Skip to content

Commit 1d415a6

Browse files
agibsoncccagibsonccc
committed
Update version
1 parent 9078f48 commit 1d415a6

File tree

18 files changed

+99
-20
lines changed

18 files changed

+99
-20
lines changed

‎README.md

+1-1
Original file line numberDiff line numberDiff line change
@@ -19,7 +19,7 @@
1919
The **Eclipse Deeplearning4J** (DL4J) ecosystem is a set of projects intended to support all the needs of a JVM based deep learning application. This means starting with the raw data, loading and preprocessing it from wherever and whatever format it is in to building and tuning a wide variety of simple and complex deep learning networks.
2020

2121
The DL4J stack comprises of:
22-
- **DL4J**: High level API to build MultiLayerNetworks and ComputationGraphs with a variety of layers, including custom ones. Supports importing Keras models from h5, including tf.keras models (as of 1.0.0-M1) and also supports distributed training on Apache Spark
22+
- **DL4J**: High level API to build MultiLayerNetworks and ComputationGraphs with a variety of layers, including custom ones. Supports importing Keras models from h5, including tf.keras models (as of 1.0.0-M1.1) and also supports distributed training on Apache Spark
2323
- **ND4J**: General purpose linear algebra library with over 500 mathematical, linear algebra and deep learning operations. ND4J is based on the highly-optimized C++ codebase LibND4J that provides CPU (AVX2/512) and GPU (CUDA) support and acceleration by libraries such as OpenBLAS, OneDNN (MKL-DNN), cuDNN, cuBLAS, etc
2424
- **SameDiff** : Part of the ND4J library, SameDiff is our automatic differentiation / deep learning framework. SameDiff uses a graph-based (define then run) approach, similar to TensorFlow graph mode. Eager graph (TensorFlow 2.x eager/PyTorch) graph execution is planned. SameDiff supports importing TensorFlow frozen model format .pb (protobuf) models. Import for ONNX, TensorFlow SavedModel and Keras models are planned. Deeplearning4j also has full SameDiff support for easily writing custom layers and loss functions.
2525
- **DataVec**: ETL for machine learning data in a wide variety of formats and files (HDFS, Spark, Images, Video, Audio, CSV, Excel etc)

‎android-examples/app/build.gradle

+1-1
Original file line numberDiff line numberDiff line change
@@ -27,7 +27,7 @@ android {
2727
}
2828

2929
dependencies {
30-
def dl4jVersion = '1.0.0-M1'
30+
def dl4jVersion = '1.0.0-M1.1'
3131
def openblasVersion = '0.3.10-1.5.5'
3232
def opencvVersion = '4.5.1-1.5.5'
3333
def leptonicaVersion = '1.80.0-1.5.5'

‎data-pipeline-examples/pom.xml

+2-2
Original file line numberDiff line numberDiff line change
@@ -23,12 +23,12 @@
2323

2424
<groupId>org.deeplearning4j</groupId>
2525
<artifactId>data-pipeline-examples</artifactId>
26-
<version>1.0.0-M1</version>
26+
<version>1.0.0-M1.1</version>
2727
<name>Building a data pipeline prior to modeling</name>
2828
<description>Loading raw data and processing it before training</description>
2929

3030
<properties>
31-
<dl4j-master.version>1.0.0-M1</dl4j-master.version>
31+
<dl4j-master.version>1.0.0-M1.1</dl4j-master.version>
3232
<!-- Change the nd4j.backend property to nd4j-cuda-X-platform to use CUDA GPUs -->
3333
<!-- <nd4j.backend>nd4j-cuda-10.2-platform</nd4j.backend> -->
3434
<nd4j.backend>nd4j-native</nd4j.backend>

‎dl4j-distributed-training-examples/pom.xml

+2-2
Original file line numberDiff line numberDiff line change
@@ -23,12 +23,12 @@
2323

2424
<groupId>org.deeplearning4j</groupId>
2525
<artifactId>dl4j-distributed-training-examples</artifactId>
26-
<version>1.0.0-M1</version>
26+
<version>1.0.0-M1.1</version>
2727
<name>Introduction to Distributed Training with DL4J</name>
2828
<description>A set of examples introducing distributed training with the DL4J framework</description>
2929

