diff --git a/PyTorch/built-in/audio/ESPnet2_for_PyTorch/README.md b/PyTorch/built-in/audio/ESPnet2_for_PyTorch/README.md
index 2d673ee02540f9df590094cfb497f804b4061bcb..09c31729223e2af488a5ca5495b2ec1c83037cb8 100644
--- a/PyTorch/built-in/audio/ESPnet2_for_PyTorch/README.md
+++ b/PyTorch/built-in/audio/ESPnet2_for_PyTorch/README.md
@@ -333,16 +333,16 @@ Conformer是将CNN用于增强Transformer来做ASR的结构
```
npu-smi info
- #该设备芯片名为Ascend910A （自行替换）
+ #该设备芯片名为Atlas （自行替换）
回显如下：
+-------------------|-----------------|------------------------------------------------------+
| NPU Name | Health | Power(W) Temp(C) Hugepages-Usage(page) |
| Chip Device | Bus-Id | AICore(%) Memory-Usage(MB) |
+===================+=================+======================================================+
- | 0 910A | OK | 15.8 42 0 / 0 |
+ | 0 Atlas | OK | 15.8 42 0 / 0 |
| 0 0 | 0000:82:00.0 | 0 1074 / 21534 |
+===================+=================+======================================================+
- | 1 910A | OK | 15.4 43 0 / 0 |
+ | 1 Atlas | OK | 15.4 43 0 / 0 |
| 0 1 | 0000:89:00.0 | 0 1070 / 21534 |
+===================+=================+======================================================+
```
@@ -352,10 +352,10 @@ Conformer是将CNN用于增强Transformer来做ASR的结构
将xformer_encoder.sh，xformer_decoder.sh，transformer_lm.sh，ctc.sh放置到/root/.cache/espnet_onnx/asr_train_asr_qkv/full目录下，运行xformer_encoder.sh导出encoder`OM`模型，默认保存在当前文件夹下，其他模型类似。
```
- bash xformer_encoder.sh Ascend910A
- bash xformer_decoder.sh Ascend910A
- bash transformer_lm.sh Ascend910A
- bash ctc.sh Ascend910A
+ bash xformer_encoder.sh Atlas
+ bash xformer_decoder.sh Atlas
+ bash transformer_lm.sh Atlas
+ bash ctc.sh Atlas
```
### 2 开始推理验证
@@ -407,6 +407,6 @@ Conformer是将CNN用于增强Transformer来做ASR的结构
| 芯片型号 | 配置 | 数据集 | 精度(overall) | 性能(fps) |
| :-----------: | :------------------------------------: | :-------: | :-------------: |
- | GPU | encoder/decoder/ctc/lm(beam_size=20) | aishell | 95.27% |
- | GPU | encoder/decoder/ctc/lm(beam_size=2) | aishell | 95.08% |
- | Ascend910A | encoder/decoder/ctc/lm(default) | aishell | 95.02% |
+ | 竞品A | encoder/decoder/ctc/lm(beam_size=20) | aishell | 95.27% |
+ | 竞品A | encoder/decoder/ctc/lm(beam_size=2) | aishell | 95.08% |
+ | Atlas | encoder/decoder/ctc/lm(default) | aishell | 95.02% |
diff --git a/PyTorch/built-in/cv/classification/Beit2_for_PyTorch/README.md b/PyTorch/built-in/cv/classification/Beit2_for_PyTorch/README.md
index e5881a77cdc1721e6960c5e0e2e5ffbc1cb1a184..5fcc14711f4a7c59ebaf85e0c259cae7139828fd 100644
--- a/PyTorch/built-in/cv/classification/Beit2_for_PyTorch/README.md
+++ b/PyTorch/built-in/cv/classification/Beit2_for_PyTorch/README.md
@@ -153,7 +153,7 @@ cd ..
**表 2** 训练结果展示表
-这里使用了单机进行预训练，采用的NPU型号为910B1
+这里使用了单机进行预训练，采用的NPU型号为Atlas 900 A2 PODc
| NAME | single-step time | Iterations | DataType|Torch_Version |
|:-:|:-:|:-:|:-:|:-:|
diff --git a/PyTorch/built-in/cv/classification/CRNN_for_PyTorch/README.md b/PyTorch/built-in/cv/classification/CRNN_for_PyTorch/README.md
index 3c9c9537e5bb576ce7291e4504674bea222d11a3..441cfe34b8f91e1fe7835ef284bf29b52aca296c 100644
--- a/PyTorch/built-in/cv/classification/CRNN_for_PyTorch/README.md
+++ b/PyTorch/built-in/cv/classification/CRNN_for_PyTorch/README.md
@@ -301,7 +301,7 @@ CRNN (Convolutional Recurrent Neural Network) 于2015年由华中科技大学的
```shell
npu-smi info
```
- 该设备芯片名为Ascend910A （请根据实际芯片填入）
+ 该设备芯片名为Atlas （请根据实际芯片填入）
4. 执行atc命令
@@ -375,7 +375,7 @@ CRNN (Convolutional Recurrent Neural Network) 于2015年由华中科技大学的
| 芯片型号 | Batch Size | 数据集 | 精度 |
| -------- | ---------- | ----------- | ------ |
-| 910A | 16 | IIIT5K_lmdb | 76.57% |
+| Atlas | 16 | IIIT5K_lmdb | 76.57% |
# 公网地址说明
diff --git a/PyTorch/built-in/cv/classification/ResNet50_for_PyTorch/README.md b/PyTorch/built-in/cv/classification/ResNet50_for_PyTorch/README.md
index 1ae40c101aeb46efb79945cb9553fbd505826722..09a3baa2dfd3ec939991a6be3ab4c28f91443a24 100644
--- a/PyTorch/built-in/cv/classification/ResNet50_for_PyTorch/README.md
+++ b/PyTorch/built-in/cv/classification/ResNet50_for_PyTorch/README.md
@@ -297,16 +297,16 @@ ResNet是ImageNet竞赛中分类问题效果较好的网络，它引入了残差
```
npu-smi info
- #该设备芯片名为Ascend910A （请根据实际芯片填入）
+ #该设备芯片名为Atlas （请根据实际芯片填入）
回显如下：
+-------------------+-----------------+------------------------------------------------------+
| NPU Name | Health | Power(W) Temp(C) Hugepages-Usage(page) |
| Chip Device | Bus-Id | AICore(%) Memory-Usage(MB) |
+===================+=================+======================================================+
- | 0 910A | OK | 69.5 40 0 / 0 |
+ | 0 Atlas | OK | 69.5 40 0 / 0 |
| 0 0 | 0000:82:00.0 | 0 950 / 15137 |
+===================+=================+======================================================+
- | 1 910A | OK | 65.3 36 0 / 0 |
+ | 1 Atlas | OK | 65.3 36 0 / 0 |
| 0 1 | 0000:89:00.0 | 0 1613 / 15137 |
+===================+=================+======================================================+
```
@@ -373,7 +373,7 @@ c. 精度验证。
| 芯片型号 | Batch Size | 数据集 | 精度 |
| --------- | ---------------- | ---------- | ---------- |
-| 910A | 64 | ImageNet | top-1: 76.96%
top-5: 93.24% |
+| Atlas | 64 | ImageNet | top-1: 76.96%
top-5: 93.24% |
diff --git a/PyTorch/built-in/cv/detection/DB_ID0706_for_PyTorch/README.md b/PyTorch/built-in/cv/detection/DB_ID0706_for_PyTorch/README.md
index 42a28a1122c9f96538aec86e8c0193e9e9685a89..a2f55fa45f04154dc27b28f20ac74a937dccd61d 100644
--- a/PyTorch/built-in/cv/detection/DB_ID0706_for_PyTorch/README.md
+++ b/PyTorch/built-in/cv/detection/DB_ID0706_for_PyTorch/README.md
@@ -338,16 +338,16 @@ DB(Differentiable Binarization)是一种使用可微分二值图来实时文字
```sh
npu-smi info
- #该设备芯片名为Ascend910A （自行替换）
+ #该设备芯片名为Atlas （自行替换）
回显如下：
+-------------------+-----------------+------------------------------------------------------+
| NPU Name | Health | Power(W) Temp(C) Hugepages-Usage(page) |
| Chip Device | Bus-Id | AICore(%) Memory-Usage(MB) |
+===================+=================+======================================================+
- | 0 910A | OK | 15.8 42 0 / 0 |
+ | 0 Atlas | OK | 15.8 42 0 / 0 |
| 0 0 | 0000:82:00.0 | 0 1074 / 21534 |
+===================+=================+======================================================+
- | 1 910A | OK | 15.4 43 0 / 0 |
+ | 1 Atlas | OK | 15.4 43 0 / 0 |
| 0 1 | 0000:89:00.0 | 0 1070 / 21534 |
+===================+=================+======================================================+
```
@@ -428,7 +428,7 @@ DB(Differentiable Binarization)是一种使用可微分二值图来实时文字
| 芯片型号 | Batch Size | 数据集 | 精度 |
| :------: | :--------: | :-------: | :--: |
-| 910A | 1 | icdar2015 | 0.896 |
+| Atlas | 1 | icdar2015 | 0.896 |
# 公网地址说明
diff --git a/PyTorch/built-in/cv/semantic_segmentation/BiseNetV1_for_PyTorch/README.md b/PyTorch/built-in/cv/semantic_segmentation/BiseNetV1_for_PyTorch/README.md
index a31255d1fcb7e88b83b30c9b201d827faf21d072..280752635b45bde85470000884b03b7a357c9f0f 100644
--- a/PyTorch/built-in/cv/semantic_segmentation/BiseNetV1_for_PyTorch/README.md
+++ b/PyTorch/built-in/cv/semantic_segmentation/BiseNetV1_for_PyTorch/README.md
@@ -188,10 +188,10 @@
| Name | mIoU | FPS | Device | Npu_nums | Steps | AMP_Type | CPU |
