Deep Learning and Reinforcement Learning Library for Scientists and Engineers
This is a maintenance release. All users are suggested to update.
This release provides high-performance image augmentation API. The API is based on affine transformation. It has been proven useful to offer 80x speed up in augmenting images in the openpose-plus project.
GroupNormLayer
(PR #850)affine_rotation_matrix
(PR #857)affine_horizontal_flip_matrix
(PR #857)affine_vertical_flip_matrix
(PR #857)affine_shift_matrix
(PR #857)affine_shear_matrix
(PR #857)affine_zoom_matrix
(PR #857)affine_transform_cv2
(PR #857)affine_transform_keypoints
(PR #857)examples/data_process/tutorial_fast_affine_transform.py
(PR #857)data_format
tl.prepro.transform_matrix_offset_center
(PR #855)TensorLayer 1.10.x will be the last supported version of TL 1.X, big changes are upcoming and won't preserve backward compatibility. TensorLayer 1.10.x will only be updated with bugfixes on existing features. No additional feature will be implemented in TL 1.10.x
tests\test_timeout.py
has been added to ensure the network creation process does not freeze.tensorlayer/utils.py
with customizable tensorboard directory (PR #819)tests\test_timeout.py
has been added to ensure the network creation process does not freeze.tensorlayer/utils.py
with customizable tensorboard directory (PR #819)This release contains a memory leak issue.
It has been a very busy summer for the TensorLayer team. In this version, we start to support:
Mostly importantly, we decide to open-source a series of neural network application codes that have been used by practitioners. The first batch includes:
At the same time, just want to put a note ahead, we are working very hard towards the TensorLayer 2.0 release in order to synchronize with the coming TensorFlow 2.0.
Enjoy this release, and we would love your feedback!
tl.model.vgg19
(PR #698)tl.logging.contrib.hyperdash
(PR #739)tl.distributed.trainer
(PR #700)prefetch_buffer_size
to the tl.distributed.Trainer
(PR #766)tl.db.TensorHub
(PR #751)tl.files.save_graph
(PR #751)tl.files.load_graph_
(PR #751)tl.files.save_graph_and_params
(PR #751)tl.files.load_graph_and_params
(PR #751)tl.prepro.keypoint_random_xxx
(PR #787)data_format
(PR #809)all_dev
, all_cpu_dev
, and all_gpu_dev
(PR #739)tutorial_models_vgg19
has been introduced to show how to use tl.model.vgg19
(PR #698).tutorial_bipedalwalker_a3c_continuous_action.py
(PR #734, Issue #732)tutorial_models_vgg16
and tutorial_models_vgg19
has been changed the input scale from [0,255] to [0,1](PR #710)tutorial_mnist_distributed_trainer.py
and tutorial_cifar10_distributed_trainer.py
are added to explain the uses of Distributed Trainer (PR #700)tutorial_quanconv_cifar10.py
and tutorial_quanconv_mnist.py
(PR #735)tutorial_work_with_onnx.py
(PR #775)test_layers_convolution.py
, test_layers_normalization.py
, test_layers_core.py
(PR #735)tl.logging.warn
has been deprecated in favor of tl.logging.warning
(PR #739)conv_layers()
has been removed in both vgg16 and vgg19(PR #710)absolute_difference_error
in file: tensorlayer/cost.py
- Issue #753 (PR #759)This version is identical to 1.9.0 but fix the issue with TF 1.10.0: https://stackoverflow.com/questions/51593126/tensorlayer-fails-with-tensorflow-1-10-0rc0#51593186
This version was supposed to be release under version: 1.8.6, due to the large amount of changes introduced in this version, it has been decided to release this version under the version: 1.9.0
tl.alphas
and tl.alphas_like
added following the tf.ones/zeros and tf.zeros_like/ones_like (PR #580)tl.lazy_imports.LazyImport
to import heavy libraries only when necessary (PR #667)tl.act.leaky_relu6
and tl.layers.PRelu6Layer
have been deprecated (PR #686)tl.act.leaky_twice_relu6
and tl.layers.PTRelu6Layer
have been deprecated (PR #686)tl.decorators
API created including deprecated_alias
and private_method
(PR #660)tl.decorators
API enriched with protected_method
(PR #675)tl.decorators
API enriched with deprecated
directly raising warning and modifying documentation (PR #691)AMSGrad
added on Optimizers page added (PR #636)tf.layers
in CNN (PR #686)On the Convergence of Adam and Beyond (ICLR 2018)
(PR #636)all
, all_cpu
, and all_gpu
