def main():
test_status = "PASS!"
# CHECK-LABEL: test_op_report_vgg_style_lenet
# CHECK: PASS!
print("test_op_report_vgg_style_lenet")
device = torch_mlir.mlir_device()
model = Net().to(device)
ref_tensor = torch.randn((8, 1, 30, 30))
tensor = ref_tensor.clone().to(device)
result = model(tensor)
target = torch.ones((8), dtype=torch.long).to(device)
loss = F.nll_loss(result, target)
loss.backward()
mlir0 = torch_mlir.get_mlir(model.conv1.weight.grad)
print(mlir0)
report = torch_mlir.op_report(mlir0)
print(report)
report_dict = report
expected = 32
if (len(report_dict) != expected):
print("### ERROR: Expecting", expected,
"items in the report, but got ", len(report_dict))
test_status = "FAIL!"
# Every item should have a read and a write
for key, value in report_dict.items():
if not 'reads' in value:
print(
f"### ERROR: {key} does not contain the required reads field")
test_status = "FAIL!"
if not 'writes' in value:
print(
f"### ERROR: {key} does not contain the required writes field")
test_status = "FAIL!"
if "convolution" in key:
if not 'ops:MAC' in value:
print(
f"### ERROR: convolution {key} does not contain the required MAC field"
)
test_status = "FAIL!"
if "mm" in key:
if not 'ops:MAC' in value:
print(
f"### ERROR: mm {key} does not contain the required MAC field"
)
test_status = "FAIL!"
print(test_status)
# -*- Python -*-
# This file is licensed under a pytorch-style license
# See frontends/pytorch/LICENSE for license information.
import torch
import npcomp.frontends.pytorch as torch_mlir
# RUN: %PYTHON %s | FileCheck %s
dev = torch_mlir.mlir_device()
t0 = torch.randn((1, 2, 3, 4), device=dev)
t1 = torch.randn((1, 2, 3, 4), device=dev)
t2 = torch.randn((1, 2, 3, 4), device=dev)
t3 = t0 + t1 + t2
#
# Generate and check the MLIR for the result tensor
#
t3_mlir = torch_mlir.get_mlir(t3)
# CHECK-LABEL: test_export_add3
# CHECK: %1 = "aten.add"(%arg0, %arg1, %0) {layer_name = "L0-add-0"} : (tensor<1x2x3x4xf32>, tensor<1x2x3x4xf32>, i32) -> tensor<1x2x3x4xf32>
# CHECK: %2 = "aten.add"(%1, %arg2, %0) {layer_name = "L1-add-1"} : (tensor<1x2x3x4xf32>, tensor<1x2x3x4xf32>, i32) -> tensor<1x2x3x4xf32>
print("test_export_add3")
print(t3_mlir)
def _test_model(self, model, model_args):
result = model(model_args)
mlir = torch_mlir.get_mlir(result)
printWithCurrentFunctionName(mlir)
return True
# -*- Python -*-
# This file is licensed under a pytorch-style license
# See frontends/pytorch/LICENSE for license information.
import torch
import npcomp.frontends.pytorch as torch_mlir
import torchvision.models as models
# RUN: python %s | FileCheck %s
dev = torch_mlir.mlir_device()
model = models.vgg11_bn().to(dev)
model.training = False
result = model(torch.ones(32,3,32,32).to(dev))
mlir = torch_mlir.get_mlir( result )
# for now we just check the output shape
# CHECK-LABEL: test_export_vgg11
# CHECK: return %{{.*}} : tensor<32x1000xf32>
print("test_export_vgg11")
print(mlir)
# -*- Python -*-
# This file is licensed under a pytorch-style license
# See frontends/pytorch/LICENSE for license information.
import torch
import npcomp.frontends.pytorch as torch_mlir
# RUN: python %s | FileCheck %s
dev = torch_mlir.mlir_device()
t0 = torch.randn(4, device=dev)
t1 = torch.randn(4, device=dev)
t2 = torch.randn(4, device=dev)
t4 = t0 + t1 + t2
t5 = t4 + t1
t6 = t5 + t4
# CHECK-LABEL: test_multi_out
# CHECK: return %2, %3, %4 : tensor<4xf32>, tensor<4xf32>, tensor<4xf32>
mlir = torch_mlir.get_mlir([t4, t5, t6])
print("test_multi_out")
print(mlir)
ref_model.weight.data = model.weight.clone()
ref_model.bias.data = model.bias.clone()
model = model.to(dev)
softmax = torch.nn.LogSoftmax(dim=1)
loss = torch.nn.NLLLoss()
tensor = torch.randn(N, Cin, h, w, device=dev)
result = model(tensor)
# CHECK-LABEL: test_export_conv2d
# CHECK: aten.convolution_overrideable
print("test_export_conv2d")
print(torch_mlir.get_mlir(result))
target = torch.empty(N, 8, 8, dtype=torch.long).random_(0, Cout)
ref_target = target.clone()
target = target.to(dev)
test_loss = loss(softmax(result), target)
test_loss.backward()
# CHECK-LABEL: test_export_conv2d_back
# CHECK: aten.convolution_overrideable
# CHECK: aten._log_softmax
# CHECK: aten.nll_loss2d_forward
print("test_export_conv2d_back")
print(torch_mlir.get_mlir(test_loss))