|-----------|:-----:|:---:|:--------:|:--------:|:-----:|:--------:|:---:|
| 1p-*PU | - | 9 | - | - | 400 | O1 | x86 |
-| 1p-NPU1.8 | - | 12 | 910A | 1 | 400 | O1 | ARM |
+| 1p-NPU1.8 | - | 12 | Atlas | 1 | 400 | O1 | ARM |
| 8p-*PU | 75.80 | 62 | - | - | 40000 | O1 | x86 |
-| 8p-NPU1.8 | - | 88 | 910A | 8 | 400 | O1 | ARM |
-| 8p-NPU1.8 | 76.03 | 88 | 910A | 8 | 40000 | O1 | ARM |
+| 8p-NPU1.8 | - | 88 | Atlas | 8 | 400 | O1 | ARM |
+| 8p-NPU1.8 | 76.03 | 88 | Atlas | 8 | 40000 | O1 | ARM |
# 公网地址说明
diff --git a/PyTorch/built-in/diffusion/diffusers/README.md b/PyTorch/built-in/diffusion/diffusers/README.md
index 7fc330772340e16199ba9070b5885100673a9da3..72d9523c8ea9927628044eba5978c0327389b56e 100644
--- a/PyTorch/built-in/diffusion/diffusers/README.md
+++ b/PyTorch/built-in/diffusion/diffusers/README.md
@@ -60,16 +60,16 @@
- [推理任务](#推理任务-4)
- [获取预训练模型](#获取预训练模型-4)
- [开始推理](#开始推理-4)
-- [SD3](#SD3)
- - [准备环境](#准备环境-5)
+- [SD3](#sd3)
+ - [准备环境](#准备环境-2)
- [安装模型环境](#安装模型环境-5)
- [安装昇腾环境](#安装昇腾环境-5)
- [快速开始](#快速开始-5)
- - [训练任务](#训练任务-5)
+ - [训练任务](#训练任务)
- [获取预训练模型](#获取预训练模型-5)
- - [开始训练](#开始训练-1)
+ - [开始训练](#开始训练-1)
- [推理任务](#推理任务-5)
- - [开始推理](#开始推理-5)
+ - [开始推理](#开始推理-5)
- [公网地址说明](#公网地址说明)
- [变更说明](#变更说明)
- [变更](#变更)
@@ -313,11 +313,11 @@ https://huggingface.co/docs/diffusers/installation
| 芯片 | 卡数 | 任务 | FPS | batch_size | AMP_Type | Torch_Version | deepspeed |
|:---:|:---:|:----------:|:-----:|:----------:|:---:|:---:|:---:|
-| GPU | 8p | LoRA | 23.38 | 7 | fp16 | 2.1 | ✔ |
+| 竞品A | 8p | LoRA | 23.38 | 7 | fp16 | 2.1 | ✔ |
| Atlas A2 |8p | LoRA | 28.75 | 7 | fp16 | 2.1 | ✔ |
-| GPU | 8p | Controlnet | 32.5 | 5 | fp16 | 2.1 | ✔ |
+| 竞品A | 8p | Controlnet | 32.5 | 5 | fp16 | 2.1 | ✔ |
| Atlas A2 |8p | Controlnet | 28.42 | 5 | fp16 | 2.1 | ✔ |
-| GPU | 8p | Finetune | 142.7 | 24 | fp16 | 2.1 | ✔ |
+| 竞品A | 8p | Finetune | 142.7 | 24 | fp16 | 2.1 | ✔ |
| Atlas A2 |8p | Finetune | 172.9 | 24 | fp16 | 2.1 | ✔ |
### 推理任务
diff --git a/PyTorch/built-in/diffusion/sd-scripts-xl/README.md b/PyTorch/built-in/diffusion/sd-scripts-xl/README.md
index 17fdb827791ffb00851180d33a1445a7752ce7db..1519e09b51a96847801115a1e662af83ce0d6771 100644
--- a/PyTorch/built-in/diffusion/sd-scripts-xl/README.md
+++ b/PyTorch/built-in/diffusion/sd-scripts-xl/README.md
@@ -1,18 +1,26 @@
# sd-scripts-xl for PyTorch
# 目录
-- [简介](#简介)
- - [模型介绍](#模型介绍)
- - [支持任务列表](#支持任务列表)
- - [代码实现](#代码实现)
-- [sd-scripts-xl](#sd-scripts-xl)
- - [准备训练环境](#准备训练环境)
- - [准备数据集](#准备数据集)
- - [快速开始](#快速开始)
- - [预训练任务(SDXL+CLIP)](#预训练任务sdxlclip)
- - [预训练任务(SDXL+MT5)](#预训练任务sdxlmt5)
-- [公网地址说明](#公网地址说明)
-- [变更说明](#变更说明)
-- [FAQ](#FAQ)
+- [sd-scripts-xl for PyTorch](#sd-scripts-xl-for-pytorch)
+- [目录](#目录)
+- [简介](#简介)
+ - [模型介绍](#模型介绍)
+ - [支持任务列表](#支持任务列表)
+ - [代码实现](#代码实现)
+- [sd-scripts-xl](#sd-scripts-xl)
+ - [准备训练环境](#准备训练环境)
+ - [安装模型环境](#安装模型环境)
+ - [安装昇腾环境](#安装昇腾环境)
+ - [准备数据集](#准备数据集)
+ - [获取预训练模型](#获取预训练模型)
+- [快速开始](#快速开始)
+ - [预训练任务(SDXL+CLIP)](#预训练任务sdxlclip)
+ - [开始训练](#开始训练)
+ - [预训练任务(SDXL+MT5)](#预训练任务sdxlmt5)
+ - [开始训练](#开始训练-1)
+- [训练结果展示](#训练结果展示)
+- [公网地址说明](#公网地址说明)
+- [变更说明](#变更说明)
+- [FAQ](#faq)
@@ -309,7 +317,7 @@ replace_token_length $mt5_tokenizer_path/tokenizer_config.json
| NAME | sd版本 | FPS | batch_size | AMP_Type | Torch_Version |
| :------: | :---: | :--: | :------: | :-----------: | :-----------: |
-| GPU | xl | 20.2 | 4 | fp16 | 1.13 |
+| 竞品A | xl | 20.2 | 4 | fp16 | 1.13 |
| Atlas A2 | xl | 10.4 | 4 | fp16 | 1.11 |
@@ -317,7 +325,7 @@ replace_token_length $mt5_tokenizer_path/tokenizer_config.json
| NAME | sd版本 | FPS | batch_size | AMP_Type | Torch_Version |
| :------: | :---: | :--: |:----------:|:--------:| :-----------: |
-| GPU | xl | 9.754 | 2 | bf16 | 1.13 |
+| 竞品A | xl | 9.754 | 2 | bf16 | 1.13 |
| Atlas A2 | xl | 10.71| 2 | bf16 | 1.11 |
diff --git a/PyTorch/built-in/foundation/Aquila2/README.md b/PyTorch/built-in/foundation/Aquila2/README.md
index 220ac3d7a96c726a05f3bf94ade0d9ab72af04ae..c24212b0de5dab559eceb030e0ec9a77b3714323 100644
--- a/PyTorch/built-in/foundation/Aquila2/README.md
+++ b/PyTorch/built-in/foundation/Aquila2/README.md
@@ -1,16 +1,25 @@
# Aquila2 for Pytorch
# 目录
-- [简介](#简介)
- - [模型介绍](#模型介绍)
- - [代码实现](#代码实现)
-- [Aquila2](#Aquila2)
- - [准备训练环境](#准备训练环境)
- - [快速开始](#快速开始)
- - [预训练任务](#预训练任务)
-- [公网地址说明](#公网地址说明)
-- [变更说明](#变更说明)
-- [FAQ](#FAQ)
+- [Aquila2 for Pytorch](#aquila2-for-pytorch)
+- [目录](#目录)
+- [简介](#简介)
+ - [模型介绍](#模型介绍)
+ - [代码实现](#代码实现)
+- [Aquila2](#aquila2)
+ - [准备训练环境](#准备训练环境)
+ - [安装模型环境](#安装模型环境)
+ - [安装昇腾环境](#安装昇腾环境)
+ - [准备数据集](#准备数据集)
+ - [预训练数据集准备](#预训练数据集准备)
+ - [快速开始](#快速开始)
+ - [预训练任务](#预训练任务)
+ - [开始训练](#开始训练)
+ - [训练结果](#训练结果)
+- [公网地址说明](#公网地址说明)
+- [变更说明](#变更说明)
+ - [变更](#变更)
+- [FAQ](#faq)
# 简介
## 模型介绍
@@ -144,9 +153,9 @@ Aquila2是智源发布的业内领先的大语言模型，在多个领域都有
**表 3** 训练结果展示表
| 芯片 | 卡数 | 参数规模 | seq_length | micro_batch_size | global_batch_size | 单步迭代时间 (s/step) | tokens吞吐 (tokens/s/p)
|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
-| GPU | 16p | 34B | 4096 | 1 | 32 | 10.8 | 756 |
+| 竞品A | 16p | 34B | 4096 | 1 | 32 | 10.8 | 756 |
| Atlas A2 | 16p | 34B | 4096 | 2 | 64 | - | - |
-| GPU | 32p | 70B | 4096 | 1 | 44 | - | - |
+| 竞品A | 32p | 70B | 4096 | 1 | 44 | - | - |
| Atlas A2 | 32p | 70B | 4096 | 1 | 44 | - | - |
# 公网地址说明
diff --git a/PyTorch/built-in/foundation/ChatGLM-6B/README.md b/PyTorch/built-in/foundation/ChatGLM-6B/README.md
index e6ef63c9ea9d12d229cdbbcab21e154b5e985e79..a55239aae09b382d8817f6a6a808e61847b770dd 100644
--- a/PyTorch/built-in/foundation/ChatGLM-6B/README.md
+++ b/PyTorch/built-in/foundation/ChatGLM-6B/README.md
@@ -210,13 +210,13 @@ bash preprocess.sh
| NAME | SamplesPerSec | Iterations | DataType | Torch_Version | Card |
|:-------------:|:-------------:|:-:|:-:|:-:|:----:|
| Finetune -NPU | 2213 | 5000 | fp16 | 1.11 | 910 |
-| Finetune -GPU | 2048 | 5000 | fp16 | 1.11 | A800 |
+| Finetune -竞品A | 2048 | 5000 | fp16 | 1.11 | 竞品A |
说明：P-Tuning 仅打通功能，无性能优化。
**表 2** 评估结果展示表
-| 评估项 | NPU | GPU |
+| 评估项 | NPU | 竞品A |
|:-------:|:-------:|:-------:|
| BLEU-4 | 8.2853 | 8.1127 |
| ROUGE-1 | 31.1898 | 30.7429 |
diff --git a/PyTorch/built-in/foundation/ChatGLM3-6B/README.md b/PyTorch/built-in/foundation/ChatGLM3-6B/README.md
index a1a51980b86a8edefae864696a3505af566faf70..690e8aaf5d7672f81e9b816372dddb94ac53f3f6 100644
--- a/PyTorch/built-in/foundation/ChatGLM3-6B/README.md
+++ b/PyTorch/built-in/foundation/ChatGLM3-6B/README.md
@@ -189,9 +189,9 @@
| 芯片 | 卡数 | 模型 | Iterations | Global Batch Size | Train Samples per Second
| --------- |---| ----------- | ---------------- | ----------------------------- | ---------------------------- |
| Atlas A2 |8p| ChatGLM3-6B | 2000 | 16 |13.781 |
-| GPU |8p| ChatGLM3-6B | 2000 | 16 |15.094 |
+| 竞品A |8p| ChatGLM3-6B | 2000 | 16 |15.094 |
| Atlas A2 |8p| ChatGLM3-6B-32K | 2000 | 16 | 11.819 |
-| GPU |8p| ChatGLM3-6B-32K | 2000 | 16 |12.088 |
+| 竞品A |8p| ChatGLM3-6B-32K | 2000 | 16 |12.088 |
diff --git a/PyTorch/built-in/foundation/CodeGeeX2/README.md b/PyTorch/built-in/foundation/CodeGeeX2/README.md
index cdf7bbb9af665732b1f93252be3b59dd6c05a99b..2ec3a54629d894c286c4bf2ca405c48c54271ba4 100644
--- a/PyTorch/built-in/foundation/CodeGeeX2/README.md
