Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Tutorial][Executor] Fix the usage of executors in tutorials #8586

Merged
merged 7 commits into from
Aug 28, 2021
Merged

[Tutorial][Executor] Fix the usage of executors in tutorials #8586

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
57216cb
fix: executor usage for keras tutorial
ganler Jul 29, 2021
3a2bf81
fix: executor usage for onnx tutorial
ganler Jul 29, 2021
2efd3c6
[Tutorial][Executor] Fix executors in tutorials
junrushao Aug 8, 2021
f3d5d00
pull
ganler Aug 10, 2021
84ca7fc
Merge branch 'main' into fix-doc
ganler Aug 10, 2021
d7aadee
Merge branch 'main' into fix-doc
ganler Aug 12, 2021
9ff38ca
merge main to re-trigger ci
ganler Aug 26, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion tutorials/dev/bring_your_own_datatypes.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -257,8 +257,9 @@ def get_cat_image():
######################################################################
# It's easy to execute MobileNet with native TVM:

ex = tvm.relay.create_executor("graph", mod=module, params=params)
input = get_cat_image()
result = tvm.relay.create_executor("graph", mod=module).evaluate()(input, **params).numpy()
result = ex.evaluate()(input).numpy()
# print first 10 elements
print(result.flatten()[:10])

Expand Down
4 changes: 2 additions & 2 deletions tutorials/frontend/from_keras.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -103,14 +103,14 @@
# due to a latent bug. Note that the pass context only has an effect within
# evaluate() and is not captured by create_executor().
with tvm.transform.PassContext(opt_level=0):
model = relay.build_module.create_executor("graph", mod, dev, target).evaluate()
model = relay.build_module.create_executor("graph", mod, dev, target, params).evaluate()


######################################################################
# Execute on TVM
# ---------------
dtype = "float32"
tvm_out = model(tvm.nd.array(data.astype(dtype)), **params)
tvm_out = model(tvm.nd.array(data.astype(dtype)))
top1_tvm = np.argmax(tvm_out.numpy()[0])

#####################################################################
Expand Down
6 changes: 4 additions & 2 deletions tutorials/frontend/from_onnx.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -92,13 +92,15 @@
mod, params = relay.frontend.from_onnx(onnx_model, shape_dict)

with tvm.transform.PassContext(opt_level=1):
compiled = relay.build_module.create_executor("graph", mod, tvm.cpu(0), target).evaluate()
executor = relay.build_module.create_executor(
"graph", mod, tvm.cpu(0), target, params
).evaluate()

######################################################################
# Execute on TVM
# ---------------------------------------------
dtype = "float32"
tvm_output = compiled(tvm.nd.array(x.astype(dtype)), **params).numpy()
tvm_output = executor(tvm.nd.array(x.astype(dtype))).numpy()

######################################################################
# Display results
Expand Down