[mxnet] Update document to use MXNet 1.9.0 (#1737)

Change-Id: I990c1db63a52890010579d4896b3ea9e598abd11
deepjavalibrary · Jun 21, 2022 · fe5ee3f · fe5ee3f
1 parent fb61b7c
commit fe5ee3f
Showing 1 changed file with 12 additions and 12 deletions.
diff --git a/engines/mxnet/mxnet-engine/README.md b/engines/mxnet/mxnet-engine/README.md
@@ -50,15 +50,15 @@ Finally, you can also specify the path to a valid MXNet build using the `MXNET_L
 
 For macOS, you can use the following library:
 
-- ai.djl.mxnet:mxnet-native-mkl:1.8.0:osx-x86_64
+- ai.djl.mxnet:mxnet-native-mkl:1.9.0:osx-x86_64
 
     This package takes advantage of the Intel MKL library to boost performance.
 ```xml
 <dependency>
     <groupId>ai.djl.mxnet</groupId>
     <artifactId>mxnet-native-mkl</artifactId>
     <classifier>osx-x86_64</classifier>
-    <version>1.8.0</version>
+    <version>1.9.0</version>
     <scope>runtime</scope>
 </dependency>
 ```
@@ -76,15 +76,15 @@ Apache MXNet no longer statically link CUDNN and NCCL in it.
 Apache MXNet 1.8.0 cu110 package supports SM 5.0, 6.0, 7.0 and 8.0 cuda architectures.
 Apache MXNet 1.8.0 cu102 package supports SM 3.0, 5.0, 6.0, 7.0 and 7.5 cuda architectures.
 
-- ai.djl.mxnet:mxnet-native-cu102mkl:1.8.0:linux-x86_64 - CUDA 11.0
-- ai.djl.mxnet:mxnet-native-cu101mkl:1.8.0:linux-x86_64 - CUDA 10.2
+- ai.djl.mxnet:mxnet-native-cu102mkl:1.9.0:linux-x86_64 - CUDA 11.2
+- ai.djl.mxnet:mxnet-native-cu101mkl:1.9.0:linux-x86_64 - CUDA 10.2
 
 ```xml
 <dependency>
     <groupId>ai.djl.mxnet</groupId>
-    <artifactId>mxnet-native-cu110mkl</artifactId>
+    <artifactId>mxnet-native-cu112mkl</artifactId>
     <classifier>linux-x86_64</classifier>
-    <version>1.8.0</version>
+    <version>1.9.0</version>
     <scope>runtime</scope>
 </dependency>
 ```
@@ -94,22 +94,22 @@ Apache MXNet 1.8.0 cu102 package supports SM 3.0, 5.0, 6.0, 7.0 and 7.5 cuda arc
     <groupId>ai.djl.mxnet</groupId>
     <artifactId>mxnet-native-cu102mkl</artifactId>
     <classifier>linux-x86_64</classifier>
-    <version>1.8.0</version>
+    <version>1.9.0</version>
     <scope>runtime</scope>
 </dependency>
 ```
 
 #### Linux CPU
 
-- ai.djl.mxnet:mxnet-native-mkl:1.8.0:linux-x86_64
+- ai.djl.mxnet:mxnet-native-mkl:1.9.0:linux-x86_64
 
 ```xml
 <dependency>
     <groupId>ai.djl.mxnet</groupId>
     <artifactId>mxnet-native-mkl</artifactId>
     <classifier>linux-x86_64</classifier>
     <scope>runtime</scope>
-    <version>1.8.0</version>
+    <version>1.9.0</version>
 </dependency>
 ```
 
@@ -126,18 +126,18 @@ You can still access GPU on windows by using the [automatic runtime download](#I
 
 #### Windows GPU
 
-This package supports CUDA 11.0 and CUDA 10.2 for Windows.
+This package supports CUDA 11.2 and CUDA 10.2 for Windows.
 
 ### Windows CPU
 
-- ai.djl.mxnet:mxnet-native-mkl:1.8.0:win-x86_64
+- ai.djl.mxnet:mxnet-native-mkl:1.9.0:win-x86_64
 
 ```xml
 <dependency>
     <groupId>ai.djl.mxnet</groupId>
     <artifactId>mxnet-native-mkl</artifactId>
     <classifier>win-x86_64</classifier>
     <scope>runtime</scope>
-    <version>1.8.0</version>
+    <version>1.9.0</version>
 </dependency>
 ```