+++ b/PyTorch/built-in/foundation/CodeGeeX2/README.md
@@ -225,14 +225,14 @@ bash preprocess.sh
| NAME | SamplesPerSec | Iterations | DataType | Torch_Version | Card |
| :-----------: | :-----------: | :--------: | :------: | :-----------: | :--: |
-| Finetune -NPU | (待补充) | (待补充) | bf16? | 1.11 | 910B |
-| Finetune -GPU | (待补充) | (待补充) | bf16? | 1.11 | A800 |
+| Finetune -NPU | (待补充) | (待补充) | bf16? | 1.11 | Atlas 900 A2 PODc |
+| Finetune -竞品A | (待补充) | (待补充) | bf16? | 1.11 | 竞品A |
说明：P-Tuning 仅打通功能，无性能优化。
**表 2** 评估结果展示表
-| 评估项 | NPU | GPU |
+| 评估项 | NPU | 竞品A |
| :-----: | :--: | :--: |
| human pass@1 | 0.37 | 0.35 |
diff --git a/PyTorch/built-in/foundation/CodeShell-7B/README.md b/PyTorch/built-in/foundation/CodeShell-7B/README.md
index 2153c61de20329497246df5075eb4357d548d696..f5a729a64c9d39b94628542dcdd6567d1b3b5cb9 100644
--- a/PyTorch/built-in/foundation/CodeShell-7B/README.md
+++ b/PyTorch/built-in/foundation/CodeShell-7B/README.md
@@ -2,16 +2,25 @@
## 目录
-- [简介](#简介)
- - [模型介绍](#模型介绍)
- - [支持任务列表](#支持任务列表)
- - [代码实现](#代码实现)
-- [CodeShell](#CodeShell)
- - [准备训练环境](#准备训练环境)
- - [快速开始](#快速开始)
-- [公网地址说明](#公网地址说明)
-- [变更说明](#变更说明)
-- [FAQ](#FAQ)
+- [CodeShell-7B for PyTorch](#codeshell-7b-for-pytorch)
+ - [目录](#目录)
+- [简介](#简介)
+ - [模型介绍](#模型介绍)
+ - [支持任务列表](#支持任务列表)
+ - [代码实现](#代码实现)
+- [CodeShell](#codeshell)
+ - [准备训练环境](#准备训练环境)
+ - [安装环境](#安装环境)
+ - [安装昇腾环境](#安装昇腾环境)
+ - [准备预训练权重](#准备预训练权重)
+ - [准备数据集](#准备数据集)
+ - [快速开始](#快速开始)
+ - [训练任务](#训练任务)
+ - [开始训练](#开始训练)
+ - [训练结果](#训练结果)
+- [公网地址说明](#公网地址说明)
+- [变更说明](#变更说明)
+- [FAQ](#faq)
# 简介
@@ -156,7 +165,7 @@ python convert_alpaca.py --in-file finetune/alpaca_data.json --out-file finetune
| 芯片 | 卡数 | Batch size | Steps | Train_Samples_Per_Second |
|----------|:--------:|:----------:|:-----:|:------------------------:|
-| GPU | 8p | 6 | 2000 | 40.952 |
+| 竞品A | 8p | 6 | 2000 | 40.952 |
| Atlas-A2 | 8p | 6 | 2000 | 36.801 |
diff --git a/PyTorch/built-in/foundation/GPT-NeoX/README.md b/PyTorch/built-in/foundation/GPT-NeoX/README.md
index 2d9d24c1c02d059775df9820715ae7a924a98b8a..cb6868e85a573cbf8d059d664be901202c864705 100644
--- a/PyTorch/built-in/foundation/GPT-NeoX/README.md
+++ b/PyTorch/built-in/foundation/GPT-NeoX/README.md
@@ -323,7 +323,7 @@ GPT-NeoX-20B 是由EleutherAI和Hugging face合作开发的一个超大规模的
| NAME | tflops | Iterations | DataType | Torch_Version | Card |
|:-------------:|:-------------:|:-:|:-:|:-:|:----:|
-| GPU-2pp4mp2dp | 100 | 5000 | fp16 | 1.5 | A100 |
+| 竞品A-2pp4mp2dp | 100 | 5000 | fp16 | 1.5 | 竞品A |
| NPU-2pp4mp2dp | 150 | 5000 | fp16 | 1.5 | 910 |
diff --git a/PyTorch/built-in/foundation/LLaMA-13B/README.md b/PyTorch/built-in/foundation/LLaMA-13B/README.md
index cfe099663effc31cce333d3c15e34aa5d6c3a1b1..049fe1ce35b763e575f7ac1990d3ef484039d61d 100644
--- a/PyTorch/built-in/foundation/LLaMA-13B/README.md
+++ b/PyTorch/built-in/foundation/LLaMA-13B/README.md
@@ -201,7 +201,7 @@ per_bs * grad_acc * seq_len / time
| 13B-NPU(单机20层) | 1619 | 3 |
| 13B-竞品A(单机20层) | 1740 | 3 |
-注：这里vicuna 7B/13B在NPU上使用910B3（313T）训练，竞品使用A800训练
+注：这里vicuna 7B/13B在NPU上使用Atlas 800T A2训练
# 推理
## 推理环境搭建
这里要替换transformers库中的部分文件，用于推理（评估）场景，后续如果要进行训练再更改为transformers_modify中的文件。
diff --git a/PyTorch/built-in/mm/AltCLIP/README.md b/PyTorch/built-in/mm/AltCLIP/README.md
index 85d155a08d7e7ff90afba66b05f9d9c83beb8069..86b0b3b2c5445c4c238410e4f282cc38fb102ec4 100644
--- a/PyTorch/built-in/mm/AltCLIP/README.md
+++ b/PyTorch/built-in/mm/AltCLIP/README.md
@@ -2,17 +2,25 @@
## 目录
-- [简介](#简介)
- - [模型介绍](#模型介绍)
- - [支持任务列表](#支持任务列表)
- - [代码实现](#代码实现)
-- [AltCLIP](#AltCLIP)
- - [准备训练环境](#准备训练环境)
- - [快速开始](#快速开始)
- - [CIFAR10微调任务](#CIFAR10微调任务)
-- [公网地址说明](#公网地址说明)
-- [变更说明](#变更说明)
-- [FAQ](#FAQ)
+- [AltCLIP for PyTorch](#altclip-for-pytorch)
+ - [目录](#目录)
+- [简介](#简介)
+ - [模型介绍](#模型介绍)
+ - [支持任务列表](#支持任务列表)
+ - [代码实现](#代码实现)
+- [AltCLIP](#altclip)
+ - [准备训练环境](#准备训练环境)
+ - [安装环境](#安装环境)
+ - [安装昇腾环境](#安装昇腾环境)
+ - [准备预训练权重](#准备预训练权重)
+ - [准备数据集](#准备数据集)
+ - [快速开始](#快速开始)
+ - [CIFAR10微调任务](#cifar10微调任务)
+ - [开始训练](#开始训练)
+ - [训练结果](#训练结果)
+- [公网地址说明](#公网地址说明)
+- [变更说明](#变更说明)
+- [FAQ](#faq)
# 简介
@@ -166,7 +174,7 @@ clip_benchmark_datasets
#### 训练结果
| 芯片 | 卡数 | 精度acc | 性能FPS | batch size | Precision | Torch Version |
| -------------------- | :--: | :-----: | :-----: | :--------: | :-------: | :-----------: |
-| GPU | 8p | 0.9737 | 338 | 512 | bf16 | 2.1 |
+| 竞品A | 8p | 0.9737 | 338 | 512 | bf16 | 2.1 |
| Atlas A200T A2 Box16 | 8p | 0.9732 | 295 | 512 | bf16 | 2.1 |
diff --git a/PyTorch/built-in/mm/AnimateDiff/README.md b/PyTorch/built-in/mm/AnimateDiff/README.md
index 068eaf6956ebe486ca39c0c1c35aa1cddedc7a40..6e806d871bc84be99cc9323336fa8ed8ad412ca9 100644
--- a/PyTorch/built-in/mm/AnimateDiff/README.md
+++ b/PyTorch/built-in/mm/AnimateDiff/README.md
@@ -4,10 +4,12 @@
# 目录
-- [AnimateDiff](#animatediff-for-pytorch)
+- [AnimateDiff for PyTorch](#animatediff-for-pytorch)
+- [目录](#目录)
- [概述](#概述)
+ - [模型介绍](#模型介绍)
- [准备训练环境](#准备训练环境)
- - [创建Python环境](#创建Python环境)
+ - [创建Python环境](#创建python环境)
- [准备数据集](#准备数据集)
- [准备预训练权重](#准备预训练权重)
- [准备推理权重](#准备推理权重)
@@ -211,7 +213,7 @@ AnimateDiff提出了一个有效的框架，可将现有的大多数个性化文
| 芯片 | 卡数 | samples per second | batch_size | AMP_Type | Torch_Version |
|:---:|:---:|:------------------:|:----------:|:--------:|:---:|
-| GPU | 8p | 469.1 | 64 | fp16 | 2.1 |
+| 竞品A | 8p | 469.1 | 64 | fp16 | 2.1 |
| Atlas A2 | 8p | 410.7 | 64 | fp16 | 2.1 |
### 模型推理
diff --git a/PyTorch/built-in/mm/DiT/README.md b/PyTorch/built-in/mm/DiT/README.md
index b9da5949d199632c179c8fc5b2a34345c17cc632..f3d2274aa45f39c61217f89fc81742103a4669dc 100644
--- a/PyTorch/built-in/mm/DiT/README.md
+++ b/PyTorch/built-in/mm/DiT/README.md
@@ -2,18 +2,27 @@
## 目录
-- [简介](#简介)
- - [模型介绍](#模型介绍)
- - [支持任务列表](#支持任务列表)
- - [代码实现](#代码实现)
-- [DiT](#DiT)
- - [准备训练环境](#准备训练环境)
- - [快速开始](#快速开始)
- - [训练任务](#训练任务)
- - [在线推理](#在线推理)
-- [公网地址说明](#公网地址说明)
-- [变更说明](#变更说明)
-- [FAQ](#FAQ)
+- [DiT for PyTorch](#dit-for-pytorch)
+ - [目录](#目录)
+- [简介](#简介)
+ - [模型介绍](#模型介绍)
+ - [支持任务列表](#支持任务列表)
+ - [代码实现](#代码实现)
+- [DiT](#dit)
+ - [准备训练环境](#准备训练环境)
+ - [安装环境](#安装环境)
+ - [安装昇腾环境](#安装昇腾环境)
+ - [准备预训练权重](#准备预训练权重)
+ - [准备数据集](#准备数据集)
+ - [快速开始](#快速开始)
+ - [训练任务](#训练任务)
+ - [开始训练](#开始训练)
+ - [训练结果](#训练结果)
+ - [在线推理](#在线推理)
+ - [开始推理](#开始推理)
+- [公网地址说明](#公网地址说明)
+- [变更说明](#变更说明)
+- [FAQ](#faq)
# 简介
@@ -156,13 +165,13 @@ Scalable Diffusion Models with Transformers，是完全基于transformer架构
#### 训练结果
| 芯片 | 卡数 | image size | global batch size | Precision | 性能FPS |
| ------------- | :--: | :--------: | :---------------: | :-------: | :-----: |
-| GPU | 8p | 256 | 256 | fp32 | 432 |
+| 竞品A | 8p | 256 | 256 | fp32 | 432 |
| Atlas 800T A2 | 8p | 256 | 256 | fp32 | 376 |
-| GPU | 8p | 256 | 512 | bf16 | 727 |
+| 竞品A | 8p | 256 | 512 | bf16 | 727 |
| Atlas 800T A2 | 8p | 256 | 512 | bf16 | 586 |
-| GPU | 8p | 512 | 64 | fp32 | 80 |
+| 竞品A | 8p | 512 | 64 | fp32 | 80 |
| Atlas 800T A2 | 8p | 512 | 64 | fp32 | 77 |
-| GPU | 8p | 512 | 128 | bf16 | 151 |
+| 竞品A | 8p | 512 | 128 | bf16 | 151 |
| Atlas 800T A2 | 8p | 512 | 128 | bf16 | 122 |
### 在线推理
diff --git a/PyTorch/built-in/mm/LLaVA/README.md b/PyTorch/built-in/mm/LLaVA/README.md
