Merge branch 'develop'

CHANGELOG.md

@@ -6,6 +6,18 @@ Most recent releases are shown at the top. Each release shows:
 - **Changed**: Additional parameters, changes to inputs or outputs, etc
 - **Fixed**: Bug fixes that don't change documented behaviour
 
+## 0.7.0 (2019-12-10)
+
+### New:
+- *ktrain* now uses tf.keras (`tensorflow>=1.14,<=2.0) instead of stand-alone Keras.  
+
+### Changed:
+- N/A
+
+### Fixed:
+- N/A
+
+
 ## 0.6.2 (2019-12-02)
 
 ### New:

README.md

@@ -1,11 +1,17 @@
+
+
+
 ### News and Announcements
+- **2019-12-10:**  
+  - ***ktrain*** **v0.7.x is released and now uses TensorFlow Keras** (i.e., `tf.keras`) instead of stand-alone Keras.  If you're using custom Keras models with *ktrain*, you must change all `keras` references to `tensorflow.keras`. That is, don't import Keras like this: `from keras.layers import Dense`. Do this instead:  `from tensorflow.keras.layers import Dense`.  If you mix calls to tf.keras with Keras, you will experience problems.  Supported versions of TensorFlow include 1.14 and 2.0.
 - **2019-11-12:**  
   - *ktrain* v0.6.x is released and includes pre-canned support for [learning from unlabeled or partially labeled text data](https://nbviewer.jupyter.org/github/amaiya/ktrain/blob/master/tutorial-05-learning_from_unlabeled_text_data.ipynb).
-- **2019-10-16:**  
-  - *ktrain* v0.5.x is released and includes pre-canned support for [node classification in graphs](https://nbviewer.jupyter.org/github/amaiya/ktrain/blob/master/examples/graphs/hateful_twitter_users-GraphSAGE.ipynb).
 - **Coming Soon**:
   - better support for custom data formats and models
-  - support for using *ktrain* with `tf.keras` and TensorFlow 2.0
+  - ability to train [HuggingFace Transformer models](https://github.com/huggingface/transformers) within *ktrain*
+
+
+
 ----
 
 
@@ -177,9 +183,20 @@ Additional examples can be found [here](https://github.com/amaiya/ktrain/tree/ma
 
 ### Installation
 
-```
-pip3 install ktrain
-```
+Make sure pip is up-to-date with: `pip3 install -U pip`.
+
+1. Ensure Tensorflow 1.14 or TensorFlow 2 [is installed](https://www.tensorflow.org/install/pip?lang=python3) if it is not already
+
+> For GPU: `pip3 install "tensorflow_gpu>=1.14,<=2"`
+
+> For CPU: `pip3 install "tensorflow>=1.14,<=2"`
+
+
+2. Install *ktrain*: `pip3 install ktrain`
+
+The *ktrain* package can be used with TensorFlow versions 1.14 and 2.0.  If using TensorFlow 2.0, *ktrain*
+presently runs in 1.x mode using [tf.compat.v1.disable_v2_behavior](https://www.tensorflow.org/api_docs/python/tf/compat/v1/disable_v2_behavior).  In the future, this will be removed and **only** TensorFlow 2 will be supported.
+
 
 <!--
 ### Requirements
@@ -197,8 +214,7 @@ The following software/libraries should be installed:
 -->
 
 
-
-This code was tested on Ubuntu 18.04 LTS using Keras 2.2.4 with a TensorFlow 1.14 backend.
+This code was tested on Ubuntu 18.04 LTS using TensorFlow 1.14 and TensorFlow 2 (Keras version 2.2.4-tf).
 
 ----
 **Creator:  [Arun S. Maiya](http://arun.maiya.net)**

examples/text/IMDb-fasttext.ipynb

@@ -70,8 +70,8 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from keras.models import Sequential\n",
-    "from keras.layers import Dense, Embedding, GlobalAveragePooling1D\n",
+    "from tensorflow.keras.models import Sequential\n",
+    "from tensorflow.keras.layers import Dense, Embedding, GlobalAveragePooling1D\n",
     "def get_model():\n",
     "    model = Sequential()\n",
     "    model.add(Embedding(20000+1, 50, input_length=400)) # add 1 for padding token\n",

