diff --git a/docs/mindspore/source_en/design/images/multi_level_compilation/jit_level_memory_pool.png b/docs/mindspore/source_en/design/images/multi_level_compilation/jit_level_memory_pool.png new file mode 100644 index 0000000000000000000000000000000000000000..2b8aaa9b1595fd91b55c75f396ad9c8827f6a438 Binary files /dev/null and b/docs/mindspore/source_en/design/images/multi_level_compilation/jit_level_memory_pool.png differ diff --git a/tutorials/source_en/model_infer/ms_infer/quantization.md b/tutorials/source_en/model_infer/ms_infer/quantization.md index c53a0d598cfe4c63b542f076572afdc7be700cf8..429220ab392f3a7dcc417c5d6c879f8a594816e5 100644 --- a/tutorials/source_en/model_infer/ms_infer/quantization.md +++ b/tutorials/source_en/model_infer/ms_infer/quantization.md @@ -150,7 +150,7 @@ The following provides a complete process of quantization and deployment of the ### Perceptual Quantization Training -- [SimQAT algorithm](https://www.mindspore.cn/golden_stick/docs/en/r1.0.0/quantization/simulated_quantization.html): a basic quantization aware algorithm based on the pseudo-quantization technology. +- [SimQAT algorithm](https://www.mindspore.cn/golden_stick/docs/en/r1.0.0/quantization/simqat.html): a basic quantization aware algorithm based on the pseudo-quantization technology. - [SLB quantization algorithm](https://www.mindspore.cn/golden_stick/docs/en/r1.0.0/quantization/slb.html): a non-linear low-bit quantization aware algorithm. ### Pruning diff --git a/tutorials/source_en/parallel/dynamic_cluster.md b/tutorials/source_en/parallel/dynamic_cluster.md index f49f603b9492736aae1f7ecc45ed03432995490d..6d32b81023baa823eb63b30730bc50a6b0778bce 100644 --- a/tutorials/source_en/parallel/dynamic_cluster.md +++ b/tutorials/source_en/parallel/dynamic_cluster.md @@ -402,7 +402,7 @@ That is, you can perform 2-machine 8-card distributed training tasks. ## Disaster Recovery -Dynamic cluster supports disaster recovery under data parallel. In a parallel training scenario with multi-card data, if a process quits abnormally, the training can be continued after pulling up the corresponding script of the corresponding process again, and the accuracy convergence will not be affected. Disaster recovery configuration and samples can be found in the [Disaster Recovery in Dynamic Cluster Scenarios](https://www.mindspore.cn/tutorials/en/r2.6.0/train_availability/disaster_recover.html) tutorial. +Dynamic cluster supports disaster recovery under data parallel. In a parallel training scenario with multi-card data, if a process quits abnormally, the training can be continued after pulling up the corresponding script of the corresponding process again, and the accuracy convergence will not be affected. Disaster recovery configuration and samples can be found in the [Disaster Recovery in Dynamic Cluster Scenarios](https://www.mindspore.cn/tutorials/en/r2.6.0/train_availability/fault_recover.html) tutorial. ## Security Authentication diff --git a/tutorials/source_zh_cn/model_infer/ms_infer/quantization.md b/tutorials/source_zh_cn/model_infer/ms_infer/quantization.md index 4484d522d10d249f9ec85dcc3fb6cd66e99a854f..b238e4428c8d6629c380a638b47e09c685a81bbb 100644 --- a/tutorials/source_zh_cn/model_infer/ms_infer/quantization.md +++ b/tutorials/source_zh_cn/model_infer/ms_infer/quantization.md @@ -150,7 +150,7 @@ print(output) ### 感知量化训练实例讲解 -- [SimQAT算法示例](https://www.mindspore.cn/golden_stick/docs/zh-CN/r1.0.0/quantization/simulated_quantization.html)：一种基础的基于伪量化技术的感知量化算法。 +- [SimQAT算法示例](https://www.mindspore.cn/golden_stick/docs/zh-CN/r1.0.0/quantization/simqat.html)：一种基础的基于伪量化技术的感知量化算法。 - [SLB量化算法示例](https://www.mindspore.cn/golden_stick/docs/zh-CN/r1.0.0/quantization/slb.html)：一种非线性的低比特感知量化算法。 ### 剪枝方法实例讲解 diff --git a/tutorials/source_zh_cn/parallel/dynamic_cluster.md b/tutorials/source_zh_cn/parallel/dynamic_cluster.md index c6ce425969a1c5d177ac3fb3936a83ce95847a74..02d3f1453a861f2ba2c06af90bfb558a89cea62b 100644 --- a/tutorials/source_zh_cn/parallel/dynamic_cluster.md +++ b/tutorials/source_zh_cn/parallel/dynamic_cluster.md @@ -402,7 +402,7 @@ bash run_dynamic_cluster_2.sh ## 容灾恢复 -动态组网支持数据并行下容灾恢复。在多卡数据并行训练场景下，发生进程异常退出，重新拉起对应进程对应的脚本后训练可继续，并且不影响精度收敛。容灾恢复配置和样例可参考[动态组网场景下故障恢复](https://www.mindspore.cn/tutorials/zh-CN/r2.6.0/train_availability/disaster_recover.html)教程。 +动态组网支持数据并行下容灾恢复。在多卡数据并行训练场景下，发生进程异常退出，重新拉起对应进程对应的脚本后训练可继续，并且不影响精度收敛。容灾恢复配置和样例可参考[动态组网场景下故障恢复](https://www.mindspore.cn/tutorials/zh-CN/r2.6.0/train_availability/fault_recover.html)教程。 ## 安全认证