index a86a53a088994c8f49312a93d917e434f9a840f9..8044bfc5cbcc883c80d1a1a00cbfb3f7bbde0d22 100644
--- a/PyTorch/built-in/mm/LLaVA/README.md
+++ b/PyTorch/built-in/mm/LLaVA/README.md
@@ -4,8 +4,12 @@
# 目录
-- [LLaVA](#llava-for-pytorch)
+- [LLaVA for PyTorch](#llava-for-pytorch)
+- [目录](#目录)
- [概述](#概述)
+ - [模型介绍](#模型介绍)
+ - [支持任务列表](#支持任务列表)
+ - [代码实现](#代码实现)
- [准备训练环境](#准备训练环境)
- [创建Python环境](#创建python环境)
- [准备数据集](#准备数据集)
@@ -17,7 +21,7 @@
- [模型推理](#模型推理)
- [公网地址说明](#公网地址说明)
- [变更说明](#变更说明)
- - [FQA](#faq)
+ - [FAQ](#faq)
@@ -143,7 +147,7 @@ LLaVA是一种新颖的端到端训练的大型多模态模型，它结合了视
| 芯片 | 卡数 | samples per second | batch_size | AMP_Type | Torch_Version |
|:---:|:---:|:------------------:|:----------:|:---:|:---:|
-| GPU | 8p | 18.62 | 16 | bf16 | 2.1 |
+| 竞品A | 8p | 18.62 | 16 | bf16 | 2.1 |
| Atlas A2 | 8p | 20.13 | 16 | bf16 | 2.1 |
### 模型评估
diff --git a/PyTorch/built-in/mm/OpenSora-master/docs/zh_CN/README.md b/PyTorch/built-in/mm/OpenSora-master/docs/zh_CN/README.md
index 43753e357197dc5697de01d0cc92453b684b7fae..8b4c4203767485a01ee25bc9751aadf55243dcd9 100644
--- a/PyTorch/built-in/mm/OpenSora-master/docs/zh_CN/README.md
+++ b/PyTorch/built-in/mm/OpenSora-master/docs/zh_CN/README.md
@@ -213,7 +213,7 @@ docker run -ti --gpus all -v {MOUNT_DIR}:/data opensora
查看更多
-| 分辨率 | 模型大小 | 数据 | 迭代次数 | 批量大小 | GPU 天数 (H800) | 网址
+| 分辨率 | 模型大小 | 数据 | 迭代次数 | 批量大小 | 竞品A 天数 (H800) | 网址
| ---------- | ---------- | ------ | ----------- | ---------- | --------------- |
| 16×512×512 | 700M | 20K HQ | 20k | 2×64 | 35 | [:link:](https://huggingface.co/hpcai-tech/Open-Sora/blob/main/OpenSora-v1-HQ-16x512x512.pth) |
| 16×256×256 | 700M | 20K HQ | 24k | 8×64 | 45 | [:link:](https://huggingface.co/hpcai-tech/Open-Sora/blob/main/OpenSora-v1-HQ-16x256x256.pth) |
diff --git a/PyTorch/built-in/mm/OpenSora1.0/docs/zh_CN/README.md b/PyTorch/built-in/mm/OpenSora1.0/docs/zh_CN/README.md
index 8e52abf16933b7876b409d566feca6869ee809ba..401281c19bda8e9abd02d8ab366ba25478972f52 100644
--- a/PyTorch/built-in/mm/OpenSora1.0/docs/zh_CN/README.md
+++ b/PyTorch/built-in/mm/OpenSora1.0/docs/zh_CN/README.md
@@ -115,7 +115,7 @@ pip install -v .
## 模型权重
-| 分辨率 | 数据 | 迭代次数 | 批量大小 | GPU 天数 (H800) | 网址 |
+| 分辨率 | 数据 | 迭代次数 | 批量大小 | 竞品A 天数 (H800) | 网址 |
| ---------- | ------ | ----------- | ---------- | --------------- | ---------- |
| 16×256×256 | 366K | 80k | 8×64 | 117 | [:link:]() |
| 16×256×256 | 20K HQ | 24k | 8×64 | 45 | [:link:]() |
diff --git a/PyTorch/built-in/mm/OpenSora1.1/docs/zh_CN/README.md b/PyTorch/built-in/mm/OpenSora1.1/docs/zh_CN/README.md
index 21f8c6d79976c3181c9e621370ffffa2bcb61296..cdc5633e4a53eb2aad384717da326bd2e19b3fe8 100644
--- a/PyTorch/built-in/mm/OpenSora1.1/docs/zh_CN/README.md
+++ b/PyTorch/built-in/mm/OpenSora1.1/docs/zh_CN/README.md
@@ -116,7 +116,7 @@ pip install -v .
## 模型权重
-| 分辨率 | 数据 | 迭代次数 | 批量大小 | GPU 天数 (H800) | 网址 |
+| 分辨率 | 数据 | 迭代次数 | 批量大小 | 竞品A 天数 (H800) | 网址 |
| ---------- | ------ | ----------- | ---------- | --------------- | ---------- |
| 16×256×256 | 366K | 80k | 8×64 | 117 | [:link:]() |
| 16×256×256 | 20K HQ | 24k | 8×64 | 45 | [:link:]() |
diff --git a/PyTorch/built-in/mm/OpenSoraPlan1.0/README.md b/PyTorch/built-in/mm/OpenSoraPlan1.0/README.md
index db822195324ac9d8ec6edde912251f22e4729587..f95520bbd37cd4792c730321adaea2aedff1151c 100644
--- a/PyTorch/built-in/mm/OpenSoraPlan1.0/README.md
+++ b/PyTorch/built-in/mm/OpenSoraPlan1.0/README.md
@@ -1,27 +1,37 @@
# OpenSoraPlan1.0 for PyTorch
# 目录
-- [简介](#简介)
- - [模型介绍](#模型介绍)
- - [支持任务列表](#支持任务列表)
- - [代码实现](#代码实现)
-
-- [准备训练环境](#准备训练环境)
-- [VideoGPT](#VideoGPT)
- - [训练数据集准备](#训练数据集准备)
- - [快速开始](#快速开始)
- - [训练任务](#训练任务)
- - [性能展示](#性能展示)
-- [LatteT2V](#LatteT2V)
- - [训练数据集准备](#训练数据集准备)
- - [准备预训练模型](#准备预训练模型)
- - [快速开始](#快速开始)
- - [训练任务](#训练任务)
- - [性能展示](#性能展示)
- - [在线推理任务](#在线推理任务)
-- [公网地址说明](#公网地址说明)
-- [变更说明](#变更说明)
-- [FAQ](#FAQ)
+- [OpenSoraPlan1.0 for PyTorch](#opensoraplan10-for-pytorch)
+- [目录](#目录)
+- [简介](#简介)
+ - [模型介绍](#模型介绍)
+ - [支持任务列表](#支持任务列表)
+ - [代码实现](#代码实现)
+- [准备训练环境](#准备训练环境)
+ - [安装模型环境](#安装模型环境)
+ - [安装昇腾环境](#安装昇腾环境)
+ - [训练数据集准备](#训练数据集准备)
+- [VideoGPT](#videogpt)
+ - [训练数据集准备](#训练数据集准备-1)
+ - [快速开始](#快速开始)
+ - [训练任务](#训练任务)
+ - [开始训练](#开始训练)
+ - [性能展示](#性能展示)
+ - [性能](#性能)
+- [LatteT2V](#lattet2v)
+ - [训练数据集准备](#训练数据集准备-2)
+ - [准备预训练模型](#准备预训练模型)
+ - [快速开始](#快速开始-1)
+ - [训练任务](#训练任务-1)
+ - [开始训练](#开始训练-1)
+ - [性能展示](#性能展示-1)
+ - [性能](#性能-1)
+ - [在线推理任务](#在线推理任务)
+ - [开始推理](#开始推理)
+- [公网地址说明](#公网地址说明)
+- [变更说明](#变更说明)
+ - [变更](#变更)
+- [FAQ](#faq)
# 简介
## 模型介绍
@@ -305,7 +315,7 @@ python dataset/preprocess_msrvtt.py --data_path dataset/msrvtt/train/annotations
| 芯片 | 卡数 | 单步迭代时间(s/step) | batch_size | AMP_Type | Torch_Version |
|:---:|:---:|:----:|:----------:|:---:|:---:|
-| GPU | 8p | 1.84 | 4 | bf16 | 2.1 |
+| 竞品A | 8p | 1.84 | 4 | bf16 | 2.1 |
| Atlas A2 | 8p | 1.95 | 4 | bf16 | 2.1 |
### 在线推理任务
diff --git a/PyTorch/built-in/mm/Qwen-VL/README.md b/PyTorch/built-in/mm/Qwen-VL/README.md
index 2e1f22d92f69a78096262a270d46523404cbb7ff..ddc017e358a51aa06aba2ae8e999ca6e45767134 100644
--- a/PyTorch/built-in/mm/Qwen-VL/README.md
+++ b/PyTorch/built-in/mm/Qwen-VL/README.md
@@ -197,7 +197,7 @@
##### 性能
| 芯片 | 卡数 | model_max_length | batch_size | gradient_accumulation_steps | AMP_Type | Torch_Version | tokens/p/s |
|:---:|:---:|:----:|:----------:|:---:|:---:|:---:|:---:|
-| GPU | 8p | 2048 | 1 | 16 | bf16 | 2.1 | 1796 |
+| 竞品A | 8p | 2048 | 1 | 16 | bf16 | 2.1 | 1796 |
| Atlas A2 | 8p | 2048 | 1 | 16 | bf16 | 2.1 | 1910 |
# 公网地址说明
diff --git a/PyTorch/built-in/nlp/Bert_Chinese_ID3433_for_PyTorch/README.md b/PyTorch/built-in/nlp/Bert_Chinese_ID3433_for_PyTorch/README.md
index a21c3ae3aa1ac5fad989434a0dc350fc3f10d69b..c80299a7250027c62ea5eb9f8d82121ddbb59023 100644
--- a/PyTorch/built-in/nlp/Bert_Chinese_ID3433_for_PyTorch/README.md
+++ b/PyTorch/built-in/nlp/Bert_Chinese_ID3433_for_PyTorch/README.md
@@ -467,16 +467,16 @@ BERT的全称是Bidirectional Encoder Representation from Transformers，即双
```
npu-smi info
- #该设备芯片名为Ascend910A （自行替换）
+ #该设备芯片名为Atlas （自行替换）
回显如下：
+-------------------|-----------------|------------------------------------------------------+
| NPU Name | Health | Power(W) Temp(C) Hugepages-Usage(page) |
| Chip Device | Bus-Id | AICore(%) Memory-Usage(MB) |
+===================+=================+======================================================+
- | 0 910A | OK | 15.8 42 0 / 0 |
+ | 0 Atlas | OK | 15.8 42 0 / 0 |
| 0 0 | 0000:82:00.0 | 0 1074 / 21534 |
+===================+=================+======================================================+
- | 1 910A | OK | 15.4 43 0 / 0 |
+ | 1 Atlas | OK | 15.4 43 0 / 0 |
| 0 1 | 0000:89:00.0 | 0 1070 / 21534 |
+===================+=================+======================================================+
```
diff --git a/PyTorch/contrib/audio/tdnn/README.md b/PyTorch/contrib/audio/tdnn/README.md
index 02836ee4f6c478d3799b6c2e9604f5bfaac331ed..347b3b98d3e552abac1f478b2d0525abd0ab3598 100644
--- a/PyTorch/contrib/audio/tdnn/README.md
+++ b/PyTorch/contrib/audio/tdnn/README.md
@@ -306,16 +306,16 @@ TDNN是一种经典的语音识别网络结构，主要由Conv1D+Relu+BN组成
```
npu-smi info
- #该设备芯片名为Ascend910A （自行替换）
+ #该设备芯片名为Atlas （自行替换）
