils.py:53
↓ 7 callersMethod_get_model
(self, compile_model=True)
keras/src/callbacks/tensorboard_test.py:135
↓ 7 callersFunction_promote_binary_op_types
(x1, x2)
keras/src/backend/openvino/numpy.py:27
↓ 7 callersFunction_random_uniform
Wrapper for `ov_opset.random_uniform` that sanitizes seed values. Pybind11 will sign-flip values >= 2**31, and int32 wrap-around can produce
keras/src/backend/openvino/random.py:46
↓ 7 callersFunction_ref_log_softmax
(values)
keras/src/activations/activations_test.py:18
↓ 7 callersMethod_register_call_context_args
Registers call-context args for this layer. If this layer declares a `call()` method that accepts one or more of the given args, thos
keras/src/layers/layer.py:1803
↓ 7 callersMethodbuild
(self, input_shape)
keras/src/layers/layer.py:428
↓ 7 callersMethodbuild
(self, input_shape)
keras/src/layers/convolutional/conv_test.py:726
↓ 7 callersMethodcall
(self, input)
keras/src/export/saved_model_test.py:27
↓ 7 callersFunctioncast
(x, dtype)
keras/src/backend/openvino/core.py:990
↓ 7 callersFunctioncompute_conv_transpose_padding_args_for_jax
( input_shape, kernel_shape, strides, padding, output_padding, dilation_rate, )
keras/src/backend/common/backend_utils.py:119
↓ 7 callersMethodcompute_loss
Compute the total loss, validate it, and return it. Subclasses can optionally override this method to provide custom loss computation
keras/src/trainers/trainer.py:325
↓ 7 callersMethodcompute_metrics
Update metric states and collect all metrics to be returned. Subclasses can optionally override this method to provide custom metric
keras/src/trainers/trainer.py:483
↓ 7 callersMethodenable_lora
( self, rank, lora_alpha=None, a_initializer="he_uniform", b_initializ
keras/src/layers/core/embedding.py:214
↓ 7 callersMethodfrom_config
(cls, config)
keras/src/layers/preprocessing/pipeline.py:70
↓ 7 callersMethodfunction
(iterator)
keras/src/backend/tensorflow/trainer.py:237
↓ 7 callersFunctionget
Retrieves a Keras `DTypePolicy` instance. The `identifier` may be the string name of a `DTypePolicy` class. >>> policy = dtype_policies.get(
keras/src/dtype_policies/__init__.py:65
↓ 7 callersFunctionget_example_array
()
keras/src/constraints/constraints_test.py:8
↓ 7 callersMethodjax_state_sync
(self)
keras/src/backend/jax/trainer.py:880
↓ 7 callersMethodon_predict_batch_end
(self, batch, logs=None)
keras/src/callbacks/callback_list.py:206
↓ 7 callersMethodon_test_batch_end
(self, batch, logs=None)
keras/src/callbacks/callback_list.py:200
↓ 7 callersFunctionsafe_get_h5_group
Retrieve a Group within a given File or Group. Args: parent: the parent h5py.File or h5py.Group. name: the name of the Group to r
keras/src/saving/saving_lib.py:1191
↓ 7 callersMethodsetup
(self, axis=1)
keras/src/losses/losses_test.py:519
↓ 7 callersMethodstateless_update_state
(self, metric_variables, *args, **kwargs)
keras/src/metrics/metric.py:121
↓ 6 callersMethod_assert_sym_convert
( self, value, expected_dtype, expected_item=None, expected_shape=None
keras/src/backend/torch/core_test.py:39
↓ 6 callersMethod_check_reloading_model
(self, ref_input, model)
keras/src/legacy/saving/legacy_h5_format_test.py:117
↓ 6 callersFunction_combine
(a_flat, b_flat)
keras/src/backend/tensorflow/core.py:413
↓ 6 callersFunction_compute_adaptive_pooling_gather_indices
Compute gather indices for Two-Pool Gather method.
keras/src/backend/jax/nn.py:299
↓ 6 callersFunction_compute_static_gather_indices
Compute gather indices for Two-Pool Gather method (corrected).
keras/src/backend/tensorflow/nn.py:290
↓ 6 callersFunction_convert_data_format
(data_format, ndim)
keras/src/backend/tensorflow/nn.py:770
↓ 6 callersFunction_dict_to_ordered_dict
(structure)
keras/src/tree/torchtree_impl.py:23
↓ 6 callersFunction_get_backed_symint
Create a backed SymInt via torch.export dynamic shapes.
keras/src/backend/torch/core_test.py:12
↓ 6 callersMethod_get_jax_state
( self, trainable_variables=False, non_trainable_variables=False, optimizer_va
keras/src/backend/jax/trainer.py:1039
↓ 6 callersFunction_get_test_layer
(layer_type, kernel_shape)
keras/src/quantizers/gptq_test.py:60
↓ 6 callersMethod_is_improvement
(self, monitor_value, reference_value)
keras/src/callbacks/monitor_callback.py:101
↓ 6 callersMethod_overwrite_variable_with_gradient
(self, variable)
keras/src/optimizers/base_optimizer.py:207
↓ 6 callersMethod_restore_h5_file
Ensure the current shard is the last one created.
keras/src/saving/saving_lib.py:1674
↓ 6 callersMethod_run_layer_test
Helper: test layer output shape matches compute_output_shape().
keras/src/layers/pooling/adaptive_pooling3d_test.py:10
↓ 6 callersMethod_run_layer_test
Helper: test layer output shape matches compute_output_shape().
