Available software - Revision history

Rdickson: /* IBM-AI Deep Learning Anaconda Channel */ change URL on advice from CWhidden

2022-03-08T15:35:20Z

‎IBM-AI Deep Learning Anaconda Channel: change URL on advice from CWhidden

Bgeetika at 14:45, 6 January 2021

2021-01-06T14:45:29Z

Bgeetika at 14:32, 6 January 2021

2021-01-06T14:32:14Z

Lyang: /* IBM WMLA Deep Learning Packages */

2020-12-22T13:34:08Z

‎IBM WMLA Deep Learning Packages

Lyang: /* IBM-AI Deep Learning Anaconda Channel */

2020-12-22T13:27:51Z

‎IBM-AI Deep Learning Anaconda Channel

Bgeetika at 19:10, 4 December 2020

2020-12-04T19:10:29Z

Bgeetika at 19:05, 4 December 2020

2020-12-04T19:05:33Z

Lyang: /* IBM PowerAI Deep Learning Packages */

2020-10-19T16:15:26Z

‎IBM PowerAI Deep Learning Packages

Lyang: /* Anaconda Python */

2020-10-19T16:14:49Z

‎Anaconda Python

Lyang at 16:02, 19 October 2020

2020-10-19T16:02:43Z

@@ Line 30: / Line 30: @@
 To use deep learning packages like Tensorflow, pytorch, caffe on DeepSense you need to add the IBM-AI anaconda channel to your list of available software channels using below command.
-  conda config --prepend channels https://public.dhe.ibm.com/ibmdl/export/pub/software/server/ibm-ai/conda/
+  conda config --prepend channels https://ftp.osuosl.org/pub/open-ce//1.2.2/
-The above channel is still working, but the supported software version will not be updated any more. For example, you can only install TensorFlow 2.1.2 and PyTorch 1.3.1 using this channel. If you would like to install higher versions of software, please refer to [https://docs.deepsense.ca/index.php?title=Installing_Software#3._Installation_of_Deep_Learning_packages| Installing Software].
+See also [https://docs.deepsense.ca/index.php?title=Installing_Software#3._Installation_of_Deep_Learning_packages| Installing Software].
 == IBM WMLA Deep Learning Packages ==

@@ Line 35: / Line 35: @@
 == IBM WMLA Deep Learning Packages ==
-Anaconda under the WMLA install is a global install that can be accessed from any systems on DeepSense's platform. It is different from the anaconda install in users' home directories. Therefore, users cannot modify anything under the install of WMLA. However, users can use (i.e., read permission) the conda environments that have been created. If the environments created cannot satisfy your requirements, please feel free to ask the DeepSense support to create the environments for you. WMLA has following pre-developed environments which are ready to use by the users
+[https://developer.ibm.com/linuxonpower/deep-learning-powerai/ WMLA] includes multiple open source deep learning frameworks compiled for IBM Power8 systems.
   dlinsights
@@ Line 55: / Line 56: @@
   /software/WMLA/anaconda3/envs/dlipy36-wmlce170/bin/python
   (dlipy36-wmlce170) [luy@ds-lg-01 ~]$ conda list
 {| class="wikitable"
 !Framework

← Older revision		Revision as of 14:32, 6 January 2021
Line 35:		Line 35:
	== IBM WMLA Deep Learning Packages ==		== IBM WMLA Deep Learning Packages ==

		+	Anaconda under the WMLA install is a global install that can be accessed from any systems on DeepSense's platform. It is different from the anaconda install in users' home directories. Therefore, users cannot modify anything under the install of WMLA. However, users can use (i.e., read permission) the conda environments that have been created. If the environments created cannot satisfy your requirements, please feel free to ask the DeepSense support to create the environments for you. WMLA has following pre-developed environments which are ready to use by the users
		+
		+	dlinsights
		+	dlipy36-wmlce161
		+	dlipy36-wmlce162
		+	dlipy36-wmlce170
		+	dlipy37-wmlce170
		+
		+	Users can check the version of the packages under these environments by activating the environment.
		+	Below is the method for using these environments.
		+
		+	[luy@ds-lg-01 ~]$ ls /software/WMLA/anaconda3/envs
		+	dlinsights dlipy36-wmlce161 dlipy36-wmlce162 dlipy36-wmlce170 dlipy37-wmlce170
		+	[luy@ds-lg-01 ~]$ source /software/WMLA/anaconda3/etc/profile.d/conda.sh
		+	[luy@ds-lg-01 ~]$ which conda
		+	/software/WMLA/anaconda3/bin/conda
		+	[luy@ds-lg-01 ~]$ conda activate dlipy36-wmlce170
		+	(dlipy36-wmlce170) [luy@ds-lg-01 ~]$ which python
		+	/software/WMLA/anaconda3/envs/dlipy36-wmlce170/bin/python
		+	(dlipy36-wmlce170) [luy@ds-lg-01 ~]$ conda list
	[https://developer.ibm.com/linuxonpower/deep-learning-powerai/ IBM PowerAI] includes multiple open source deep learning frameworks compiled for IBM Power8 systems.		[https://developer.ibm.com/linuxonpower/deep-learning-powerai/ IBM PowerAI] includes multiple open source deep learning frameworks compiled for IBM Power8 systems.