回显如下：
+-------------------+-----------------+------------------------------------------------------+
| NPU Name | Health | Power(W) Temp(C) Hugepages-Usage(page) |
| Chip Device | Bus-Id | AICore(%) Memory-Usage(MB) |
+===================+=================+======================================================+
- | 0 910A | OK | 15.8 42 0 / 0 |
+ | 0 Atlas | OK | 15.8 42 0 / 0 |
| 0 0 | 0000:82:00.0 | 0 1074 / 32768 |
+===================+=================+======================================================+
- | 1 910A | OK | 15.4 43 0 / 0 |
+ | 1 Atlas | OK | 15.4 43 0 / 0 |
| 0 1 | 0000:89:00.0 | 0 1070 / 32768 |
+===================+=================+======================================================+
```
diff --git a/PyTorch/contrib/cv/classification/HRNet_ID1780_for_PyTorch/README.md b/PyTorch/contrib/cv/classification/HRNet_ID1780_for_PyTorch/README.md
index bb2409d89b44e6a6c59f0dc9749d4d82cd1c534d..5c441b969095703c0049ed2369dd61895fb77cdf 100644
--- a/PyTorch/contrib/cv/classification/HRNet_ID1780_for_PyTorch/README.md
+++ b/PyTorch/contrib/cv/classification/HRNet_ID1780_for_PyTorch/README.md
@@ -322,16 +322,16 @@ HRNet（High-Resolution Net）是针对2D人体姿态估计（Human Pose Estimat
```
npu-smi info
- #该设备芯片名为Ascend 910A （自行替换）
+ #该设备芯片名为Atlas （自行替换）
回显如下：
+-------------------+-----------------+------------------------------------------------------+
| NPU Name | Health | Power(W) Temp(C) Hugepages-Usage(page) |
| Chip Device | Bus-Id | AICore(%) Memory-Usage(MB) |
+===================+=================+======================================================+
- | 0 910A | OK | 15.8 42 0 / 0 |
+ | 0 Atlas | OK | 15.8 42 0 / 0 |
| 0 0 | 0000:82:00.0 | 0 1074 / 21534 |
+===================+=================+======================================================+
- | 1 910A | OK | 15.4 43 0 / 0 |
+ | 1 Atlas | OK | 15.4 43 0 / 0 |
| 0 1 | 0000:89:00.0 | 0 1070 / 21534 |
+===================+=================+======================================================+
```
@@ -418,7 +418,7 @@ HRNet（High-Resolution Net）是针对2D人体姿态估计（Human Pose Estimat
| 芯片型号 | Batch Size | 数据集 | 精度 |
| --------- |------------| ---------- |-----------------------|
-| 910A | 1 | ImageNet | 76.02/Top1 91.72/Top5 |
+| Atlas | 1 | ImageNet | 76.02/Top1 91.72/Top5 |
# 公网地址说明
diff --git a/PyTorch/contrib/cv/classification/InceptionV3_ID1596_for_PyTorch/README.md b/PyTorch/contrib/cv/classification/InceptionV3_ID1596_for_PyTorch/README.md
index 177fc1cf417181a6fba206d2cdbdcb84534c3664..562498e74e4886e5dcf6406a07309e4b4645b212 100644
--- a/PyTorch/contrib/cv/classification/InceptionV3_ID1596_for_PyTorch/README.md
+++ b/PyTorch/contrib/cv/classification/InceptionV3_ID1596_for_PyTorch/README.md
@@ -292,16 +292,16 @@ InceptionV3 模型是谷歌 Inception 系列里面的第三代模型，在 Incep
```bash
npu-smi info
```
- 例如该设备芯片名为 910A，回显如下：
+ 例如该设备芯片名为 Atlas，回显如下：
```
+-------------------+-----------------+------------------------------------------------------+
| NPU Name | Health | Power(W) Temp(C) Hugepages-Usage(page) |
| Chip Device | Bus-Id | AICore(%) Memory-Usage(MB) |
+===================+=================+======================================================+
- | 0 910A | OK | 15.8 42 0 / 0 |
+ | 0 Atlas | OK | 15.8 42 0 / 0 |
| 0 0 | 0000:82:00.0 | 0 1074 / 21534 |
+===================+=================+======================================================+
- | 1 910A | OK | 15.4 43 0 / 0 |
+ | 1 Atlas | OK | 15.4 43 0 / 0 |
| 0 1 | 0000:89:00.0 | 0 1070 / 21534 |
+===================+=================+======================================================+
```
@@ -391,11 +391,11 @@ InceptionV3 模型是谷歌 Inception 系列里面的第三代模型，在 Incep
----
# 性能&精度
-在910A设备上，OM模型的精度为 **{Top1@Acc=77.31%, Top5@Acc=93.46%}**。
+在Atlas设备上，OM模型的精度为 **{Top1@Acc=77.31%, Top5@Acc=93.46%}**。
| 芯片型号 | BatchSize | 数据集 | 精度 |
| --------- | --------- | ----------- | --------------- |
-|Ascend910A| 128 | ILSVRC2012 | Top1Acc=78.06% Top5@Acc=93.81%
+|Atlas| 128 | ILSVRC2012 | Top1Acc=78.06% Top5@Acc=93.81%
# 公网地址说明
diff --git a/PyTorch/contrib/cv/classification/MAE_for_PyTorch/README.md b/PyTorch/contrib/cv/classification/MAE_for_PyTorch/README.md
index 313ec83f73a4fa72d62221ecc5736a046e2b9a3f..948807e0285c72ffb02bcbf210e318a62d8012f9 100644
--- a/PyTorch/contrib/cv/classification/MAE_for_PyTorch/README.md
+++ b/PyTorch/contrib/cv/classification/MAE_for_PyTorch/README.md
@@ -1,10 +1,18 @@
# MAE for PyTorch
-- [概述](#概述)
-- [准备训练环境](#准备训练环境)
-- [开始训练](#开始训练)
-- [训练结果展示](#训练结果展示)
-- [版本说明](#版本说明)
+- [MAE for PyTorch](#mae-for-pytorch)
+- [概述](#概述)
+ - [简述](#简述)
+- [准备训练环境](#准备训练环境)
+ - [准备环境](#准备环境)
+ - [准备数据集](#准备数据集)
+- [开始训练](#开始训练)
+ - [训练模型](#训练模型)
+- [训练结果展示](#训练结果展示)
+- [版本说明](#版本说明)
+ - [变更](#变更)
+ - [FAQ](#faq)
+- [公网地址说明](#公网地址说明)
@@ -116,30 +124,30 @@ MAE的设计虽然简单，但已被证明是一个强大的、可扩展的视
1. 预训练
```bash
# pre-training 1p performance，单p上运行1个epoch，运行时间约为1h
- # 输出性能日志./output_pretrain_1p/910A_1p_pretrain.log、总结性日志./output_pretrain_1p/log.txt
+ # 输出性能日志./output_pretrain_1p/Atlas_1p_pretrain.log、总结性日志./output_pretrain_1p/log.txt
bash ./test/pretrain_performance_1p.sh --data_path=real_data_path
# pre-training 8p performance，8p上运行1个epoch，运行时间约为9min
- # 输出性能日志./output_pretrain_8p/910A_8p_pretrain.log、总结性日志./output_pretrain_8p/log.txt
+ # 输出性能日志./output_pretrain_8p/Atlas_8p_pretrain.log、总结性日志./output_pretrain_8p/log.txt
bash ./test/pretrain_performance_8p.sh --data_path=real_data_path
# pre-training 8p full，8p上运行400个epoch，运行时间约为60h
- # 输出完整预训练日志./output_pretrain_full_8p/910A_8p_pretrain_full.log、总结性日志./output_pretrain_full_8p/log.txt
+ # 输出完整预训练日志./output_pretrain_full_8p/Atlas_8p_pretrain_full.log、总结性日志./output_pretrain_full_8p/log.txt
bash ./test/pretrain_full_8p.sh --data_path=real_data_path
```
2. fine-tuning
```bash
# fine-tuning 1p performance，单p上运行1个epoch，运行时间约为1h15min，
- # 输出性能日志./output_finetune_1p/910A_1p_finetune.log、总结性日志./output_finetune_1p/log.txt
+ # 输出性能日志./output_finetune_1p/Atlas_1p_finetune.log、总结性日志./output_finetune_1p/log.txt
bash ./test/finetune_performance_1p.sh --data_path=real_data_path --finetune_pth=pretrained_model_path
# fine-tuning 8p performance，8p上运行1个epoch，运行时间约为11min
- # 输出性能日志./output_finetune_8p/910A_8p_finetune.log、总结性日志./output_finetune_8p/log.txt
+ # 输出性能日志./output_finetune_8p/Atlas_8p_finetune.log、总结性日志./output_finetune_8p/log.txt
bash ./test/finetune_performance_8p.sh --data_path=real_data_path --finetune_pth=pretrained_model_path
# fine-tuning 8p full，8p上运行100个epoch，运行时间约为18h
- # 输出完整微调日志./output_finetune_full_8p/910A_8p_finetune_full.log、总结性日志./output_finetune_full_8p/log.txt
+ # 输出完整微调日志./output_finetune_full_8p/Atlas_8p_finetune_full.log、总结性日志./output_finetune_full_8p/log.txt
bash ./test/finetune_full_8p.sh --data_path=real_data_path --finetune_pth=pretrained_model_path
# fine-tuning_large 8p performance，8p上运行1个epoch，910运行时间约为14min
@@ -152,7 +160,7 @@ MAE的设计虽然简单，但已被证明是一个强大的、可扩展的视
bash ./test/finetune_full_large_16p.sh --data_path=real_data_path --finetune_pth=pretrained_model_path
# 8p Base_eval，运行时间约为3min
- # 输出eval日志./output_finetune_eval_8p/910A_8p_finetune_eval.log
+ # 输出eval日志./output_finetune_eval_8p/Atlas_8p_finetune_eval.log
