[Samples] add lenet and mobilenetv2 example; update corresponding readme instruction

README.md

@@ -82,12 +82,10 @@ $ scalehls-opt resnet18.tmp -print-op-graph 2> resnet18.gv
 $ dot -Tpng resnet18.gv > resnet18.png
 
 $ # Legalize the output of ONNX-MLIR, optimize and emit C++ code.
-$ scalehls-opt resnet18.mlir -legalize-onnx -affine-loop-normalize \
-    -legalize-dataflow -split-function -convert-linalg-to-affine-loops \
-    -affine-loop-perfection -affine-loop-normalize \
+$ scalehls-opt resnet18.mlir -legalize-onnx -affine-loop-normalize -canonicalize \
+    -legalize-dataflow="min-gran=2 insert-copy=false" -split-function \
+    -convert-linalg-to-affine-loops -affine-loop-fusion \
     -convert-to-hlscpp="top-function=main_graph" \
-    -store-op-forward -simplify-memref-access -cse -canonicalize \
-    -qor-estimation="target-spec=../../../config/target-spec.ini" \
     | scalehls-translate -emit-hlscpp
 ```
 

samples/onnx-mlir/lenet/export_lenet.py

@@ -0,0 +1,34 @@
+'''LeNet in PyTorch.
+
+Modified based on (https://github.com/kuangliu/pytorch-cifar/blob/master/models/lenet.py)
+
+'''
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+
+class LeNet(nn.Module):
+    def __init__(self):
+        super(LeNet, self).__init__()
+        self.conv1 = nn.Conv2d(3, 6, 5)
+        self.conv2 = nn.Conv2d(6, 16, 5)
+        self.fc1 = nn.Linear(16*5*5, 120)
+        self.fc2 = nn.Linear(120, 84)
+        self.fc3 = nn.Linear(84, 10)
+
+    def forward(self, x):
+        out = F.relu(self.conv1(x))
+        out = F.max_pool2d(out, 2)
+        out = F.relu(self.conv2(out))
+        out = F.max_pool2d(out, 2)
+        out = torch.flatten(out, 1)  # out.view(out.size(0), -1)
+        out = F.relu(self.fc1(out))
+        out = F.relu(self.fc2(out))
+        out = self.fc3(out)
+        return out
+
+
+input_random = torch.randn((1, 3, 32, 32))
+torch.onnx.export(LeNet(), input_random, 'lenet.onnx', opset_version=7)

samples/onnx-mlir/mnist/export_mnist.py

@@ -1,5 +0,0 @@
-import torch
-import torch.nn as nn
-import torch.nn.functional as F
-import torchvision.models as models
-import numpy as np

samples/onnx-mlir/mobilenetv2/export_mobilenetv2.py

@@ -0,0 +1,85 @@
+'''MobileNetV2 in PyTorch.
+
+Modified based on (https://github.com/kuangliu/pytorch-cifar/blob/master/models/mobilenetv2.py)
+
+'''
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+
+class Block(nn.Module):
+    '''expand + depthwise + pointwise'''
+
+    def __init__(self, in_planes, out_planes, expansion, stride):
+        super(Block, self).__init__()
+        self.stride = stride
+
+        planes = expansion * in_planes
+        self.conv1 = nn.Conv2d(
+            in_planes, planes, kernel_size=1, stride=1, padding=0, bias=False)
+        self.conv2 = nn.Conv2d(planes, planes, kernel_size=3,
+                               stride=stride, padding=1, groups=planes, bias=False)
+        self.conv3 = nn.Conv2d(
+            planes, out_planes, kernel_size=1, stride=1, padding=0, bias=False)
+
+        self.shortcut = nn.Sequential()
+        if stride == 1 and in_planes != out_planes:
+            self.shortcut = nn.Sequential(
+                nn.Conv2d(in_planes, out_planes, kernel_size=1,
+                          stride=1, padding=0, bias=False)
+            )
+
+    def forward(self, x):
+        out = F.relu(self.conv1(x))
+        out = F.relu(self.conv2(out))
+        out = self.conv3(out)
+        out = out + self.shortcut(x) if self.stride == 1 else out
+        return out
+
+
+class MobileNetV2(nn.Module):
+    # (expansion, out_planes, num_blocks, stride)
+    cfg = [(1,  16, 1, 1),
+           (6,  24, 2, 1),  # NOTE: change stride 2 -> 1 for CIFAR10
+           (6,  32, 3, 2),
+           (6,  64, 4, 2),
+           (6,  96, 3, 1),
+           (6, 160, 3, 2),
+           (6, 320, 1, 1)]
+
+    def __init__(self, num_classes=10):
+        super(MobileNetV2, self).__init__()
+        # NOTE: change conv1 stride 2 -> 1 for CIFAR10
+        self.conv1 = nn.Conv2d(3, 32, kernel_size=3,
+                               stride=1, padding=1, bias=False)
+        self.layers = self._make_layers(in_planes=32)
+        self.conv2 = nn.Conv2d(320, 1280, kernel_size=1,
+                               stride=1, padding=0, bias=False)
+        self.avgpool = nn.AdaptiveAvgPool2d((1, 1))
+        self.linear = nn.Linear(1280, num_classes)
+
+    def _make_layers(self, in_planes):
+        layers = []
+        for expansion, out_planes, num_blocks, stride in self.cfg:
+            strides = [stride] + [1]*(num_blocks-1)
+            for stride in strides:
+                layers.append(Block(in_planes, out_planes, expansion, stride))
+                in_planes = out_planes
+        return nn.Sequential(*layers)
+
+    def forward(self, x):
+        out = F.relu(self.conv1(x))
+        out = self.layers(out)
+        out = F.relu(self.conv2(out))
+        # NOTE: change pooling kernel_size 7 -> 4 for CIFAR10
+        out = self.avgpool(out)
+        out = torch.flatten(out, 1)  # out.view(out.size(0), -1)
+        out = self.linear(out)
+        return out
+
+
+input_random = torch.randn((1, 3, 32, 32))
+torch.onnx.export(MobileNetV2(), input_random,
+                  'mobilenetv2.onnx', opset_version=7)

samples/onnx-mlir/resnet18/export_resnet18.py

@@ -1,8 +1,12 @@
+'''ResNet in PyTorch.
+
+Modified based on (https://github.com/kuangliu/pytorch-cifar/blob/master/models/resnet.py)
+
+'''
+
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
-import torchvision.models as models
-import numpy as np
 
 
 class BasicBlock(nn.Module):