Skip to content

Commit b5d8036

Browse files
louie-tsailouie-tsai
committed
Update README.md
1 parent 10dc876 commit b5d8036

File tree

7 files changed

+20
-17
lines changed

7 files changed

+20
-17
lines changed

‎Libraries/oneCCL/README.md‎

Lines changed: 3 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -11,8 +11,9 @@ The code samples are licensed under MIT license
1111

1212
| Type | Name | Description |
1313
| --------- | ----------------------- | ------------------------------------------------------------ |
14-
| Component | oneCCL_Getting_Started | Those C++ & C API example demonstrates basic of oneCCL programming model by invoking different collective operations such as allreduce. |
15-
| Component | oneCCL_Getting_Started.ipynb |This Jupyter Notebook demonstrates how to compile a oneCCL sample with different releases and how to port a oneCCL sample from CPU-only version to CPU&GPU version by using DPC++ via batch jobs on the Intel oneAPI DevCloud (check below Notice)|
14+
| Component | [oneCCL_Getting_Started](oneCCL_Getting_Started) | Those C++ & C API example demonstrates basic of oneCCL programming model by invoking different collective operations such as allreduce. |
15+
| Component | [tutorials](tutorials) | Hands-on Jupyter notebook tutorials among different topics. |
16+
| Component | [oneCCL_Getting_Started](oneCCL_Getting_Started.ipynb) |This Jupyter Notebook demonstrates how to compile a oneCCL sample with different releases and how to port a oneCCL sample from CPU-only version to CPU&GPU version by using DPC++ via batch jobs on the Intel oneAPI DevCloud (check below Notice)|
1617
> Notice : Please use Intel oneAPI DevCloud as the environment for jupyter notebook samples. \
1718
Users can refer to [DevCloud Getting Started](https://devcloud.intel.com/oneapi/get-started/) for using DevCloud \
1819
Users can use JupyterLab from DevCloud via "One-click Login in", and download samples via "git clone" or the "oneapi-cli" tool \

‎Libraries/oneCCL/oneCCL_Getting_Started/README.md‎

Lines changed: 4 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -6,7 +6,7 @@ By using all reduce collective operation samples, users can understand how to co
66
|:--- |:---
77
| OS | Linux Ubuntu 18.04;
88
| Hardware | Kaby Lake with GEN9 or newer
9-
| Software | Intel oneAPI Collective Communications Library (oneCCL), Intel oneAPI DPC++ Compiler, Intel oneAPI DPC++ Library (oneDPL), GNU Compiler
9+
| Software | Intel oneAPI Collective Communications Library (oneCCL), Intel oneAPI DPC++/C++ Compiler, Intel oneAPI DPC++ Library (oneDPL), GNU Compiler
1010
| What you will learn | basic oneCCL programming model for both Intel CPU and GPU
1111
| Time to complete | 15 minutes
1212

@@ -15,7 +15,6 @@ By using all reduce collective operation samples, users can understand how to co
1515
| ------ | ------ | ------ |
1616
| sycl_allreduce_cpp_test.cpp | sycl_allreduce_test.cpp |[Allreduce](https://intel.github.io/oneccl/spec/communication_primitives.html#allreduce) |
1717
| cpu_allreduce_cpp_test.cpp | cpu_allreduce_test.cpp/cpu_allreduce_bfp16.c |[Allreduce](https://intel.github.io/oneccl/spec/communication_primitives.html#allreduce) |
18-
|oneCCL_Getting_Started.ipynb (check below Notice)| | |
1918
> Notice : Please use Intel oneAPI DevCloud as the environment for jupyter notebook samples. \
2019
Users can refer to [DevCloud Getting Started](https://devcloud.intel.com/oneapi/get-started/) for using DevCloud \
2120
Users can use JupyterLab from DevCloud via "One-click Login in", and download samples via "git clone" or the "oneapi-cli" tool \
@@ -47,7 +46,7 @@ You can refer to this page [oneAPI](https://software.intel.com/en-us/oneapi) for
4746

4847
The samples below require the following components, which are part of the [Intel oneAPI Base Tookit](https://software.intel.com/en-us/oneapi/oneapi-kit)
4948
* Intel oneAPI Collective Communications Library (oneCCL)
50-
* Intel oneAPI DPC++ Compiler
49+
* Intel oneAPI DPC++/C++ Compiler
5150
* Intel oneAPI DPC++ Library (oneDPL)
5251