bash ./test/finetune_eval_8p.sh --data_path=real_data_path --resume_pth=finetuned_model_path
```
diff --git a/PyTorch/contrib/cv/classification/SE-ResNext-101-32x4d/README.md b/PyTorch/contrib/cv/classification/SE-ResNext-101-32x4d/README.md
index 78b4bf7f6df7316b9e14011e2baf98393641d53d..afef738ce3c0c0436f01d2f194b650da3df71b90 100644
--- a/PyTorch/contrib/cv/classification/SE-ResNext-101-32x4d/README.md
+++ b/PyTorch/contrib/cv/classification/SE-ResNext-101-32x4d/README.md
@@ -52,9 +52,9 @@ test/output/devie_id/Se-ResNext101_bs1024_8p_acc.log # 8p training accuracy re
| Acc@1 | FPS | Platform| Device_nums| Epochs | Type |
| :------: | :------: | :------ | :------ | :------: | :------: |
-| - | 221 | GPU | 1 | 1 | O2 |
+| - | 221 | 竞品A | 1 | 1 | O2 |
| - | 395 | NPU | 1 | 1 | O2 |
-| 78.34 | 1480 | GPU | 8 | 100 | O2 |
+| 78.34 | 1480 | 竞品A | 8 | 100 | O2 |
| 77.75 | 1978 | NPU | 8 | 100 | O2 |
# 公网地址说明
diff --git a/PyTorch/contrib/cv/detection/CTPN/README.md b/PyTorch/contrib/cv/detection/CTPN/README.md
index 8c91c1284ce2f262030930b632176bdf311e7370..f2b523e99eac97833c699a35bf0dd2d8db8adb26 100644
--- a/PyTorch/contrib/cv/detection/CTPN/README.md
+++ b/PyTorch/contrib/cv/detection/CTPN/README.md
@@ -43,6 +43,6 @@ Calculated!{"precision": 0.7331386861313869, "recall": 0.7094063926940639, "hmea
| 名称 | 精度 | 性能 |
| :----: | :--: | :------: |
| NPU-8p | 72.1 | 17.66fps |
-| GPU-8p | 72.4 | 13.25fps |
+| 竞品A-8p | 72.4 | 13.25fps |
| NPU-1p | | 1.695fps |
-| GPU-1p | | 6.295fps|
\ No newline at end of file
+| 竞品A-1p | | 6.295fps|
\ No newline at end of file
diff --git a/PyTorch/contrib/cv/detection/DSFD/README.md b/PyTorch/contrib/cv/detection/DSFD/README.md
index 20fdb5e814b7df17395fbdf4a28b91e53ba496e0..dda16aba3d5502a02a8b7fb53f97680058c7b4ad 100644
--- a/PyTorch/contrib/cv/detection/DSFD/README.md
+++ b/PyTorch/contrib/cv/detection/DSFD/README.md
@@ -100,7 +100,7 @@ Reference:
| | Acc |
| :-------------- | ---------------------------- |
| 参考精度 | E:0.951 M:0.936 H:0.837 |
-| GPU 8P 自测精度 | E 0.9473, M 0.9362, H 0.8651 |
+| 竞品A 8P 自测精度 | E 0.9473, M 0.9362, H 0.8651 |
# Statement
diff --git a/PyTorch/contrib/cv/detection/GCNet/README.md b/PyTorch/contrib/cv/detection/GCNet/README.md
index cedc3bfa0103c97d4de0b49ea1827ca89f8151cb..a4ac91c314cdac3bd7695d9e1a468f1849562b58 100644
--- a/PyTorch/contrib/cv/detection/GCNet/README.md
+++ b/PyTorch/contrib/cv/detection/GCNet/README.md
@@ -129,8 +129,8 @@ bash ./test/eval.sh --weight_path=数据集路径
| 名称 | 精度(mAP) | 性能(fps) |
| ------ | --------- | --------- |
-| GPU-1p | - | 8.47 |
-| GPU-8p | 39.9 | 44.62 |
+| 竞品A-1p | - | 8.47 |
+| 竞品A-8p | 39.9 | 44.62 |
| NPU-1p | - | 0.52 |
| NPU-8p | 39.1 | 2.35 |
diff --git a/PyTorch/contrib/cv/detection/RetinaMask/README.md b/PyTorch/contrib/cv/detection/RetinaMask/README.md
index a42bb5da0196c0decb91556454461c7654dfb15f..43b6c5706fdd7b15bbaaf0b10a7f94888de9133a 100644
--- a/PyTorch/contrib/cv/detection/RetinaMask/README.md
+++ b/PyTorch/contrib/cv/detection/RetinaMask/README.md
@@ -72,8 +72,8 @@ bash test/train_eval_1p.sh --data_path=./dataset/ --weight_path=./model_0044999.
| NAME | Steps | BBOX-MAP | SEGM-MAP | FPS |
| :----: | :----: | :------: | :------: | :--: |
-| GPU-1p | 360000 | - | - | 8.7 |
-| GPU-8p | 20000 | 29.0 | 25.7 | 55.1 |
+| 竞品A-1p | 360000 | - | - | 8.7 |
+| 竞品A-8p | 20000 | 29.0 | 25.7 | 55.1 |
| NPU-1p | 400 | - | - | 4.6 |
| NPU-8p | 20000 | 28.8 | 25.7 | 34.8 |
diff --git a/PyTorch/contrib/cv/others/3D_EDSR_ID3005_for_PyTorch/README.md b/PyTorch/contrib/cv/others/3D_EDSR_ID3005_for_PyTorch/README.md
index 4283d5e49af324567ff1d4321194fab1506ea5a9..291bdba2c73aec6e3b85125ac861828d0601556f 100644
--- a/PyTorch/contrib/cv/others/3D_EDSR_ID3005_for_PyTorch/README.md
+++ b/PyTorch/contrib/cv/others/3D_EDSR_ID3005_for_PyTorch/README.md
@@ -39,8 +39,8 @@ Log path: test/output/devie_id/train_${device_id}.log or obs://cann-idxxx/npu/wo
| | PSNR (dB) | Npu_nums | Epochs | AMP_Type | FPS |single step cost|
| --------- | --------- | -------- | ------ | -------- |-------- |--------|
| NPU | 23.0884 | 1 | 50 | O2 | 1.12 | 0.43 |
-| | PSNR (dB) | Gpu_nums | Epochs | AMP_Type | FPS |single step cost|
-| GPU | 23.0939 | 1 | 50 | O2 | 4.54 | 0.65|
+| | PSNR (dB) | 竞品A_nums | Epochs | AMP_Type | FPS |single step cost|
+| 竞品A | 23.0939 | 1 | 50 | O2 | 4.54 | 0.65|
diff --git a/PyTorch/contrib/cv/others/3D_Nested_Unet/README.md b/PyTorch/contrib/cv/others/3D_Nested_Unet/README.md
index e2c3f46dc1721ab07289faa42eccfbf602f78220..40096261ccf33ffb1ab500ad09013407e61088e4 100644
--- a/PyTorch/contrib/cv/others/3D_Nested_Unet/README.md
+++ b/PyTorch/contrib/cv/others/3D_Nested_Unet/README.md
@@ -26,10 +26,10 @@ test文件夹
其他附件（不在本代码仓中获得）
├── v100_1p.log //GPU 1P训练日志
├── v100_8p.log //GPU 8P训练日志
-├── 910A_1p.log //NPU 1P训练日志
-├── 910A_8p.log //NPU 8P训练日志
+├── Atlas_1p.log //NPU 1P训练日志
+├── Atlas_8p.log //NPU 8P训练日志
├── v100_1p.prof //GPU 1P prof文件
-├── 910A_1p.prof //NPU 1P prof文件
+├── Atlas_1p.prof //NPU 1P prof文件
└── gpu_code.tar //GPU 1P及GPU 8P训练代码
```
**关键环境：**
@@ -310,10 +310,10 @@ python -m torch.distributed.launch --master_port=1234 --nproc_per_node=8 run/run
| :------: | :------: | :------: | :------: |
| [UNET++官方汇报](https://github.com/MrGiovanni/UNetPlusPlus/tree/master/pytorch) | 95.80 | 65.60 | --- |
| 使用作者提供的fold_0预训练权重 | 96.55 | 71.97 | --- |
-| GPU 1P bs=1 | 6.86 | 0.08 | 1.931 |
-| GPU 1P bs=2 | --- | --- | 1.450 |
-| GPU 8P bs=8 | 96.59 | 71.43 | 6.922 |
-| GPU 8P bs=16 | 96.68 | 70.43 | 6.283 |
+| 竞品A 1P bs=1 | 6.86 | 0.08 | 1.931 |
+| 竞品A 1P bs=2 | --- | --- | 1.450 |
+| 竞品A 8P bs=8 | 96.59 | 71.43 | 6.922 |
+| 竞品A 8P bs=16 | 96.68 | 70.43 | 6.283 |
| NPU 1P bs=1 | 6.02 | 0.05 | 2.477 |
| NPU 1P bs=2 | --- | --- | 2.509 |
| NPU 8P bs=8 | 96.67 | 71.42 | 4.209 |
diff --git a/PyTorch/contrib/cv/others/FSRCNN_ID2990_for_PyTorch/README.md b/PyTorch/contrib/cv/others/FSRCNN_ID2990_for_PyTorch/README.md
index 7c601e618435a43c575f59f390e8428eb6244937..247886d01b7bf3b21193096f03b13a255e2ea340 100644
--- a/PyTorch/contrib/cv/others/FSRCNN_ID2990_for_PyTorch/README.md
+++ b/PyTorch/contrib/cv/others/FSRCNN_ID2990_for_PyTorch/README.md
@@ -73,7 +73,7 @@ python3 prepare.py --images-dir "images-dir" \
2、精度指标
- | PSNR | 论文 | GPU | NPU |
+ | PSNR | 论文 | 竞品A | NPU |
| ------ | ----- | -------- | ----- |
| Scale2 | 37.12 | 37.06 | 37.03 |
| Scale3 | 33.22 | 33.61 | 33.56 |
@@ -81,7 +81,7 @@ python3 prepare.py --images-dir "images-dir" \
3、性能指标
- | GPU | NPU |
+ | 竞品A | NPU |
| --------- | --------- |
| 2500 it/s | 3800 it/s |
diff --git a/PyTorch/contrib/cv/others/LPTN_ID2780_for_PyTorch/README.md b/PyTorch/contrib/cv/others/LPTN_ID2780_for_PyTorch/README.md
index a0dfcaaafe8f5134e77e8ec4a6b190bd50fee4ac..4fcb9c53362255a8f107049e80bd5f0e0147bbc1 100644
--- a/PyTorch/contrib/cv/others/LPTN_ID2780_for_PyTorch/README.md
+++ b/PyTorch/contrib/cv/others/LPTN_ID2780_for_PyTorch/README.md
@@ -313,7 +313,7 @@ PYTHONPATH="./:${PYTHONPATH}" python3 scripts/data_preparation/create_lmdb.py
- 精度结果比对
-| 精度指标项 | 论文发布 | GPU实测 | NPU实测 |
+| 精度指标项 | 论文发布 | 竞品A实测 | NPU实测 |
| ---------- | -------- | ------- | ------- |
| PSNR | 22.12 | 22.8 | 22.3 |
| SSIM | 0.878 | 0.885 | 0.8715 |
