You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by GitBox <gi...@apache.org> on 2022/02/14 15:56:18 UTC
[GitHub] [tvm] jacobbohlin opened a new pull request #10242: [microNPU] Add MobileNetv2 test to network tests
jacobbohlin opened a new pull request #10242:
URL: https://github.com/apache/tvm/pull/10242
This adds MobileNetv2 tflite testcase to the microNPU's network tests.
Co-authored-by: Jacob Bohlin <ja...@arm.com>
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] lhutton1 commented on a change in pull request #10242: [microNPU] Add MobileNetv2 test to network tests
Posted by GitBox <gi...@apache.org>.
lhutton1 commented on a change in pull request #10242:
URL: https://github.com/apache/tvm/pull/10242#discussion_r806313919
##########
File path: tests/python/contrib/test_ethosu/test_networks.py
##########
@@ -65,5 +65,31 @@ def test_forward_mobilenet_v1(accel_type):
infra.verify_source(compiled_models, accel_type)
+def test_forward_mobilenet_v2(accel_type="ethos-u55-256"):
+ """Test the Mobilenet V2 TF Lite model."""
+ np.random.seed(23)
+ tflite_model_file = tf_testing.get_workload_official(
+ "https://storage.googleapis.com/download.tensorflow.org/models/"
+ "tflite_11_05_08/mobilenet_v2_1.0_224_quant.tgz",
+ "mobilenet_v2_1.0_224_quant.tflite",
+ )
+ with open(tflite_model_file, "rb") as f:
+ tflite_model_buf = f.read()
+ input_tensor = "input"
+ input_dtype = "uint8"
+ input_shape = (1, 224, 224, 3)
+ in_min, in_max = util.get_range_for_dtype_str(input_dtype)
+ input_data = np.random.randint(in_min, high=in_max, size=input_shape, dtype=input_dtype)
+ relay_mod, params = convert_to_relay(tflite_model_buf)
+ input_data = {input_tensor: input_data}
+ output_data = generate_ref_data(relay_mod, input_data)
+
+ mod = partition_for_ethosu(relay_mod, params)
+ compiled_models = infra.build_source(
+ mod, input_data, output_data, accel_type, output_tolerance=10
+ )
+ infra.verify_source(compiled_models, accel_type)
+
+
if __name__ == "__main__":
- test_forward_mobilenet_v1()
+ test_forward_mobilenet_v1(ACCEL_TYPES[0])
Review comment:
I think we could just remove this section completely?
##########
File path: tests/python/contrib/test_ethosu/test_networks.py
##########
@@ -65,5 +65,31 @@ def test_forward_mobilenet_v1(accel_type):
infra.verify_source(compiled_models, accel_type)
+def test_forward_mobilenet_v2(accel_type="ethos-u55-256"):
Review comment:
Is it worth parameterising this network like the above?
##########
File path: tests/python/contrib/test_ethosu/test_networks.py
##########
@@ -65,5 +65,31 @@ def test_forward_mobilenet_v1(accel_type):
infra.verify_source(compiled_models, accel_type)
+def test_forward_mobilenet_v2(accel_type="ethos-u55-256"):
+ """Test the Mobilenet V2 TF Lite model."""
+ np.random.seed(23)
+ tflite_model_file = tf_testing.get_workload_official(
+ "https://storage.googleapis.com/download.tensorflow.org/models/"
+ "tflite_11_05_08/mobilenet_v2_1.0_224_quant.tgz",
+ "mobilenet_v2_1.0_224_quant.tflite",
+ )
+ with open(tflite_model_file, "rb") as f:
+ tflite_model_buf = f.read()
+ input_tensor = "input"
+ input_dtype = "uint8"
+ input_shape = (1, 224, 224, 3)
+ in_min, in_max = util.get_range_for_dtype_str(input_dtype)
+ input_data = np.random.randint(in_min, high=in_max, size=input_shape, dtype=input_dtype)
+ relay_mod, params = convert_to_relay(tflite_model_buf)
+ input_data = {input_tensor: input_data}
+ output_data = generate_ref_data(relay_mod, input_data)
+
+ mod = partition_for_ethosu(relay_mod, params)
+ compiled_models = infra.build_source(
+ mod, input_data, output_data, accel_type, output_tolerance=10
Review comment:
It would be interesting to see if the output tolerance could be lowered now that we support more operators
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] jacobbohlin closed pull request #10242: [microNPU] Add MobileNetv2 test to network tests
Posted by GitBox <gi...@apache.org>.