5352
The samples also require OpenCL driver. Please refer [System Requirements](https://software.intel.com/en-us/articles/intel-oneapi-base-toolkit-system-requirements) for OpenCL driver installation.
@@ -66,7 +65,7 @@ You can refer to this page [oneAPI](https://software.intel.com/en-us/oneapi) for
6665

6766
- Build the samples with GCC for CPU only \
6867
please replace ${ONEAPI_ROOT} for your installation path. \
69-
ex : /opt/intel/inteloneapi \
68+
ex : /opt/intel/oneapi \
7069
Don't need to replace {DPCPP_CMPLR_ROOT}
7170
```
7271
source ${ONEAPI_ROOT}/setvars.sh --ccl-configuration=cpu_icc
@@ -84,7 +83,7 @@ Users can rebuild the cpu_allreduce_cpp_test.cpp by typing "make cpu_allreduce_c
8483

8584
- Build the samples with SYCL for GPU and CPU \
8685
please replace ${ONEAPI_ROOT} for your installation path. \
87-
ex : /opt/intel/inteloneapi \
86+
ex : /opt/intel/oneapi \
8887
Don't need to replace {DPCPP_CMPLR_ROOT}
8988
```
9089
source ${ONEAPI_ROOT}/setvars.sh --ccl-configuration=cpu_gpu_dpcpp

‎Libraries/oneDNN/README.md‎

Lines changed: 3 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -16,9 +16,10 @@ The code samples are licensed under MIT license.
1616

1717
| Type | Name | Description
1818
| --------- | ------------------------------------------------ | -
19-
| Component | [getting_started](getting_started) | A C++ sample demonstrating basics of oneDNN programming model. The sample also includes a Jupyter notebook with step by step instructions on building code with different compilers and runtime configurations oneDNN support.
19+
| Component | [getting_started](getting_started) | A C++ sample demonstrating basics of oneDNN programming model.
2020
| Component | [dpcpp_interoparibility](dpcpp_interoperability) | A DPC++ example demonstrating interoperaility of oneDNN with DPC++ application code.
21-
| Component | [simple_model](simple_model) | A C++ example demonstrating implmentation of simple convolutional model with oneDNN. The samples also include a Jupyter notebook with step by step instructions on running oneDNN-based application on a GPU.
21+
| Component | [simple_model](simple_model) | A C++ example demonstrating implmentation of simple convolutional model with oneDNN.
22+
| Component | [tutorials](tutorials) | Hands-on Jupyter notebook tutorials among different topics.
2223

2324
# Using Samples in Intel oneAPI DevCloud
2425

‎Libraries/oneDNN/dpcpp_interoperability/README.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -8,7 +8,7 @@ and this interface also helps users to execute a custom SYCL kernel with oneDNN
88
| :--- | :---
99
| OS | Linux Ubuntu 18.04;
1010
| Hardware | Kaby Lake with GEN9 or newer
11-
| Software | Intel oneAPI Deep Neural Network Library (oneDNN), Intel oneAPI DPC++ Compiler, Intel oneAPI Threading Building Blocks (oneTBB)
11+
| Software | Intel oneAPI Deep Neural Network Library (oneDNN), Intel oneAPI DPC++/C++ Compiler, Intel oneAPI Threading Building Blocks (oneTBB)
1212
| What you will learn | Using oneDNN in DPC++ application targeting Intel CPU or Intel GPU
1313
| Time to complete | 15 minutes
1414

‎Libraries/oneDNN/getting_started/README.md‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -17,7 +17,7 @@ in Intel oneAPI DevCloud environment.
1717
| :--- | :---
1818
| OS | Linux* Ubuntu* 18.04; Windows 10
1919
| Hardware | Skylake with GEN9 or newer
20-
| Software | Intel oneAPI Deep Neural Network Library (oneDNN), Intel oneAPI DPC++ Compiler, Intel oneAPI Threading Building Blocks (oneTBB), GNU Compiler Collection, Intel C++ Compiler
20+
| Software | Intel oneAPI Deep Neural Network Library (oneDNN), Intel oneAPI DPC++/C++ Compiler, Intel oneAPI Threading Building Blocks (oneTBB), GNU Compiler Collection, Intel C++ Compiler
2121
| What you will learn | Running a simple convolutional model on Intel CPU or Intel GPU
2222
| Time to complete | 15 minutes
2323