(PR #660)test_utils_predict.py
added to reproduce and fix issue #288 (PR #566)Layer_DeformableConvolution_Test
added to reproduce issue #572 with deformable convolution (PR #573)Array_Op_Alphas_Test
and Array_Op_Alphas_Like_Test
added to test tensorlayer/array_ops.py
file (PR #580)test_optimizer_amsgrad.py
added to test AMSGrad
optimizer (PR #636)test_logging.py
added to insure robustness of the logging API (PR #645)test_decorators.py
added (PR #660)test_activations.py
added (PR #686)tutorial_tfslim
has been introduced to show how to use SlimNetsLayer
(PR #560).tf.logging.set_verbosity(tf.logging.DEBUG)
tl.logging.set_verbosity(tl.logging.DEBUG)
tl.files
refactored into a directory with numerous files (PR #657)tl.files.voc_dataset
fixed because of original Pascal VOC website was down (PR #657)requirements/
directory (PR #657)tf.identity
as activation is ignored, thus reducing the size of the graph by removing useless operation (PR #667)Layer
Base Class (PR #667)Layer
Base Class now presenting methods to update faultlessly all_layers
, all_params
, and all_drop
(PR #675)tl.layers.core
and added to tl.layers.inputs
(PR #675)prev_layer
into self.inputs
(PR #675)tl.layers.TimeDistributedLayer
argurment args
is deprecated in favor of layer_args
(PR #667)tl.act.leaky_relu
have been deprecated in favor of tf.nn.leaky_relu
(PR #686)assert()
calls remove and replaced by raise AssertionError()
(PR #667)tl.identity
is removed, not used anymore and deprecated for a long time (PR #667)TF.__version__ < "1.6"
have been removed (PR #675)tl.layers.RNNLayer
with inspect
(PR #574)tl.files
with truth value of an empty array is ambiguous (PR #575)tl.utils.predict
fixed - np.hstack
problem in which the results for multiple batches are stacked along axis=1
(PR #566)tl.layers.DeformableConv2d
fixed (PR #573)tl.layers.ConvLSTMLayer
fixed (PR #676)tl.layers.TernaryConv2d
fixed - self.inputs not defined (PR #658)tl.layers.binary._compute_threshold()
(PR #658)tf.logging
replaced by tl.logging
(PR #661)tensorlayer.third_party.roi_pooling.roi_pooling.roi_pooling_ops
is now lazy loaded to prevent systematic error raised (PR #675)tutorial_word2vec_basic.py
saving issue #476 fixed (PR #635)tl.alphas
and tl.alphas_like
added following the tf.ones/zeros and tf.zeros_like/ones_like (PR #580)tl.lazy_imports.LazyImport
to import heavy libraries only when necessary (PR #667)tl.act.leaky_relu6
and tl.layers.PRelu6Layer
have been deprecated (PR #686)tl.act.leaky_twice_relu6
and tl.layers.PTRelu6Layer
have been deprecated (PR #686)tl.decorators
API created including deprecated_alias
and private_method
(PR #660)tl.decorators
API enriched with protected_method
(PR #675)tl.decorators
API enriched with deprecated
directly raising warning and modifying documentation (PR #691)AMSGrad
added on Optimizers page added (PR #636)tf.layers
in CNN (PR #686)On the Convergence of Adam and Beyond (ICLR 2018)
(PR #636)all
, all_cpu
, and all_gpu
(PR #660)test_utils_predict.py
added to reproduce and fix issue #288 (PR #566)Layer_DeformableConvolution_Test
added to reproduce issue #572 with deformable convolution (PR #573)Array_Op_Alphas_Test
and Array_Op_Alphas_Like_Test
added to test tensorlayer/array_ops.py
file (PR #580)test_optimizer_amsgrad.py
added to test AMSGrad
optimizer (PR #636)test_logging.py
added to insure robustness of the logging API (PR #645)test_decorators.py
added (PR #660)test_activations.py
added (PR #686)tutorial_tfslim
has been introduced to show how to use SlimNetsLayer
(PR #560).tf.logging.set_verbosity(tf.logging.DEBUG)
tl.logging.set_verbosity(tl.logging.DEBUG)
tl.files
refactored into a directory with numerous files (PR #657)tl.files.voc_dataset
fixed because of original Pascal VOC website was down (PR #657)requirements/
directory (PR #657)tf.identity
as activation is ignored, thus reducing the size of the graph by removing useless operation (PR #667)Layer
Base Class (PR #667)Layer
Base Class now presenting methods to update faultlessly all_layers
, all_params
, and all_drop
(PR #675)tl.layers.core
and added to tl.layers.inputs
(PR #675)prev_layer
into self.inputs
(PR #675)tl.layers.TimeDistributedLayer
argurment args
is deprecated in favor of layer_args
