Merge pull request tflearn#618 from tflearn/1.0

TFLearn v0.3 release

.travis.yml

@@ -2,8 +2,8 @@ sudo: required
 dist: trusty
 language: python
 python:
-  - 2.7.11
-  - 3.5.1
+  - 2.7
+  - 3.5
 virtualenv:
   system_site_package: true
 addons:
@@ -17,12 +17,12 @@ install:
   - echo $TRAVIS_PYTHON_VERSION
   - pip install -U pytest-cov
   - pip install -U h5py Pillow six numpy
-  - if [[ $TRAVIS_PYTHON_VERSION == 2* ]]; then travis_retry pip install https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.11.0rc0-cp27-none-linux_x86_64.whl; fi
-  - if [[ $TRAVIS_PYTHON_VERSION == 3.5* ]]; then travis_retry pip3 install https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.11.0rc0-cp35-cp35m-linux_x86_64.whl; fi
+  - if [[ $TRAVIS_PYTHON_VERSION == 2* ]]; then travis_retry pip install https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.0.0-cp27-none-linux_x86_64.whl; fi
+  - if [[ $TRAVIS_PYTHON_VERSION == 3.5* ]]; then travis_retry pip3 install https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.0.0-cp35-cp35m-linux_x86_64.whl; fi
   - python setup.py install
 
 script:
   - python setup.py test
 
 matrix:
-  fast_finish: true
+  fast_finish: true

README.md

@@ -18,7 +18,7 @@ TFLearn features include:
 
 The high-level API currently supports most of recent deep learning models, such as Convolutions, LSTM, BiRNN, BatchNorm, PReLU, Residual networks, Generative networks... In the future, TFLearn is also intended to stay up-to-date with latest deep learning techniques.
 
-Note: This is the first release of TFLearn. Contributions are more than welcome!
+Note: Latest TFLearn (v0.3) is only compatible with TensorFlow v1.0 and over.
 
 ## Overview
 ```python
@@ -54,7 +54,7 @@ There are many more examples available *[here](http://tflearn.org/examples)*.
 
 **TensorFlow Installation**
 
-TFLearn requires Tensorflow (version >= 0.11) to be installed: *[Tensorflow installation instructions](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/g3doc/get_started/os_setup.md)*.
+TFLearn requires Tensorflow (version 1.0) to be installed: *[Tensorflow installation instructions](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/g3doc/get_started/os_setup.md)*.
 
 **TFLearn Installation**
 

RELEASE.md

@@ -1,3 +1,34 @@
+# Release 0.3.0
+
+Major changes:
+- TensorFlow 1.0 compatibility
+
+Minor changes:
+- Documents refactoring.
+- Inception-ResNet-v2 Example.
+- CIFAR-100 Dataset.
+- Added time monitoring.
+- Various bug fix.
+
+# Release 0.2.2
+
+Minor changes:
+- Support for 3D conv ops
+- New layers: time_distributed, l2_normalize
+- RNNs support for batch norm
+- Added an option to save the best model
+- Seq2seq and Reinforcement learning examples
+- Beginner tutorial
+- Others minor changes
+- Various bug fix
+
+# Release 0.2.1
+
+Minor changes:
+- Various bug fix
+- State of the art result for Residual Networks
+- FCN objective & Upscore layer
+
 # Release 0.2.0
 
 Major changes:

docs/templates/getting_started.md

@@ -375,7 +375,7 @@ tflearn.is_training(False)
 ### Training Callbacks
 
 During the training cycle, TFLearn gives you the possibility to track and interact with the metrics of the training throughout a set of functions given by the [Callback](https://github.com/tflearn/tflearn/blob/master/tflearn/callbacks.py#L10) interface.
-To simplify the metrics retreivement, each callback method received a [TrainingState](https://github.com/tflearn/tflearn/blob/master/tflearn/helpers/trainer.py#L976) which track the state (e.g. : current epoch, step, batch iteration) and metrics (e.g. : current validation accuracy, global accuracy etc..)
+To simplify the metrics retrieval, each callback method received a [TrainingState](https://github.com/tflearn/tflearn/blob/master/tflearn/helpers/trainer.py#L976) which track the state (e.g. : current epoch, step, batch iteration) and metrics (e.g. : current validation accuracy, global accuracy etc..)
 
 Callback methods which relate to the training cycle : 
 - `on_train_begin(training_state)`

docs/templates/index.md

@@ -13,7 +13,7 @@ TFLearn features include:
 
 The high-level API currently supports most of recent deep learning models, such as Convolutions, LSTM, BiRNN, BatchNorm, PReLU, Residual networks, Generative networks... In the future, TFLearn is also intended to stay up-to-date with latest deep learning techniques.
 
-Note: TF Learn is at an early development stage. Contributions are more than welcome!
+Note: Latest TFLearn (v0.3) is only compatible with TensorFlow v1.0 and over.
 
 # Quick overview
 
@@ -82,7 +82,7 @@ GitHub: [https://github.com/tflearn/tflearn](https://github.com/tflearn/tflearn)
 
 This is the first release of TFLearn, if you find any bug, please report it in the GitHub issues section.
 
-Improvements and requests for new features are more than welcome! Do not hesitate to twist and tweak TFLearn, and send pull-requests.
+Improvements and requests for new features are more than welcome! Do not hesitate to twist and tweak TF Learn, and send pull-requests.
 
 For more info: *[Contribute to TFLearn](contributions)*.
 

docs/templates/installation.md

@@ -2,42 +2,56 @@
 
 ## Tensorflow Installation
 
-TFLearn requires Tensorflow (version >= 0.9.0) to be installed.
+TFLearn requires Tensorflow (version 1.0) to be installed.
 
 First, select the correct binary to install (according to your system):
 ```python
 # Ubuntu/Linux 64-bit, CPU only, Python 2.7
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0-cp27-none-linux_x86_64.whl
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.0.0-cp27-none-linux_x86_64.whl
 