jacobbohlin closed pull request #10242:
URL: https://github.com/apache/tvm/pull/10242
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] jacobbohlin commented on a change in pull request #10242: [microNPU] Add MobileNetv2 test to network tests
Posted by GitBox <gi...@apache.org>.
jacobbohlin commented on a change in pull request #10242:
URL: https://github.com/apache/tvm/pull/10242#discussion_r806660333
##########
File path: tests/python/contrib/test_ethosu/test_networks.py
##########
@@ -65,5 +65,31 @@ def test_forward_mobilenet_v1(accel_type):
infra.verify_source(compiled_models, accel_type)
+def test_forward_mobilenet_v2(accel_type="ethos-u55-256"):
Review comment:
I cannot say I have a definitive answer for this. Probably not the most efficient way to improve test coverage relative to test time.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] jacobbohlin commented on a change in pull request #10242: [microNPU] Add MobileNetv2 test to network tests
Posted by GitBox <gi...@apache.org>.
jacobbohlin commented on a change in pull request #10242:
URL: https://github.com/apache/tvm/pull/10242#discussion_r806653030
##########
File path: tests/python/contrib/test_ethosu/test_networks.py
##########
@@ -65,5 +65,31 @@ def test_forward_mobilenet_v1(accel_type):
infra.verify_source(compiled_models, accel_type)
+def test_forward_mobilenet_v2(accel_type="ethos-u55-256"):
+ """Test the Mobilenet V2 TF Lite model."""
+ np.random.seed(23)
+ tflite_model_file = tf_testing.get_workload_official(
+ "https://storage.googleapis.com/download.tensorflow.org/models/"
+ "tflite_11_05_08/mobilenet_v2_1.0_224_quant.tgz",
+ "mobilenet_v2_1.0_224_quant.tflite",
+ )
+ with open(tflite_model_file, "rb") as f:
+ tflite_model_buf = f.read()
+ input_tensor = "input"
+ input_dtype = "uint8"
+ input_shape = (1, 224, 224, 3)
+ in_min, in_max = util.get_range_for_dtype_str(input_dtype)
+ input_data = np.random.randint(in_min, high=in_max, size=input_shape, dtype=input_dtype)
+ relay_mod, params = convert_to_relay(tflite_model_buf)
+ input_data = {input_tensor: input_data}
+ output_data = generate_ref_data(relay_mod, input_data)
+
+ mod = partition_for_ethosu(relay_mod, params)
+ compiled_models = infra.build_source(
+ mod, input_data, output_data, accel_type, output_tolerance=10
+ )
+ infra.verify_source(compiled_models, accel_type)
+
+
if __name__ == "__main__":
- test_forward_mobilenet_v1()
+ test_forward_mobilenet_v1(ACCEL_TYPES[0])
Review comment:
Agreed, that's probably better.
##########
File path: tests/python/contrib/test_ethosu/test_networks.py
##########
@@ -65,5 +65,31 @@ def test_forward_mobilenet_v1(accel_type):
infra.verify_source(compiled_models, accel_type)
+def test_forward_mobilenet_v2(accel_type="ethos-u55-256"):
+ """Test the Mobilenet V2 TF Lite model."""