3030
<properties>
31-
<dl4j-master.version>1.0.0-M1</dl4j-master.version>
31+
<dl4j-master.version>1.0.0-M1.1</dl4j-master.version>
3232
<!-- Change the nd4j.backend property to nd4j-cuda-X-platform to use CUDA GPUs -->
3333
<!-- <nd4j.backend>nd4j-cuda-10.2-platform</nd4j.backend> -->
3434
<nd4j.backend>nd4j-native</nd4j.backend>

‎dl4j-distributed-training-examples/scripts/patentExamplePreproc.sh

+1-1
Original file line numberDiff line numberDiff line change
@@ -32,7 +32,7 @@ MINIBATCH=32 #Minibatch size for preprocessed
3232

3333
#Other variables. Don't modify these
3434
SCRIPTDIR=$(dirname "$0")
35-
JARFILE=${SCRIPTDIR}/../target/dl4j-spark-1.0.0-M1-bin.jar
35+
JARFILE=${SCRIPTDIR}/../target/dl4j-spark-1.0.0-M1.1-bin.jar
3636
AZURE_ACCT=fs.azure.account.key.${AZURE_STORAGE_ACCT}.blob.core.windows.net
3737

3838

‎dl4j-distributed-training-examples/scripts/patentExampleTrain.sh

+1-1
Original file line numberDiff line numberDiff line change
@@ -40,7 +40,7 @@ AERON_BUFFER=33554432
4040

4141
#Other variables. Don't modify these
4242
SCRIPTDIR=$(dirname "$0")
43-
JARFILE=${SCRIPTDIR}/../target/dl4j-spark-1.0.0-M1-bin.jar
43+
JARFILE=${SCRIPTDIR}/../target/dl4j-spark-1.0.0-M1.1-bin.jar
4444
AZURE_ACCT=fs.azure.account.key.${AZURE_STORAGE_ACCT}.blob.core.windows.net
4545

4646

‎dl4j-distributed-training-examples/scripts/tinyImagenetTrain.sh

+1-1
Original file line numberDiff line numberDiff line change
@@ -32,7 +32,7 @@ EXECUTOR_MEMORY=12G
3232
MASTER_PORT=7077 # Port for the spark master. Default is 7077
3333
MINIBATCH=32 # Minibatch size for preprocessed datasets
3434
SCRIPTDIR=$(dirname "$0")
35-
JARFILE=${SCRIPTDIR}/../target/dl4j-spark-1.0.0-M1-bin.jar
35+
JARFILE=${SCRIPTDIR}/../target/dl4j-spark-1.0.0-M1.1-bin.jar
3636

3737
#Memory
3838
OFFHEAP_MEM_JAVACPP=20G # Maximum amount of off-heap memory

‎dl4j-examples/pom.xml

+1-1
Original file line numberDiff line numberDiff line change
@@ -28,7 +28,7 @@
2828
<description>A set of examples introducing the DL4J framework</description>
2929

3030
<properties>
31-
<dl4j-master.version>1.0.0-M1</dl4j-master.version>
31+
<dl4j-master.version>1.0.0-M1.1</dl4j-master.version>
3232
<!-- Change the nd4j.backend property to nd4j-cuda-X-platform to use CUDA GPUs -->
3333
<!-- <nd4j.backend>nd4j-cuda-10.2-platform</nd4j.backend> -->
3434
<nd4j.backend>nd4j-cuda-11.2</nd4j.backend>

‎mvn-project-template/pom.xml

+2-2
Original file line numberDiff line numberDiff line change
@@ -24,11 +24,11 @@
2424
<!-- Group-ID, artifact ID and version of the project. You can modify these as you want -->
2525
<groupId>org.deeplearning4j</groupId>
2626
<artifactId>deeplearning4j-example-sample</artifactId>
27-
<version>1.0.0-M1</version>
27+
<version>1.0.0-M1.1</version>
2828

2929
<!-- Properties Section. Change ND4J versions here, if required -->
3030
<properties>
31-
<dl4j-master.version>1.0.0-M1</dl4j-master.version>
31+
<dl4j-master.version>1.0.0-M1.1</dl4j-master.version>
3232
<logback.version>1.2.3</logback.version>
3333
<java.version>1.8</java.version>
3434
<maven-shade-plugin.version>2.4.3</maven-shade-plugin.version>