@@ -67,7 +67,7 @@ make
6767
./bin/simple_model
6868
```
6969

70-
By default the sample uses oneAPI DPC++ Compiler and can execute on CPUs or
70+
By default the sample uses oneAPI DPC++/C++ Compiler and can execute on CPUs or
7171
Intel GPUs. You can build the sample with CPU support with other compilers
7272
and threading runtimes:
7373
* GNU C++ Compiler and GNU OpenMP runtime

‎Libraries/oneDNN/simple_model/README.md‎

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -9,7 +9,7 @@ in Intel oneAPI DevCloud environment.
99
| :--- | :---
1010
| OS | Linux* Ubuntu* 18.04; Windows 10
1111
| Hardware | Skylake with GEN9 or newer
12-
| Software | Intel oneAPI Deep Neural Network Library (oneDNN), Intel oneAPI DPC++ Compiler, Intel oneAPI Threading Building Blocks (oneTBB), GNU Compiler Collection, Intel C++ Compiler
12+
| Software | Intel oneAPI Deep Neural Network Library (oneDNN), Intel oneAPI DPC++/C++ Compiler, Intel oneAPI Threading Building Blocks (oneTBB), GNU Compiler Collection, Intel C++ Compiler
1313
| What you will learn | Running a simple convolutional model on Intel CPU or Intel GPU
1414
| Time to complete | 15 minutes
1515

@@ -22,7 +22,7 @@ unit (ReLU), linear response normalization (LRN), and inner product.
2222

2323
With this sample you will learn:
2424
* How to run a simple convolutional network on Intel CPU or Intel GPU
25-
* How to compile examples with Intel oneAPI DPC++ Compiler, Intel C++ Compiler,
25+
* How to compile examples with Intel oneAPI DPC++/C++ Compiler, Intel C++ Compiler,
2626
and GNU C++ Compiler
2727
* How to switch between OpenMP and TBB for CPU parallelization
2828
* How to describe tensors with oneDNN memory objects
@@ -64,7 +64,7 @@ make
6464
./bin/cnn-inference-f32-cpp
6565
```
6666

67-
By default the sample uses oneAPI DPC++ Compiler and can execute on CPUs or
67+
By default the sample uses oneAPI DPC++/C++ Compiler and can execute on CPUs or
6868
Intel GPUs. You can build the sample with CPU support with other compilers
6969
and threading runtimes:
7070
* GNU C++ Compiler and GNU OpenMP runtime

‎Libraries/oneDNN/tutorials/README.md‎

Lines changed: 4 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -11,8 +11,10 @@ The code samples are licensed under MIT license
1111

1212
| Type | Name | Description |
1313
| --------- | ----------------------- | ------------------------------------------------------------ |
14-
| Component | oneDNN_Profiling_VerboseMode_JITDump.ipynb | This Jupyter Notebook demonstrates how to use Verbose Mode and JIT Dump to profile oneDNN samples. |
15-
| Component | oneDNN_Analyze_ISA_with_DispatcherControl.ipynb | This Jupyter Notebook demonstrates how to use CPU Dispatch Control to generate JIT codes among different ISA on CPU and also analyze JIT kernels among ISAs.|
14+
| Component | [getting_started](getting_started.ipynb) | The sample also includes a Jupyter notebook with step by step instructions on building code with different compilers and runtime configurations oneDNN support. |
15+
| Component | [simple_model](simple_model.ipynb)| A Jupyter notebook with step by step instructions on running oneDNN-based application on a GPU. |
16+
| Component | [oneDNN_Profiling_VerboseMode_JITDump](oneDNN_Profiling_VerboseMode_JITDump.ipynb) | This Jupyter Notebook demonstrates how to use Verbose Mode and JIT Dump to profile oneDNN samples. |
17+
| Component | [oneDNN_Analyze_ISA_with_DispatcherControl](oneDNN_Analyze_ISA_with_DispatcherControl.ipynb) | This Jupyter Notebook demonstrates how to use CPU Dispatch Control to generate JIT codes among different ISA on CPU and also analyze JIT kernels among ISAs.|
1618
> Notice : Please use Intel oneAPI DevCloud as the environment for jupyter notebook samples. \
1719
Users can refer to [DevCloud Getting Started](https://devcloud.intel.com/oneapi/get-started/) for using DevCloud \
1820
Users can use JupyterLab from DevCloud via "One-click Login in", and download samples via "git clone" or the "oneapi-cli" tool \

0 commit comments

Comments
 (0)