@@ -321,7 +321,7 @@ PYTHONPATH="./:${PYTHONPATH}" python3 scripts/data_preparation/create_lmdb.py
- 性能结果比对
-| 性能指标项 | GPU实测 | NPU实测 |
+| 性能指标项 | 竞品A实测 | NPU实测 |
| ---------- | ------- | ------- |
|average duration(秒) | 0.06675|0.07552 |
diff --git a/PyTorch/contrib/cv/others/Lifespan_ID2972_for_pytorch/README.md b/PyTorch/contrib/cv/others/Lifespan_ID2972_for_pytorch/README.md
index 82d01faa71fbe3ff8c7d56584b57630b6378f3ec..974701fd70d53c4f7a38317ef607395c8e68f40b 100644
--- a/PyTorch/contrib/cv/others/Lifespan_ID2972_for_pytorch/README.md
+++ b/PyTorch/contrib/cv/others/Lifespan_ID2972_for_pytorch/README.md
@@ -153,7 +153,7 @@ Lifespan Age Transformation Synthesis 是一种基于GAN的方法，用于从单
| NAME | - | FPS | Epochs | sec/epoch | acc |
| ------ | --------------------- | --------- | ------ | -------- | ---- |
| NPU_1p | torch1.5+Ascend910 | 0.001385 | 15 | 2943.1 | None |
-| GPU_1p | torch1.5+V100 | 0.001019 | 15 | 2166.1 | None |
+| 竞品A_1p | torch1.5+V100 | 0.001019 | 15 | 2166.1 | None |
# 公网地址说明
diff --git a/PyTorch/contrib/cv/others/Pix2Pix/README.md b/PyTorch/contrib/cv/others/Pix2Pix/README.md
index eaff186688ee08101d70ab38495e5f989fc3f340..ed6006504f78e534be78df8745fd04b8bb85d0bf 100644
--- a/PyTorch/contrib/cv/others/Pix2Pix/README.md
+++ b/PyTorch/contrib/cv/others/Pix2Pix/README.md
@@ -6,8 +6,8 @@
| 名称 | 精度 | 性能 |
| :------: | :------: | :------: |
- | GPU-1p | - | 15 |
- | GPU-8p | - | 31 |
+ | 竞品A-1p | - | 15 |
+ | 竞品A-8p | - | 31 |
| NPU-1p | - | 8 |
| NPU-8p | - | 8 |
# 自验报告
diff --git a/PyTorch/contrib/cv/others/Pix2PixHD/README.md b/PyTorch/contrib/cv/others/Pix2PixHD/README.md
index 4c83213b082f0657e0cce9514fd2d5de3f148b71..2f8253ea72e6cda36672f3e15b255e04e3d132fa 100644
--- a/PyTorch/contrib/cv/others/Pix2PixHD/README.md
+++ b/PyTorch/contrib/cv/others/Pix2PixHD/README.md
@@ -128,9 +128,9 @@ bash ./test/train_eval_1p.sh --data_path=./datasets
| 名称 | 精度 | 性能 |
| ------ | ----- | -------- |
-| GPU-1p | - | 4.55 fps |
+| 竞品A-1p | - | 4.55 fps |
| NPU-1p | - | 3.76 fps |
-| GPU-8p | - | 19.14 fps |
+| 竞品A-8p | - | 19.14 fps |
| NPU-8p | - | 13.79 fps |
diff --git a/PyTorch/contrib/cv/others/Pytorch-Super-Resolution-Implementations_ID3004_for_Pytorch/README.md b/PyTorch/contrib/cv/others/Pytorch-Super-Resolution-Implementations_ID3004_for_Pytorch/README.md
index fb5c3190007bb3573cdd9791c5505dd6da7ea76e..792570c5b0fe966e7e8021a878f6db577f191555 100644
--- a/PyTorch/contrib/cv/others/Pytorch-Super-Resolution-Implementations_ID3004_for_Pytorch/README.md
+++ b/PyTorch/contrib/cv/others/Pytorch-Super-Resolution-Implementations_ID3004_for_Pytorch/README.md
@@ -233,13 +233,13 @@ https://e-share.obs-website.cn-north-1.myhuaweicloud.com?token=kMZ5rC1dFBGYHI6Yc
```
3. 精度指标。
- | 精度指标项 | 论文发布 | GPU实测 | NPU实测 |
+ | 精度指标项 | 论文发布 | 竞品A实测 | NPU实测 |
| ---------- | -------- | ------- | ------- |
| PSNR | 32.47 | 16.978 | 17.35 |
4. 性能指标。
- | 性能指标项 | 论文发布 | GPU实测 | NPU实测 |
+ | 性能指标项 | 论文发布 | 竞品A实测 | NPU实测 |
| ---------- | -------- | ------- | ------- |
| s/epoch | 无 | 26.136 | 22.506 |
diff --git a/PyTorch/contrib/cv/others/SRGAN/README.md b/PyTorch/contrib/cv/others/SRGAN/README.md
index 0a8125ae6cd9743a1b16793bca72bb9dc4dbf29b..acbf33676e6a1357236644c4996bd7a9bb64c65b 100644
--- a/PyTorch/contrib/cv/others/SRGAN/README.md
+++ b/PyTorch/contrib/cv/others/SRGAN/README.md
@@ -79,8 +79,8 @@ bash ./test/train_full_8p.sh --data_path=../data
| ---------- | ---- | ------ | -------- | ------- | ------ |
| NPU 1p_1.5 | 270 | 100 | O1 | 33.0558 | 0.9226 |
| NPU 8P_1.5 | 1200 | 100 | O1 | 32.1882 | 0.9172 |
-| GPU 1p | 360 | 100 | O1 | 33.4604 | 0.9308 |
-| GPU 8P | 1400 | 100 | O1 | 31.0824 | 0.9191 |
+| 竞品A 1p | 360 | 100 | O1 | 33.4604 | 0.9308 |
+| 竞品A 8P | 1400 | 100 | O1 | 31.0824 | 0.9191 |
| NPU 1p_1.8 | 180 | 100 | O1 | 33.3234 | 0.9302 |
| NPU 8p_1.8 | 1200 | 100 | O1 | 33.2284 | 0.9312 |
diff --git a/PyTorch/contrib/cv/others/Srcnn_x2_for_Pytorch/README.md b/PyTorch/contrib/cv/others/Srcnn_x2_for_Pytorch/README.md
index 6dda360d80d1ef63e2cb37d81787184d5b6d0e5c..91a3dfb2879b5afd817eaa5194327674edb680c3 100644
--- a/PyTorch/contrib/cv/others/Srcnn_x2_for_Pytorch/README.md
+++ b/PyTorch/contrib/cv/others/Srcnn_x2_for_Pytorch/README.md
@@ -49,8 +49,8 @@ Log path:
| 名称 | 精度(PSNR)| 性能(FPS) |
| :---: | :----:| :--------: |
-| GPU-1P | - | 4060.2765 |
-| GPU-8P | - | 12944.0177 |
+| 竞品A-1P | - | 4060.2765 |
+| 竞品A-8P | - | 12944.0177 |
| NPU-1P | - | 5736.3541 |
| NPU-8P | 36.62 | 13953.0316 |
diff --git a/PyTorch/contrib/cv/others/Super-Resolution_CNN_ID3003_for_Pytorch/README.md b/PyTorch/contrib/cv/others/Super-Resolution_CNN_ID3003_for_Pytorch/README.md
index 3e1f16903ae4b86cda3d88e69afb53d544472d48..36fdeac42d6e4e569154cf3bef190d9846b34437 100644
--- a/PyTorch/contrib/cv/others/Super-Resolution_CNN_ID3003_for_Pytorch/README.md
+++ b/PyTorch/contrib/cv/others/Super-Resolution_CNN_ID3003_for_Pytorch/README.md
@@ -228,7 +228,7 @@ pip3 install requirements.txt
3. 精度指标。
```
- | 精度指标项 | 论文发布 | GPU实测 | NPU实测 |
+ | 精度指标项 | 论文发布 | 竞品A实测 | NPU实测 |
| ---------- | -------- | ------- | ------- |
| PSNR | 32.75 | 33.27 | 33.27 |
diff --git a/PyTorch/contrib/cv/pose_estimation/DeepPose/README.md b/PyTorch/contrib/cv/pose_estimation/DeepPose/README.md
index 83f381fc6129fcd55cb18929f0324ed3fbc8fc8e..95500b03fd79b60584c9d67ad84ad5e909e004e1 100644
--- a/PyTorch/contrib/cv/pose_estimation/DeepPose/README.md
+++ b/PyTorch/contrib/cv/pose_estimation/DeepPose/README.md
@@ -51,8 +51,8 @@ python3 pthtar2onnx.py
| 名称 | 精度 | 性能 | AMP_Type |
| :----: | ----- | ------- | -------- |
-| GPU-1p | - | 194 | O2 |
-| GPU-8p | 52.50 | 1160 | O2 |
+| 竞品A-1p | - | 194 | O2 |
+| 竞品A-8p | 52.50 | 1160 | O2 |
| NPU-1p | - | 117 | O2 |
| NPU-8p | 52.65 | 650-830 | O2 |
diff --git a/PyTorch/contrib/cv/pose_estimation/HigherHRNet/README.md b/PyTorch/contrib/cv/pose_estimation/HigherHRNet/README.md
index bc2d5d139a21519bec57d9567cf2b48bf0ea295f..395cf0a04793bc08a02eaf079c7594f00ce9d20c 100644
--- a/PyTorch/contrib/cv/pose_estimation/HigherHRNet/README.md
+++ b/PyTorch/contrib/cv/pose_estimation/HigherHRNet/README.md
@@ -88,9 +88,9 @@ bash test/train_eval_8p.sh --data_path=real_data_path --pth_path=real_pre_train_
| 名称 | 精度 | 性能 |
| :----: | :--: | :------: |
| NPU-8p | 66.9 | 2.2s/step |
-| GPU-8p | 67.1 | 1.2s/step |
+| 竞品A-8p | 67.1 | 1.2s/step |
| NPU-1p | | 1.1s/step |
-| GPU-1p | | 0.7s/step|
+| 竞品A-1p | | 0.7s/step|
# Statement
diff --git a/PyTorch/contrib/cv/pose_estimation/TransPose/README.md b/PyTorch/contrib/cv/pose_estimation/TransPose/README.md
index c15f65eaae85a6d2f992e618e9157511d51b2759..60e26f998e617b23ff315976fed1d09651bc22e1 100644
--- a/PyTorch/contrib/cv/pose_estimation/TransPose/README.md
+++ b/PyTorch/contrib/cv/pose_estimation/TransPose/README.md
@@ -74,8 +74,8 @@ bash test/train_finetune_1p.sh --data_path=real_data_path --pth_path=real_pre_tr
| 名称 | 精度 | 性能 | AMP_Type |
| :----: | :--: | :--: | :------: |
-| GPU-1p | - | 0.34s/step | O1 |
-| GPU-8p | 71.7 | 0.98s/step | O1 |
+| 竞品A-1p | - | 0.34s/step | O1 |
+| 竞品A-8p | 71.7 | 0.98s/step | O1 |
| NPU-1p | - | 0.34s/step | O1 |
| NPU-8p | 72.5 | 0.95s/step | O1 |