‎nd4j-ndarray-examples/pom.xml

+5-1
Original file line numberDiff line numberDiff line change
@@ -28,6 +28,10 @@
2828
<description>Working with NDArrays</description>
2929

3030
<repositories>
31+
<repository>
32+
<id>orgdeeplearning4j-1177</id>
33+
<url>https://oss.sonatype.org/content/repositories/orgdeeplearning4j-1177</url>
34+
</repository>
3135
<repository>
3236
<id>sonatype-nexus-snapshots</id>
3337
<name>Sonatype Nexus Snapshots</name>
@@ -44,7 +48,7 @@
4448

4549

4650
<properties>
47-
<dl4j-master.version>1.0.0-M1</dl4j-master.version>
51+
<dl4j-master.version>1.0.0-M1.1.1</dl4j-master.version>
4852
<!-- Change the nd4j.backend property to nd4j-cuda-X-platform to use CUDA GPUs -->
4953
<!-- <nd4j.backend>nd4j-cuda-10.2-platform</nd4j.backend> -->
5054
<nd4j.backend>nd4j-native</nd4j.backend>

‎rl4j-examples/pom.xml

+7-3
Original file line numberDiff line numberDiff line change
@@ -6,10 +6,10 @@
66

77
<groupId>org.deeplearning4j</groupId>
88
<artifactId>rl4j-examples</artifactId>
9-
<version>1.0.0-M1</version>
9+
<version>1.0.0-M1.1</version>
1010
<properties>
11-
<nd4j.version>1.0.0-M1</nd4j.version>
12-
<rl4j.version>1.0.0-M1</rl4j.version>
11+
<nd4j.version>1.0.0-M1.1.1</nd4j.version>
12+
<rl4j.version>1.0.0-M1.1.1</rl4j.version>
1313
<logback.version>1.1.7</logback.version>
1414
<java.version>1.8</java.version>
1515
<maven-compiler-plugin.version>3.8.1</maven-compiler-plugin.version>
@@ -20,6 +20,10 @@
2020

2121

2222
<repositories>
23+
<repository>
24+
<id>orgdeeplearning4j-1177</id>
25+
<url>https://oss.sonatype.org/content/repositories/orgdeeplearning4j-1177</url>
26+
</repository>
2327
<repository>
2428
<id>sonatype-nexus-snapshots</id>
2529
<name>Sonatype Nexus Snapshots</name>

‎samediff-examples/pom.xml

+7-2
Original file line numberDiff line numberDiff line change
@@ -23,12 +23,12 @@
2323

2424
<groupId>org.deeplearning4j</groupId>
2525
<artifactId>samediff-examples</artifactId>
26-
<version>1.0.0-M1</version>
26+
<version>1.0.0-M1.1</version>
2727
<name>Quickstart to SameDiff</name>
2828
<description>A set of beginner examples introducing the SameDiff framework</description>
2929

3030
<properties>
31-
<dl4j-master.version>1.0.0-M1</dl4j-master.version>
31+
<dl4j-master.version>1.0.0-M1.1.1</dl4j-master.version>
3232
<!-- Change the nd4j.backend property to nd4j-cuda-X-platform to use CUDA GPUs -->
3333
<!-- <nd4j.backend>nd4j-cuda-10.2-platform</nd4j.backend> -->
3434
<nd4j.backend>nd4j-native</nd4j.backend>
@@ -46,6 +46,11 @@
4646
</properties>
4747

4848
<repositories>
49+
<repository>
50+
<id>orgdeeplearning4j-1177</id>
51+
<url>https://oss.sonatype.org/content/repositories/orgdeeplearning4j-1177</url>
52+
</repository>
53+
4954
<repository>
5055
<id>sonatype-nexus-snapshots</id>
5156
<name>Sonatype Nexus Snapshots</name>

‎tensorflow-keras-import-examples/README.md

+1-1
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,6 @@
11
## Eclipse Deeplearning4j: Model Import
22