 # Ubuntu/Linux 64-bit, GPU enabled, Python 2.7
-# Requires CUDA toolkit 7.5 and CuDNN v5. For other versions, see "Install from sources" below.
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-0.10.0-cp27-none-linux_x86_64.whl
+# Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below.
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.0.0-cp27-none-linux_x86_64.whl
 
 # Mac OS X, CPU only, Python 2.7:
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.10.0-py2-none-any.whl
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-1.0.0-py2-none-any.whl
 
 # Mac OS X, GPU enabled, Python 2.7:
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow-0.10.0-py2-none-any.whl
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow_gpu-1.0.0-py2-none-any.whl
+
+# Ubuntu/Linux 64-bit, CPU only, Python 3.3
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.0.0-cp33-cp33m-linux_x86_64.whl
+
+# Ubuntu/Linux 64-bit, GPU enabled, Python 3.3
+# Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below.
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.0.0-cp33-cp33m-linux_x86_64.whl
 
 # Ubuntu/Linux 64-bit, CPU only, Python 3.4
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0-cp34-cp34m-linux_x86_64.whl
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.0.0-cp34-cp34m-linux_x86_64.whl
 
 # Ubuntu/Linux 64-bit, GPU enabled, Python 3.4
-# Requires CUDA toolkit 7.5 and CuDNN v5. For other versions, see "Install from sources" below.
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-0.10.0-cp34-cp34m-linux_x86_64.whl
+# Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below.
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.0.0-cp34-cp34m-linux_x86_64.whl
 
 # Ubuntu/Linux 64-bit, CPU only, Python 3.5
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0-cp35-cp35m-linux_x86_64.whl
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.0.0-cp35-cp35m-linux_x86_64.whl
 
 # Ubuntu/Linux 64-bit, GPU enabled, Python 3.5
-# Requires CUDA toolkit 7.5 and CuDNN v5. For other versions, see "Install from sources" below.
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-0.10.0-cp35-cp35m-linux_x86_64.whl
+# Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below.
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.0.0-cp35-cp35m-linux_x86_64.whl
+
+# Ubuntu/Linux 64-bit, CPU only, Python 3.6
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.0.0-cp36-cp36m-linux_x86_64.whl
+
+# Ubuntu/Linux 64-bit, GPU enabled, Python 3.6
+# Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below.
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.0.0-cp36-cp36m-linux_x86_64.whl
 
 # Mac OS X, CPU only, Python 3.4 or 3.5:
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.10.0-py3-none-any.whl
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-1.0.0-py3-none-any.whl
 
 # Mac OS X, GPU enabled, Python 3.4 or 3.5:
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow-0.10.0-py3-none-any.whl
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow_gpu-1.0.0-py3-none-any.whl
 ```
 
 Then install TensorFlow:
@@ -71,7 +85,7 @@ python setup.py install
 
 ## Upgrade Tensorflow
 
-If your version of Tensorflow is too old (under 0.9.0), you may need to upgrade Tensorflow to avoid some incompatibilities with TFLearn.
+If your version of Tensorflow is too old (under 1.0), you may need to upgrade Tensorflow to avoid some incompatibilities with TFLearn.
 
 To upgrade Tensorflow, you first need to uninstall Tensorflow and Protobuf:
 
@@ -83,37 +97,51 @@ pip uninstall tensorflow
 Then you can re-install Tensorflow. Select the correct binary to install (according to your system):
 ```python
 # Ubuntu/Linux 64-bit, CPU only, Python 2.7
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0-cp27-none-linux_x86_64.whl
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.0.0-cp27-none-linux_x86_64.whl
 