(PR #667)tl.act.leaky_relu
have been deprecated in favor of tf.nn.leaky_relu
(PR #686)assert()
calls remove and replaced by raise AssertionError()
(PR #667)tl.identity
is removed, not used anymore and deprecated for a long time (PR #667)TF.__version__ < "1.6"
have been removed (PR #675)tl.layers.RNNLayer
with inspect
(PR #574)tl.files
with truth value of an empty array is ambiguous (PR #575)tl.utils.predict
fixed - np.hstack
problem in which the results for multiple batches are stacked along axis=1
(PR #566)tl.layers.DeformableConv2d
fixed (PR #573)tl.layers.ConvLSTMLayer
fixed (PR #676)tl.layers.TernaryConv2d
fixed - self.inputs not defined (PR #658)tl.layers.binary._compute_threshold()
(PR #658)tf.logging
replaced by tl.logging
(PR #661)tensorlayer.third_party.roi_pooling.roi_pooling.roi_pooling_ops
is now lazy loaded to prevent systematic error raised (PR #675)tutorial_word2vec_basic.py
saving issue #476 fixed (PR #635)tl.alphas
and tl.alphas_like
added following the tf.ones/zeros and tf.zeros_like/ones_like (by @DEKHTIARJonathan in #580)tl.lazy_imports.LazyImport
to import heavy libraries only when necessary (by @DEKHTIARJonathan in #667)tl.decorators
API created including deprecated_alias
and private_method
(by @DEKHTIARJonathan in #660)tl.decorators
API enriched with protected_method
(by @DEKHTIARJonathan in #675)AMSGrad
added on Optimizers page added (by @DEKHTIARJonathan in #636)tf.layers
in CNN (by @zsdonghao in #686)On the Convergence of Adam and Beyond (ICLR 2018)
(by @DEKHTIARJonathan in #636)all
, all_cpu
, and all_gpu
(by @DEKHTIARJonathan in #660)test_utils_predict.py
added to reproduce and fix issue #288 (by @2wins in #566)Layer_DeformableConvolution_Test
added to reproduce issue #572 with deformable convolution (by @DEKHTIARJonathan in #573)Array_Op_Alphas_Test
and Array_Op_Alphas_Like_Test
added to test tensorlayer/array_ops.py
file (by @DEKHTIARJonathan in #580)test_optimizer_amsgrad.py
added to test AMSGrad
optimizer (by @DEKHTIARJonathan in #636)test_logging.py
added to insure robustness of the logging API (by @DEKHTIARJonathan in #645)test_decorators.py
added (by @DEKHTIARJonathan in #660)tutorial_tfslim
has been introduced to show how to use SlimNetsLayer
(by @2wins in #560).tl.files
refactored into a directory with numerous files (by @DEKHTIARJonathan in #657)tl.files.voc_dataset
fixed because of original Pascal VOC website was down (by @DEKHTIARJonathan in #657)requirements/
directory (by @DEKHTIARJonathan in #657)tf.identity
as activation is ignored, thus reducing the size of the graph by removing useless operation (by @DEKHTIARJonathan in #667)Layer
Base Class (by @DEKHTIARJonathan in #667)Layer
Base Class now presenting methods to update faultlessly all_layers
, all_params
, and all_drop
(by @DEKHTIARJonathan in #675)tl.layers.core
and added to tl.layers.inputs
(by @DEKHTIARJonathan in #675)prev_layer
into self.inputs
(by @DEKHTIARJonathan in #675)tl.layers.TimeDistributedLayer
argurment args
is deprecated in favor of layer_args
(by @DEKHTIARJonathan in #667)assert()
calls remove and replaced by raise AssertionError()
(by @DEKHTIARJonathan in #667)tl.identity
is removed, not used anymore and deprecated for a long time (by @DEKHTIARJonathan in #667)TF.__version__ < "1.6"
have been removed (by @DEKHTIARJonathan in #675)tl.layers.RNNLayer
with inspect
(by @DEKHTIARJonathan in #574)tl.files
with truth value of an empty array is ambiguous (by @DEKHTIARJonathan in #575)tl.utils.predict
fixed - np.hstack
problem in which the results for multiple batches are stacked along axis=1
(by @2wins in #566)tl.layers.DeformableConv2d
fixed (by @DEKHTIARJonathan in #573)tl.layers.ConvLSTMLayer
fixed (by @dengyueyun666 in #676)tl.layers.TernaryConv2d
fixed - self.inputs not defined (by @DEKHTIARJonathan in #658)tl.layers.binary._compute_threshold()
(by @DEKHTIARJonathan in #658)tf.logging
replaced by tl.logging
(by @DEKHTIARJonathan in #661)tensorlayer.third_party.roi_pooling.roi_pooling.roi_pooling_ops
is now lazy loaded to prevent systematic error raised (by @DEKHTIARJonathan in #675)tutorial_word2vec_basic.py
saving issue #476 fixed (by @DEKHTIARJonathan in #635)@lgarithm @DEKHTIARJonathan @2wins @One-sixth @zsdonghao @luomai