3-
The DL4J supports models created in the popular Python Tensorflow and Keras frameworks. As of 1.0.0-M1, Keras models (including tf.keras) can be imported into Deeplearning. TensorFlow frozen format models can be imported into SameDiff.
3+
The DL4J supports models created in the popular Python Tensorflow and Keras frameworks. As of 1.0.0-M1.1, Keras models (including tf.keras) can be imported into Deeplearning. TensorFlow frozen format models can be imported into SameDiff.
44

55
Models in Tensorflow have to be converted to "frozen" pbs (protobuf). More information on freezing Tensorflow models can be found [here](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/tools/freeze_graph.py#L15) for Tensorflow 1.X and [here](https://leimao.github.io/blog/Save-Load-Inference-From-TF2-Frozen-Graph/) for Tensorflow 2.X. Keras models have to be saved in h5 format. More information can be found [here](https://keras.io/getting-started/faq/#how-can-i-save-a-keras-model). Importing both Keras 1 and Keras 2 models are supported. Of note - importing models saved with tf.keras is also supported. Currently general TensorFlow operations within Keras models (i.e., those not part of the tf.keras API) are currently importable but support inference only. Full training is supported for anything that is part of the Keras API.
66

‎tensorflow-keras-import-examples/ops-added-old.txt

+20
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,20 @@
1+
Placeholder,input
2+
Const,Variable
3+
Const,Variable_1
4+
Const,Variable_2
5+
Const,Variable_3
6+
Const,Variable_4
7+
Const,Variable_5
8+
Identity,Variable/read
9+
Identity,Variable_1/read
10+
Identity,Variable_2/read
11+
Identity,Variable_3/read
12+
Identity,Variable_4/read
13+
Identity,Variable_5/read
14+
MatMul,MatMul
15+
Add,Add
16+
MatMul,MatMul_1
17+
Add,Add_1
18+
MatMul,MatMul_2
19+
Add,add
20+
Softmax,output

‎tensorflow-keras-import-examples/ops-imported-old.txt

+13
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,13 @@
1+
Identity,Variable/read
2+
Identity,Variable_1/read
3+
Identity,Variable_2/read
4+
Identity,Variable_3/read
5+
Identity,Variable_4/read
6+
Identity,Variable_5/read
7+
MatMul,MatMul
8+
Add,Add
9+
MatMul,MatMul_1
10+
Add,Add_1
11+
MatMul,MatMul_2
12+
Add,add
13+
Softmax,output

‎tensorflow-keras-import-examples/ops-removed-old.txt

+20
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,20 @@
1+
input
2+
Variable
3+
Variable_1
4+
Variable_2
5+
Variable_3
6+
Variable_4
7+
Variable_5
8+
Variable/read
9+
Variable_1/read
10+
Variable_2/read
11+
Variable_3/read
12+
Variable_4/read
13+
Variable_5/read
14+
MatMul
15+
Add
16+
MatMul_1
17+
Add_1
18+
MatMul_2
19+
add
20+
output

‎tensorflow-keras-import-examples/pom.xml

+1-1
Original file line numberDiff line numberDiff line change
@@ -28,7 +28,7 @@
2828
<description>Loading models trained in keras or tensorflow</description>
2929

3030
<properties>
31-
<dl4j-master.version>1.0.0-M1</dl4j-master.version>
31+
<dl4j-master.version>1.0.0-M1.1</dl4j-master.version>
3232
<!-- Change the nd4j.backend property to nd4j-cuda-X-platform to use CUDA GPUs -->
3333
<nd4j.backend>nd4j-native</nd4j.backend>
3434
<!-- <nd4j.backend>nd4j-cuda-10.2-platform</nd4j.backend> -->

‎tensorflow-keras-import-examples/variables-added-old.txt

+13
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,13 @@
1+
Variable/read,Variable/read
2+
Variable_1/read,Variable_1/read
3+
Variable_2/read,Variable_2/read
4+
Variable_3/read,Variable_3/read
5+
Variable_4/read,Variable_4/read
6+
Variable_5/read,Variable_5/read
7+
MatMul,MatMul
8+
Add,Add
9+
MatMul_1,MatMul_1
10+
Add_1,Add_1
11+
MatMul_2,MatMul_2
12+
add,add
13+
output,output

0 commit comments

Comments
 (0)