@@ Line 45: / Line 45: @@
 |Caffe
 |/opt/DL/caffe
 |-
 |cuDNN
 |/opt/DL/cudnn
 |-
 |IBM Distributed Deep Learning (DDL)
 |/opt/DL/ddl
 |-
 | HDF5
 |/opt/DL/hdf5
 |-
 |NCCL
@@ Line 61: / Line 65: @@
 |openblas
 |/opt/DL/openblas
 |-
 |protobuf
 |/opt/DL/protobuf
 |-
 |pytorch
 |/opt/DL/pytorch
 |-
 |snap-ml
@@ Line 78: / Line 85: @@
 |Tensorboard
 |/opt/DL/tensorboard
 |}

@@ Line 31: / Line 31: @@
 To use deep learning packages like Tensorflow, pytorch, caffe on DeepSense you need to add the IBM-AI anaconda channel to your list of available software channels using below command.
   conda config --prepend channels https://public.dhe.ibm.com/ibmdl/export/pub/software/server/ibm-ai/conda/
-Check the version of the software by clicking on the above link before installation.
+The above channel is still working, but the supported software version will not be updated any more. For example, you can only install TensorFlow 2.1.2 and PyTorch 1.3.1 using this channel. If you would like to install higher versions of software, please refer to [https://docs.deepsense.ca/index.php?title=Installing_Software#3._Installation_of_Deep_Learning_packages| Installing Software].
 == IBM WMLA Deep Learning Packages ==

← Older revision		Revision as of 19:10, 4 December 2020
Line 27:		Line 27:
	See [[Installing local software]] for instructions on installing and managing your own python environments in your home directory.		See [[Installing local software]] for instructions on installing and managing your own python environments in your home directory.

		+	==IBM-AI Deep Learning Anaconda Channel==
		+
		+	To use deep learning packages like Tensorflow, pytorch, caffe on DeepSense you need to add the IBM-AI anaconda channel to your list of available software channels using below command.
		+	conda config --prepend channels https://public.dhe.ibm.com/ibmdl/export/pub/software/server/ibm-ai/conda/
		+	Check the version of the software by clicking on the above link before installation.
	== IBM WMLA Deep Learning Packages ==		== IBM WMLA Deep Learning Packages ==

@@ Line 8: / Line 8: @@
 == Anaconda Python ==
-Two Anaconda python environments are installed locally on each DeepSense compute node:
+DeepSense has two Anaconda python environments. Anaconda 2 is installed on each compute node. While Anaconda 3 is installed in a shared directory that can be accessed from any machines in the cluster.
 {| class="wikitable"
 ! Version

@@ Line 26: / Line 26: @@
 See [[Installing local software]] for instructions on installing and managing your own python environments in your home directory.
-== IBM PowerAI Deep Learning Packages ==
+== IBM WMLA Deep Learning Packages ==
 [https://developer.ibm.com/linuxonpower/deep-learning-powerai/ IBM PowerAI] includes multiple open source deep learning frameworks compiled for IBM Power8 systems.
-IBM PowerAI Enterprise includes:
+IBM WMLA Enterprise includes:
 {| class="wikitable"

@@ Line 16: / Line 16: @@
 |/opt/anaconda2
 |-
-|python 3.6.8
+|python 3.7.4
-|/opt/anaconda2/envs/py36
+|/software/WMLA/anaconda3/
 |}
@@ Line 25: / Line 25: @@
 See [[Installing local software]] for instructions on installing and managing your own python environments in your home directory.
 == IBM PowerAI Deep Learning Packages ==

@@ Line 1: / Line 1: @@
 == Basic Software ==
-* RedHat Enterprise Linux Server release 7.5 (RHEL)
+* RedHat Enterprise Linux Server release 7.7 (RHEL)
 * gcc 4.8.5
 * glibc 2.17
-* R 3.5.1
+* R 3.6.0
 == Anaconda Python ==