Update README.md
Browse files
README.md
CHANGED
@@ -13,7 +13,7 @@ tags:
|
|
13 |
|
14 |
# Model Card for Model ID
|
15 |
|
16 |
-
Based on a class of partial differential equations called **quasi-linear hyperbolic systems** [[Liu et al, 2023](https://github.com/liuyao12/ConvNets-PDE-perspective)], the QLNet breaks into uncharted waters of ConvNet model space marked by the use of (element-wise) multiplication in lieu of ReLU as the primary nonlinearity. It achieves comparable performance as ResNet50 on ImageNet-1k (acc=**78.61**), demonstrating that it has the same level of capacity/expressivity, and deserves more analysis and study (hyper-paremeter tuning, optimizer, etc.) by the academic community.
|
17 |
|
18 |
The overall architecture folllows that of the origianl ConvNet (LeCun) and ResNet (He et al.), with the use of "depthwise" as in MobileNet.
|
19 |
|
@@ -94,7 +94,7 @@ ImageNet-1k
|
|
94 |
|
95 |
### Training Procedure
|
96 |
|
97 |
-
We
|
98 |
|
99 |
```
|
100 |
python3 train.py ../datasets/imagenet/ --model resnet50 --num-classes 1000 --lr 0.1 --warmup-epochs 5 --epochs 240 --weight-decay 1e-4 --sched cosine --reprob 0.4 --recount 3 --remode pixel --aa rand-m7-mstd0.5-inc1 -b 192 -j 6 --amp --dist-bn reduce
|
|
|
13 |
|
14 |
# Model Card for Model ID
|
15 |
|
16 |
+
Based on a class of partial differential equations called **quasi-linear hyperbolic systems** [[Liu et al, 2023](https://github.com/liuyao12/ConvNets-PDE-perspective)], the **QLNet** breaks into uncharted waters of ConvNet model space marked by the use of (element-wise) multiplication in lieu of ReLU as the primary nonlinearity. It achieves comparable performance as ResNet50 on ImageNet-1k (acc=**78.61**), demonstrating that it has the same level of capacity/expressivity, and deserves more analysis and study (hyper-paremeter tuning, optimizer, etc.) by the academic community.
|
17 |
|
18 |
The overall architecture folllows that of the origianl ConvNet (LeCun) and ResNet (He et al.), with the use of "depthwise" as in MobileNet.
|
19 |
|
|
|
94 |
|
95 |
### Training Procedure
|
96 |
|
97 |
+
We used the following training script in `timm`
|
98 |
|
99 |
```
|
100 |
python3 train.py ../datasets/imagenet/ --model resnet50 --num-classes 1000 --lr 0.1 --warmup-epochs 5 --epochs 240 --weight-decay 1e-4 --sched cosine --reprob 0.4 --recount 3 --remode pixel --aa rand-m7-mstd0.5-inc1 -b 192 -j 6 --amp --dist-bn reduce
|