examples/text/toxic_comments-bigru.ipynb

@@ -182,7 +182,7 @@
    "outputs": [],
    "source": [
     "# define a custom callback for ROC-AUC\n",
-    "from keras.callbacks import Callback\n",
+    "from tensorflow.keras.callbacks import Callback\n",
     "from sklearn.metrics import roc_auc_score\n",
     "class RocAucEvaluation(Callback):\n",
     "    def __init__(self, validation_data=(), interval=1):\n",

examples/vision/cifar10-WRN22.ipynb

@@ -12,8 +12,7 @@
     "import os\n",
     "os.environ[\"CUDA_DEVICE_ORDER\"]=\"PCI_BUS_ID\";\n",
     "os.environ[\"CUDA_VISIBLE_DEVICES\"]=\"0\"; \n",
-    "import sys\n",
-    "sys.path.append('..')"
+    "import sys"
    ]
   },
   {

examples/vision/mnist-WRN22.ipynb

@@ -12,8 +12,7 @@
     "import os\n",
     "os.environ[\"CUDA_DEVICE_ORDER\"]=\"PCI_BUS_ID\";\n",
     "os.environ[\"CUDA_VISIBLE_DEVICES\"]=\"0\" \n",
-    "import sys\n",
-    "sys.path.append('..')"
+    "import sys"
    ]
   },
   {

ktrain/core.py

@@ -1257,7 +1257,7 @@ def _load_model(fpath, preproc=None, train_data=None):
     elif (preproc and (isinstance(preproc, NodePreprocessor) or \
                     type(preproc).__name__ == 'NodePreprocessor')) or \
         train_data and U.is_nodeclass(data=train_data):
-        from .graph.stellargraph.layer import MeanAggregator
+        from stellargraph.layer import MeanAggregator
         custom_objects={'MeanAggregator': MeanAggregator}
     try:
         model = load_model(fpath, custom_objects=custom_objects)

ktrain/graph/data.py

@@ -1,11 +1,3 @@
-#import networkx as nx
-#import pandas as pd
-#import stellargraph as sg
-#from stellargraph.mapper import GraphSAGENodeGenerator, GraphSAGELinkGenerator
-#from sklearn import preprocessing, feature_extraction, model_selection
-#from stellargraph.data import EdgeSplitter
-
-
 from ..imports import *
 from .. import utils as U
 from .node_generator import NodeSequenceWrapper

ktrain/graph/models.py

@@ -1,10 +1,9 @@
 from ..imports import *
 from .. import utils as U
 from .node_generator import NodeSequenceWrapper
-from . import stellargraph as sg
-from .stellargraph.mapper import GraphSAGENodeGenerator, GraphSAGELinkGenerator
-from .stellargraph.layer import GraphSAGE
-
+import stellargraph as sg
+from stellargraph.mapper import GraphSAGENodeGenerator, GraphSAGELinkGenerator
+from stellargraph.layer import GraphSAGE
 
 
 
@@ -61,7 +60,8 @@ def graph_node_classifier(name, train_data, layer_sizes=[32,32], verbose=1):
         bias=True,
         dropout=0.5,
 	)
-    x_inp, x_out = graphsage_model.default_model(flatten_output=True)
+    #x_inp, x_out = graphsage_model.default_model(flatten_output=True)
+    x_inp, x_out = graphsage_model.build()
     prediction = Dense(units=num_classes, activation=activation)(x_out)
     model = Model(inputs=x_inp, outputs=prediction)
     model.compile(optimizer='adam',

ktrain/graph/node_generator.py

@@ -1,9 +1,9 @@
-from .stellargraph.mapper import node_mappers
+from stellargraph.mapper import node_mappers
 
 class NodeSequenceWrapper(node_mappers.NodeSequence):
     def __init__(self, node_seq):
         if not isinstance(node_seq, node_mappers.NodeSequence):
-            raise ValueError('node_seq must by a stellargraph NodeSequene object')
+            raise ValueError('node_seq must by a stellargraph NodeSequence object')
         self.node_seq = node_seq
         self.targets = node_seq.targets
         self.generator = node_seq.generator

ktrain/graph/preprocessor.py

@@ -3,9 +3,9 @@
 from ..preprocessor import Preprocessor
 from .node_generator import NodeSequenceWrapper
 
-from . import stellargraph as sg
-from .stellargraph.mapper import GraphSAGENodeGenerator, GraphSAGELinkGenerator
-from .stellargraph.layer import GraphSAGE
+import stellargraph as sg
+from stellargraph.mapper import GraphSAGENodeGenerator, GraphSAGELinkGenerator
+from stellargraph.layer import GraphSAGE