keras/src/layers/pooling/adaptive_pooling2d_test.py:10
↓ 6 callersFunction_run_litert_inference
(interpreter, input_arrays)
keras/src/export/litert_torch_test.py:57
↓ 6 callersMethod_set_monitor_op
(self)
keras/src/callbacks/monitor_callback.py:62
↓ 6 callersMethod_swap_variables
(self)
keras/src/callbacks/swap_ema_weights.py:158
↓ 6 callersMethod_switch_h5_file
Switch to a different H5 file with the specified mode. This is useful for retrieving information from all shards, such as the total l
keras/src/saving/saving_lib.py:1658
↓ 6 callersMethod_token_start_index
(self)
keras/src/layers/preprocessing/index_lookup.py:1011
↓ 6 callersMethod_transform_value_range
Convert input values from `original_range` to `target_range`. This function is intended to be used in preprocessing layers that rely
keras/src/layers/preprocessing/image_preprocessing/base_image_preprocessing_layer.py:268
↓ 6 callersFunction_transpose_spatial_inputs
Transpose inputs from channels_last to channels_first format.
keras/src/backend/torch/nn.py:286
↓ 6 callersFunction_transpose_spatial_outputs
(outputs)
keras/src/backend/torch/nn.py:308
↓ 6 callersMethodactivation_quantizer_or_default
(config, default)
keras/src/quantizers/quantization_config.py:59
↓ 6 callersFunctionall
(x, axis=None, keepdims=False)
keras/src/backend/torch/numpy.py:266
↓ 6 callersMethodassertAllInRange
(self, array, min_val, max_val)
keras/src/layers/preprocessing/image_preprocessing/equalization_test.py:11
↓ 6 callersFunctionawq_quantize_matrix
Quantize a weight matrix using AWQ. This function performs the complete AWQ quantization process: 1. Find optimal per-channel scales via grid
keras/src/quantizers/awq.py:144
↓ 6 callersMethodbuild
(self, _)
keras/src/layers/layer_test.py:1545
↓ 6 callersMethodcatch_stop_iteration
Catches errors when an iterator runs out of data.
keras/src/backend/tensorflow/trainer.py:768
↓ 6 callersMethodclose
(self)
keras/src/saving/saving_lib.py:1624
↓ 6 callersFunctioncompute_conv_transpose_output_shape
( input_shape, kernel_size, filters, strides, padding, output_padding=None, data_f
keras/src/backend/common/backend_utils.py:300
↓ 6 callersFunctiondequantize_with_zero_point
Dequantizes a quantized tensor using the provided scale and zero tensors. Args: input_tensor: KerasTensor. The quantized tensor to d
keras/src/quantizers/quantizers.py:1140
↓ 6 callersFunctionenable_flash_attention
Enable flash attention. Flash attention offers performance optimization for attention layers, making it especially useful for large language
keras/src/backend/config.py:178
↓ 6 callersMethodfrom_config
Instantiates a weight constraint from a configuration dictionary. Example: ```python constraint = UnitNorm() config
keras/src/constraints/constraints.py:60
↓ 6 callersMethodget_config
(self)
keras/src/layers/pooling/base_pooling.py:72
↓ 6 callersMethodget_encoded_features
(self)
keras/src/layers/preprocessing/feature_space.py:554
↓ 6 callersFunctionget_input_signature
Get input signature for model export. Args: model: A Keras Model instance. Returns: Input signature suitable for model expor
keras/src/export/export_utils.py:9
↓ 6 callersMethodget_model
(self, dtype=None)
keras/src/saving/saving_api_test.py:249
↓ 6 callersMethodget_numpy_iterator
(self)
keras/src/trainers/data_adapters/py_dataset_adapter.py:292
↓ 6 callersMethodget_numpy_iterator
(self)
keras/src/trainers/data_adapters/tf_dataset_adapter.py:37
↓ 6 callersMethodget_recurrent_dropout_mask
(self, step_input)
keras/src/layers/rnn/dropout_rnn_cell.py:46
↓ 6 callersFunctionis_nested
(structure)
keras/src/tree/dmtree_impl.py:90
↓ 6 callersFunctionis_tensor
(x)
keras/src/backend/numpy/core.py:64
↓ 6 callersMethodmake
Make a new H5 entry group. This method is only available in write mode. It defers the creation of the H5 entry group until `__setitem
keras/src/saving/saving_lib.py:1355
↓ 6 callersMethodnamed_parameters
( self, prefix="", recurse=True, remove_duplicate=True, )
keras/src/backend/torch/layer.py:28
↓ 6 callersFunctionnp_depthwise_conv2d
( x, kernel_weights, bias_weights, strides, padding, data_format, dilation_rate, )
keras/src/layers/convolutional/depthwise_conv_test.py:88
↓ 6 callersMethodon_predict_batch_begin
(self, batch, logs=None)
keras/src/callbacks/callback_list.py:183
↓ 6 callersMethodon_predict_begin
(self, logs=None)
keras/src/callbacks/callback_list.py:266
↓ 6 callersMethodon_predict_end
(self, logs=None)
keras/src/callbacks/callback_list.py:273
↓ 6 callersMethodon_test_batch_begin
(self, batch, logs=None)
keras/src/callbacks/callback_list.py:178
↓ 6 callersMethodon_test_begin
(self, logs=None)
keras/src/callbacks/callback_list.py:249
↓ 6 callersMethodon_test_end
(self, logs=None)
keras/src/callbacks/callback_list.py:256
↓ 6 callersMethodpop
Removes the last layer in the model. Args: rebuild: `bool`. Whether to rebuild the model after removing the layer. De
keras/src/models/sequential.py:127
← previousnext →301–400 of 14,520, ranked by callers