Skip to content

Commit

Permalink
fix update (#437)
Browse files Browse the repository at this point in the history
  • Loading branch information
@shh2000
shh2000 authored Feb 1, 2024
1 parent 06abe50 commit 1382c36
Show file tree
Hide file tree
Showing 4 changed files with 36 additions and 1 deletion.
34 changes: 34 additions & 0 deletions README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -63,6 +63,7 @@
<th>昆仑芯</th>
<th>天数智芯</th>
<th>摩尔线程</th>
<th>沐曦</th>
</tr>
</thead>
<tbody>
Expand All @@ -74,6 +75,7 @@
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>2</td>
Expand All @@ -83,6 +85,7 @@
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>3</td>
Expand All @@ -92,6 +95,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/bert-paddle">paddle</a>, <a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/bert-pytorch">pytorch</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/iluvatar/bert-paddle">paddle</a>, <a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/iluvatar/bert-pytorch">pytorch</a></td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>4</td>
Expand All @@ -101,6 +105,7 @@
<td>N/A</td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/iluvatar/bert_hf-pytorch">pytorch</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/mthreads/bert_hf-pytorch">pytorch</a></td>
<td>N/A</td>
</tr>
<tr>
<td>5</td>
Expand All @@ -110,6 +115,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/bigtransfer-pytorch">pytorch</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/iluvatar/bigtransfer-pytorch">pytorch</a></td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>6</td>
Expand All @@ -119,6 +125,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/cpm-pytorch">pytorch</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/iluvatar/cpm-pytorch">pytorch</a></td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>7</td>
Expand All @@ -128,6 +135,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/detr-pytorch">pytorch</a></td>
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>8</td>
Expand All @@ -137,6 +145,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/distilbert-pytorch">pytorch</a></td>
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>9</td>
Expand All @@ -146,6 +155,7 @@
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>10</td>
Expand All @@ -155,6 +165,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/efficientnet-pytorch">pytorch</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/iluvatar/efficientnet-pytorch">pytorch</a></td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>11</td>
Expand All @@ -164,6 +175,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/faster_rcnn-pytorch">pytorch</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/iluvatar/faster_rcnn-pytorch">pytorch</a></td>
<td>N/A</td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/metax/faster_rcnn-pytorch">pytorch</a></td>
</tr>
<tr>
<td>12</td>
Expand All @@ -173,6 +185,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/glm-pytorch">pytorch</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/iluvatar/glm-pytorch">pytorch</a></td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>13</td>
Expand All @@ -182,6 +195,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/gpt2-pytorch">pytorch</a></td>
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>14</td>
Expand All @@ -191,6 +205,7 @@
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>15</td>
Expand All @@ -200,6 +215,7 @@
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>16</td>
Expand All @@ -209,6 +225,7 @@
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>17</td>
Expand All @@ -218,6 +235,7 @@
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>18</td>
Expand All @@ -227,6 +245,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/pull/348">deepspeed</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/pull/343">deepspeed</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/pull/354">deepspeed</a></td>
<td>N/A</td>
</tr>
<tr>
<td>19</td>
Expand All @@ -236,6 +255,7 @@
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>20</td>
Expand All @@ -245,6 +265,7 @@
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>21</td>
Expand All @@ -254,6 +275,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/longformer-pytorch">pytorch</a></td>
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>22</td>
Expand All @@ -263,6 +285,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/mask_rcnn-pytorch">pytorch</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/iluvatar/mask_rcnn-pytorch">pytorch</a></td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>23</td>
Expand All @@ -272,6 +295,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/mobilenetv2-pytorch">pytorch</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/iluvatar/mobilenetv2-pytorch">pytorch</a></td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>24</td>
Expand All @@ -281,6 +305,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/resnet50-pytorch">pytorch</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/iluvatar/resnet50-pytorch">pytorch</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/mthreads/resnet50-pytorch">pytorch</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/metax/resnet50-pytorch">pytorch</a></td>
</tr>
<tr>
<td>25</td>
Expand All @@ -290,6 +315,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/retinanet-pytorch">pytorch</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/iluvatar/retinanet-pytorch">pytorch</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/mthreads/retinanet-pytorch">pytorch</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/metax/retinanet-pytorch">pytorch</a></td>
</tr>
<tr>
<td>26</td>
Expand All @@ -299,6 +325,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/swin_transformer-pytorch">pytorch</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/iluvatar/swin_transformer-pytorch">pytorch</a></td>
<td>N/A</td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/metax/swin_transformer-pytorch">pytorch</a></td>
</tr>
<tr>
<td>27</td>
Expand All @@ -308,6 +335,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/t5_small-pytorch">pytorch</a></td>
<td>N/A</td>
<td>N/A</td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/metax/t5_small-pytorch">pytorch</a></td>
</tr>
<tr>
<td>28</td>
Expand All @@ -317,6 +345,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/tacotron2-pytorch">pytorch</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/iluvatar/tacotron2-pytorch">pytorch</a></td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>29</td>
Expand All @@ -326,6 +355,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/transformer-pytorch">pytorch</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/iluvatar/transformer-pytorch">pytorch</a></td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>30</td>
Expand All @@ -335,6 +365,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/transformer_xl-pytorch">pytorch</a></td>
<td>N/A</td>
<td>N/A</td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/metax/transformer_xl-pytorch">pytorch</a></td>
</tr>
<tr>
<td>31</td>
Expand All @@ -344,6 +375,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/vit-pytorch">pytorch</a></td>
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/iluvatar/vit-pytorch">pytorch</a></td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>32</td>
Expand All @@ -353,6 +385,7 @@
<td><a href="https://github.com/FlagOpen/FlagPerf/tree/main/training/kunlunxin/wav2vec2-pytorch">pytorch</a></td>
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
</tr>
<tr>
<td>33</td>
Expand All @@ -362,6 +395,7 @@
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
<td>N/A</td>
</tr>
</tbody>
</table>
Expand Down
Binary file modified assets/imgs/coop.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
2 changes: 1 addition & 1 deletion training/nvidia/detr-pytorch/README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -42,4 +42,4 @@
| ------------------- | --------- | -------------- | -------- | ------- | ------- | ------ | ----- | --------- |
| A100单机8卡（1x8） | fp32 | bs=8,lr=0.0001 | 220047 | 161.32 | 172.99 | 208.68 | 0.405 | 37.2/40.0 |
| A100单机单卡（1x1） | fp32 | bs=8,lr=0.0001 | / | 27.98 | 29.25 | 33.71 | / | 38.2/40.0 |
| A100两机8卡（2x8） | fp32 | | / | | | | / | |
| A100两机16卡（2x8） | fp32 | / | / | 309.5 | 332.7 | 378.1 | / | 39.6/40.0 |
1 change: 1 addition & 0 deletions training/nvidia/gpt2-pytorch/README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -40,3 +40,4 @@
| ------------------- | --------- | ---------------- | -------- | ------- | ------- | ------ | ----------- | --------- |
| A100单机单卡（1x1） | fp32 | / | / | 9.6 | 14.2 | 14.2 | / | 30.2/40.0 |
| A100单机8卡（1x8） | fp32 | / | 10352.7 | 68.6 | 97.5 | 97.6 | 0.60 | 31.3/40.0 |
| A100两机16卡（2x8） | fp32 | / | / | 116 | 156.5 | 156.6 | / | 29.9/40.0 |

0 comments on commit 1382c36

Please sign in to comment.