 # Ubuntu/Linux 64-bit, GPU enabled, Python 2.7
-# Requires CUDA toolkit 7.5 and CuDNN v5. For other versions, see "Install from sources" below.
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-0.10.0-cp27-none-linux_x86_64.whl
+# Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below.
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.0.0-cp27-none-linux_x86_64.whl
 
 # Mac OS X, CPU only, Python 2.7:
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.10.0-py2-none-any.whl
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-1.0.0-py2-none-any.whl
 
 # Mac OS X, GPU enabled, Python 2.7:
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow-0.10.0-py2-none-any.whl
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow_gpu-1.0.0-py2-none-any.whl
+
+# Ubuntu/Linux 64-bit, CPU only, Python 3.3
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.0.0-cp33-cp33m-linux_x86_64.whl
+
+# Ubuntu/Linux 64-bit, GPU enabled, Python 3.3
+# Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below.
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.0.0-cp33-cp33m-linux_x86_64.whl
 
 # Ubuntu/Linux 64-bit, CPU only, Python 3.4
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0-cp34-cp34m-linux_x86_64.whl
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.0.0-cp34-cp34m-linux_x86_64.whl
 
 # Ubuntu/Linux 64-bit, GPU enabled, Python 3.4
-# Requires CUDA toolkit 7.5 and CuDNN v5. For other versions, see "Install from sources" below.
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-0.10.0-cp34-cp34m-linux_x86_64.whl
+# Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below.
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.0.0-cp34-cp34m-linux_x86_64.whl
 
 # Ubuntu/Linux 64-bit, CPU only, Python 3.5
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0-cp35-cp35m-linux_x86_64.whl
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.0.0-cp35-cp35m-linux_x86_64.whl
 
 # Ubuntu/Linux 64-bit, GPU enabled, Python 3.5
-# Requires CUDA toolkit 7.5 and CuDNN v5. For other versions, see "Install from sources" below.
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-0.10.0-cp35-cp35m-linux_x86_64.whl
+# Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below.
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.0.0-cp35-cp35m-linux_x86_64.whl
+
+# Ubuntu/Linux 64-bit, CPU only, Python 3.6
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.0.0-cp36-cp36m-linux_x86_64.whl
+
+# Ubuntu/Linux 64-bit, GPU enabled, Python 3.6
+# Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below.
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.0.0-cp36-cp36m-linux_x86_64.whl
 
 # Mac OS X, CPU only, Python 3.4 or 3.5:
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.10.0-py3-none-any.whl
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-1.0.0-py3-none-any.whl
 
 # Mac OS X, GPU enabled, Python 3.4 or 3.5:
-$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow-0.10.0-py3-none-any.whl
+$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow_gpu-1.0.0-py3-none-any.whl
 ```
 
 Then install TensorFlow:
@@ -125,6 +153,6 @@ $ sudo pip install --upgrade $TF_BINARY_URL
 $ sudo pip3 install --upgrade $TF_BINARY_URL
 ```
 
-## Using Latest Tensorflow
+## Using Latest TensorFlow
 
 TFLearn is compatible with [master version](https://github.com/tensorflow/tensorflow) of Tensorflow, but some warnings may appear.

examples/nlp/seq2seq_example.py

@@ -119,8 +119,8 @@ def sequence_loss(self, y_pred, y_true):
         then use seq2seq.sequence_loss to actually compute the loss function.
         '''
         if self.verbose > 2: print ("my_sequence_loss y_pred=%s, y_true=%s" % (y_pred, y_true))
-        logits = tf.unpack(y_pred, axis=1)		# list of [-1, num_decoder_synbols] elements
-        targets = tf.unpack(y_true, axis=1)		# y_true has shape [-1, self.out_seq_len]; unpack to list of self.out_seq_len [-1] elements
+        logits = tf.unstack(y_pred, axis=1)		# list of [-1, num_decoder_synbols] elements
+        targets = tf.unstack(y_true, axis=1)		# y_true has shape [-1, self.out_seq_len]; unpack to list of self.out_seq_len [-1] elements
         if self.verbose > 2:
             print ("my_sequence_loss logits=%s" % (logits,))
             print ("my_sequence_loss targets=%s" % (targets,))
@@ -159,12 +159,12 @@ def model(self, mode="train", num_layers=1, cell_size=32, cell_type="BasicLSTMCe
 