diff --git a/PyTorch/contrib/cv/semantic_segmentation/ENet/README.md b/PyTorch/contrib/cv/semantic_segmentation/ENet/README.md
index 1d635087eff45a0aafbf3acf51ce6a70c5e0133d..f5f056b92c2507953ca77520a0be1b67dc5ff8b0 100644
--- a/PyTorch/contrib/cv/semantic_segmentation/ENet/README.md
+++ b/PyTorch/contrib/cv/semantic_segmentation/ENet/README.md
@@ -49,8 +49,8 @@ After running,you can see the results in `./NPU/stargan_full_8p/samples` or `./N
| :----: | :-----: | :----: | :------: |
| NPU-1p | 14.398 | 400 | O2 |
| NPU-8p | 74.310 | 400 | O2 |
-| GPU-1p | 21.885 | 400 | O2 |
-| GPU-8p | 161.495 | 400 | O2 |
+| 竞品A-1p | 21.885 | 400 | O2 |
+| 竞品A-8p | 161.495 | 400 | O2 |
# Statement
diff --git a/PyTorch/contrib/cv/semantic_segmentation/ErfNet/README.md b/PyTorch/contrib/cv/semantic_segmentation/ErfNet/README.md
index d49c2bf9a12965464ae38e62c284cd1e9ef99a8e..b9363f425c39d937e1de87cabf1ee6cea1a6da81 100644
--- a/PyTorch/contrib/cv/semantic_segmentation/ErfNet/README.md
+++ b/PyTorch/contrib/cv/semantic_segmentation/ErfNet/README.md
@@ -88,8 +88,8 @@ python demo.py
| 名称 | iou | fps |
| :------: | :------: | :------: |
-| GPU-1p | - | 14.52 |
-| GPU-8p | - | 94.64 |
+| 竞品A-1p | - | 14.52 |
+| 竞品A-8p | - | 94.64 |
| NPU-1p | - | 24.08 |
| NPU-8p | 71.47 | 143.15 |
diff --git a/PyTorch/contrib/cv/semantic_segmentation/MedSAM_for_PyTorch/README.md b/PyTorch/contrib/cv/semantic_segmentation/MedSAM_for_PyTorch/README.md
index dedf12cb46d7bb884553c74fa7926b3046587825..2aa70ade8325f61c9654da70dea1466398f5310e 100644
--- a/PyTorch/contrib/cv/semantic_segmentation/MedSAM_for_PyTorch/README.md
+++ b/PyTorch/contrib/cv/semantic_segmentation/MedSAM_for_PyTorch/README.md
@@ -1,11 +1,19 @@
# MedSAM for PyTorch
-- [概述](#概述)
-- [准备训练环境](#准备训练环境)
-- [开始训练](#开始训练)
-- [推理评估](#推理评估)
-- [训练结果展示](#训练结果展示)
-- [版本说明](#版本说明)
+- [MedSAM for PyTorch](#medsam-for-pytorch)
+- [概述](#概述)
+ - [简述](#简述)
+- [准备训练环境](#准备训练环境)
+ - [准备环境](#准备环境)
+ - [准备数据集](#准备数据集)
+ - [获取预训练模型](#获取预训练模型)
+- [开始训练](#开始训练)
+ - [训练模型](#训练模型)
+- [训练结果展示](#训练结果展示)
+- [版本说明](#版本说明)
+ - [变更](#变更)
+ - [已知问题](#已知问题)
+- [公网地址说明](#公网地址说明)
@@ -203,8 +211,8 @@
| NAME | DSC | MIOU | FPS | Epochs | AMP_Type|
| :-------: | :-----: | :-----: | :---: | :------: | :-------: |
-| 8p-A100 | 95.9 | 92.4 | 0.76 | 100 | O0 |
-| 8p-昇腾910 | 96.2 | 92.8 | 0.81 | 100 | O0 |
+| 8p-竞品A | 95.9 | 92.4 | 0.76 | 100 | O0 |
+| 8p-NPU | 96.2 | 92.8 | 0.81 | 100 | O0 |
# 版本说明
diff --git a/PyTorch/contrib/cv/semantic_segmentation/SeMask/README.md b/PyTorch/contrib/cv/semantic_segmentation/SeMask/README.md
index 07969c5daf09a8678962cb1802405392a6da09f5..710e7754e462a922e2b2feec74806f99391d8179 100644
--- a/PyTorch/contrib/cv/semantic_segmentation/SeMask/README.md
+++ b/PyTorch/contrib/cv/semantic_segmentation/SeMask/README.md
@@ -1,10 +1,19 @@
# SeMask
-- [概述](#概述)
-- [准备训练环境](#准备训练环境)
-- [开始训练](#开始训练)
-- [训练结果展示](#训练结果展示)
-- [版本说明](#版本说明)
+- [SeMask](#semask)
+- [概述](#概述)
+ - [简述](#简述)
+- [准备训练环境](#准备训练环境)
+ - [准备环境](#准备环境)
+ - [准备数据集](#准备数据集)
+ - [获取预训练模型](#获取预训练模型)
+- [开始训练](#开始训练)
+ - [训练模型](#训练模型)
+- [训练结果展示](#训练结果展示)
+- [版本说明](#版本说明)
+ - [变更](#变更)
+ - [已知问题](#已知问题)
+- [公网地址说明](#公网地址说明)
# 概述
@@ -159,8 +168,8 @@ SeMask是一个图像语义分割框架，通过以下两种技术将语义信
| NAME | Acc@1 | FPS | PyTorch_version |
|--------|-------|-------|-----------------|
- | GPU-1P | - | 9.521 | 1.5 |
- | GPU-8P | 79.02 | 63.15 | 1.5 |
+ | 竞品A-1P | - | 9.521 | 1.5 |
+ | 竞品A-8P | 79.02 | 63.15 | 1.5 |
| NPU-1P | - | 8.153 | 1.5 |
| NPU-8P | 79.02 | 62.126 | 1.5 |
diff --git a/PyTorch/contrib/nlp/MAG-Bert_ID2985_for_PyTorch/README.md b/PyTorch/contrib/nlp/MAG-Bert_ID2985_for_PyTorch/README.md
index 66dde338581ea8e60dd79889f1527399699fa674..2be50c5bf8c047ef33cb0e0ac099615bebc4ff65 100644
--- a/PyTorch/contrib/nlp/MAG-Bert_ID2985_for_PyTorch/README.md
+++ b/PyTorch/contrib/nlp/MAG-Bert_ID2985_for_PyTorch/README.md
@@ -253,7 +253,7 @@ pip3 install requirements.txt
3. 精度指标。
```
- | 精度指标项 | 论文发布 | GPU实测 | NPU实测 |
+ | 精度指标项 | 论文发布 | 竞品A实测 | NPU实测 |
| ---------- | -------- | ------- | ------- |
| F1_SCORE | 84.1 | 84.27 | 82.59 |
diff --git a/PyTorch/contrib/nlp/NCF_ID2943_for_PyTorch/readme.md b/PyTorch/contrib/nlp/NCF_ID2943_for_PyTorch/readme.md
index 061761259b029c79b6343e08adf7eca5089fcc25..2360539d49fee883dec1538be20ff7eab7f923ef 100644
--- a/PyTorch/contrib/nlp/NCF_ID2943_for_PyTorch/readme.md
+++ b/PyTorch/contrib/nlp/NCF_ID2943_for_PyTorch/readme.md
@@ -35,7 +35,7 @@ python3 train.py
| 200Epoch | HR | NDCG |
| -------- | ------ | ------ |
-| GPU | 0.6400 | 0.2950 |
+| 竞品A | 0.6400 | 0.2950 |
| NPU | 0.6407 | 0.3696 |
精度达标
@@ -43,7 +43,7 @@ python3 train.py
| 设备 | 单batch耗时 |
| :--- | ----------- |
-| GPU | 0.0179s |
+| 竞品A | 0.0179s |
| NPU | 0.0183s |
性能达标
diff --git a/PyTorch/contrib/nlp/albert_ID0335_for_PyTorch/README.md b/PyTorch/contrib/nlp/albert_ID0335_for_PyTorch/README.md
index 9eb3d72a6eddab87e4d918ef379640a30c89c45d..6df80ff63496240dd18efea8f3170d319e43992e 100644
--- a/PyTorch/contrib/nlp/albert_ID0335_for_PyTorch/README.md
+++ b/PyTorch/contrib/nlp/albert_ID0335_for_PyTorch/README.md
@@ -363,16 +363,16 @@ https://gitee.com/ascend/ModelZoo-PyTorch/tree/master/ACL_PyTorch/contrib/nlp/al
```
npu-smi info
- #该设备芯片名为Ascend910A （自行替换）
+ #该设备芯片名为Atlas （自行替换）
回显如下：
+-------------------|-----------------|------------------------------------------------------+
| NPU Name | Health | Power(W) Temp(C) Hugepages-Usage(page) |
| Chip Device | Bus-Id | AICore(%) Memory-Usage(MB) |
+===================+=================+======================================================+
- | 0 910A | OK | 15.8 42 0 / 0 |
+ | 0 Atlas | OK | 15.8 42 0 / 0 |
| 0 0 | 0000:82:00.0 | 0 1074 / 21534 |
+===================+=================+======================================================+
- | 1 910A | OK | 15.4 43 0 / 0 |
+ | 1 Atlas | OK | 15.4 43 0 / 0 |
| 0 1 | 0000:89:00.0 | 0 1070 / 21534 |
+===================+=================+======================================================+
```
diff --git a/PyTorch/contrib/others/Low-rank-Multimodal-Fusion_ID2983_for_Pytorch/README.md b/PyTorch/contrib/others/Low-rank-Multimodal-Fusion_ID2983_for_Pytorch/README.md
index 81a4e6d9a375072c5cad6c99317dbaf59d242f2e..11eba9fba7595533ced1592c1504ff75b2e6543f 100644
--- a/PyTorch/contrib/others/Low-rank-Multimodal-Fusion_ID2983_for_Pytorch/README.md
+++ b/PyTorch/contrib/others/Low-rank-Multimodal-Fusion_ID2983_for_Pytorch/README.md
@@ -271,7 +271,7 @@ pip3 install requirements.txt
- 精度结果比对
-| 精度指标项 | 论文发布 | GPU实测 | NPU实测 |
+| 精度指标项 | 论文发布 | 竞品A实测 | NPU实测 |
| ---------- | -------- | ------- | ------- |
| F1_angry | 89.0 | 87.86 | 89.38 |
| F1_sad | 85.9 | 84.37 | 84.06 |
diff --git a/PyTorch/dev/nlp/Textcnn_for_PyTorch/README.md b/PyTorch/dev/nlp/Textcnn_for_PyTorch/README.md
index f534634de12fa4451e3f926dd8d9750f17bc24d2..d9073e98a34d9f74d9cc2331af8c698ec9094fd0 100644
--- a/PyTorch/dev/nlp/Textcnn_for_PyTorch/README.md
+++ b/PyTorch/dev/nlp/Textcnn_for_PyTorch/README.md
@@ -86,7 +86,7 @@ Iter: 700, Train Loss: 0.21, Train Acc: 93.75%, Val Loss: 0.26, Val Acc:
#### 5、GPU/NPU loss收敛趋势：
-| step | GPU loss | NPU loss |
+| step | 竞品A loss | NPU loss |
| :--- | -------- | :------- |
| | | |
| | | |