         network = tflearn.input_data(shape=[None, self.in_seq_len + self.out_seq_len], dtype=tf.int32, name="XY")
         encoder_inputs = tf.slice(network, [0, 0], [-1, self.in_seq_len], name="enc_in")	# get encoder inputs
-        encoder_inputs = tf.unpack(encoder_inputs, axis=1)					# transform into list of self.in_seq_len elements, each [-1]
+        encoder_inputs = tf.unstack(encoder_inputs, axis=1)					# transform into list of self.in_seq_len elements, each [-1]
 
         decoder_inputs = tf.slice(network, [0, self.in_seq_len], [-1, self.out_seq_len], name="dec_in")	# get decoder inputs
-        decoder_inputs = tf.unpack(decoder_inputs, axis=1)					# transform into list of self.out_seq_len elements, each [-1]
+        decoder_inputs = tf.unstack(decoder_inputs, axis=1)					# transform into list of self.out_seq_len elements, each [-1]
 
-        go_input = tf.mul( tf.ones_like(decoder_inputs[0], dtype=tf.int32), GO_VALUE ) # insert "GO" symbol as the first decoder input; drop the last decoder input
+        go_input = tf.multiply( tf.ones_like(decoder_inputs[0], dtype=tf.int32), GO_VALUE ) # insert "GO" symbol as the first decoder input; drop the last decoder input
         decoder_inputs = [go_input] + decoder_inputs[: self.out_seq_len-1]				# insert GO as first; drop last decoder input
 
         feed_previous = not (mode=="train")
@@ -209,7 +209,7 @@ def model(self, mode="train", num_layers=1, cell_size=32, cell_type="BasicLSTMCe
 
         # model_outputs: list of the same length as decoder_inputs of 2D Tensors with shape [batch_size x output_size] containing the generated outputs.
         if self.verbose > 2: print ("model outputs: %s" % model_outputs)
-        network = tf.pack(model_outputs, axis=1)		# shape [-1, n_decoder_inputs (= self.out_seq_len), num_decoder_symbols]
+        network = tf.stack(model_outputs, axis=1)		# shape [-1, n_decoder_inputs (= self.out_seq_len), num_decoder_symbols]
         if self.verbose > 2: print ("packed model outputs: %s" % network)
 
         if self.verbose > 3:

examples/others/recommender_wide_and_deep.py

@@ -149,9 +149,9 @@ def build_model(self, learning_rate=[0.001, 0.01]):
             psize = tf.cast(tf.shape(pos)[0], tf.int64)
             nsize = tf.cast(tf.shape(neg)[0], tf.int64)
             true_positive = tf.reduce_sum(pos, name="true_positive")
-            false_negative = tf.sub(psize, true_positive, name="false_negative")
+            false_negative = tf.subtract(psize, true_positive, name="false_negative")
             false_positive = tf.reduce_sum(neg, name="false_positive")
-            true_negative = tf.sub(nsize, false_positive, name="true_negative")
+            true_negative = tf.subtract(nsize, false_positive, name="true_negative")
             overall_accuracy = tf.truediv(tf.add(true_positive, true_negative), tf.add(nsize, psize), name="overall_accuracy")
         vmset = [true_positive, true_negative, false_positive, false_negative, overall_accuracy]
 

examples/reinforcement_learning/atari_1step_qlearning.py

@@ -328,7 +328,7 @@ def build_graph(num_actions):
     # Define cost and gradient update op
     a = tf.placeholder("float", [None, num_actions])
     y = tf.placeholder("float", [None])
-    action_q_values = tf.reduce_sum(tf.mul(q_values, a), reduction_indices=1)
+    action_q_values = tf.reduce_sum(tf.multiply(q_values, a), reduction_indices=1)
     cost = tflearn.mean_square(action_q_values, y)
     optimizer = tf.train.RMSPropOptimizer(learning_rate)
     grad_update = optimizer.minimize(cost, var_list=network_params)

setup.py

@@ -27,13 +27,13 @@ def run_tests(self):
 
 
 setup(name='tflearn',
-      version='0.2.2',
+      version='0.3',
       description='Deep Learning Library featuring a higher-level API for '
                   'Tensorflow',
       author='TFLearn contributors',
       author_email='[email protected]',
       url='https://github.com/tflearn/tflearn',
-      download_url='https://github.com/tflearn/tflearn/tarball/0.2.2',
+      download_url='https://github.com/tflearn/tflearn/tarball/0.3',
       license='MIT',
       packages=find_packages(),
       install_requires=[