+ np.random.seed(23)
+ tflite_model_file = tf_testing.get_workload_official(
+ "https://storage.googleapis.com/download.tensorflow.org/models/"
+ "tflite_11_05_08/mobilenet_v2_1.0_224_quant.tgz",
+ "mobilenet_v2_1.0_224_quant.tflite",
+ )
+ with open(tflite_model_file, "rb") as f:
+ tflite_model_buf = f.read()
+ input_tensor = "input"
+ input_dtype = "uint8"
+ input_shape = (1, 224, 224, 3)
+ in_min, in_max = util.get_range_for_dtype_str(input_dtype)
+ input_data = np.random.randint(in_min, high=in_max, size=input_shape, dtype=input_dtype)
+ relay_mod, params = convert_to_relay(tflite_model_buf)
+ input_data = {input_tensor: input_data}
+ output_data = generate_ref_data(relay_mod, input_data)
+
+ mod = partition_for_ethosu(relay_mod, params)
+ compiled_models = infra.build_source(
+ mod, input_data, output_data, accel_type, output_tolerance=10
Review comment:
I can give that a try.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] manupa-arm edited a comment on pull request #10242: [microNPU] Add MobileNetv2 test to network tests
Posted by GitBox <gi...@apache.org>.
manupa-arm edited a comment on pull request #10242:
URL: https://github.com/apache/tvm/pull/10242#issuecomment-1040092717
Thanks @jacobbohlin
I think we could refactor the tests here to be a single one with stimulus being the model url and accelerator variant.
Moreover, lets write out a tuples of model url and accelerator variants for each tests -- therefore we have more control/visibility of what is being tested as opposed default parametric explosion of the test space.
Some references that we have done this at other places, also feel free to migrate any utility that you feel we can re-use to aot_test_utils
https://github.com/apache/tvm/blob/0009a308d82a321f2399923ab14b4c088461b4f2/tests/python/relay/aot/test_crt_aot_usmp.py#L205-L219
https://github.com/apache/tvm/blob/0009a308d82a321f2399923ab14b4c088461b4f2/tests/python/relay/aot/test_crt_aot_usmp.py#L232-L240
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] jacobbohlin commented on pull request #10242: [microNPU] Add MobileNetv2 test to network tests
Posted by GitBox <gi...@apache.org>.
jacobbohlin commented on pull request #10242:
URL: https://github.com/apache/tvm/pull/10242#issuecomment-1050793384
Closing this since it has been superseded by https://github.com/apache/tvm/pull/10362.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] manupa-arm edited a comment on pull request #10242: [microNPU] Add MobileNetv2 test to network tests
Posted by GitBox <gi...@apache.org>.
manupa-arm edited a comment on pull request #10242:
URL: https://github.com/apache/tvm/pull/10242#issuecomment-1040092717
Thanks @jacobbohlin
I think we could refactor the tests here to be a single one with stimulus being the model url and accelerator config.
Moreover, lets write out a tuples of model url and accelerator variants for each tests -- therefore we have more control/visibility of what is being tested as opposed default parametric explosion of the test space.
Some references that we have done this at other places, also feel free to migrate any utility that you feel we can re-use to aot_test_utils
https://github.com/apache/tvm/blob/0009a308d82a321f2399923ab14b4c088461b4f2/tests/python/relay/aot/test_crt_aot_usmp.py#L205-L219
https://github.com/apache/tvm/blob/0009a308d82a321f2399923ab14b4c088461b4f2/tests/python/relay/aot/test_crt_aot_usmp.py#L232-L240
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] jacobbohlin commented on a change in pull request #10242: [microNPU] Add MobileNetv2 test to network tests
Posted by GitBox <gi...@apache.org>.
jacobbohlin commented on a change in pull request #10242:
URL: https://github.com/apache/tvm/pull/10242#discussion_r806653030
##########
File path: tests/python/contrib/test_ethosu/test_networks.py
##########
@@ -65,5 +65,31 @@ def test_forward_mobilenet_v1(accel_type):
infra.verify_source(compiled_models, accel_type)
+def test_forward_mobilenet_v2(accel_type="ethos-u55-256"):
+ """Test the Mobilenet V2 TF Lite model."""
+ np.random.seed(23)
+ tflite_model_file = tf_testing.get_workload_official(
+ "https://storage.googleapis.com/download.tensorflow.org/models/"
+ "tflite_11_05_08/mobilenet_v2_1.0_224_quant.tgz",
+ "mobilenet_v2_1.0_224_quant.tflite",
+ )
+ with open(tflite_model_file, "rb") as f:
+ tflite_model_buf = f.read()
+ input_tensor = "input"
+ input_dtype = "uint8"
+ input_shape = (1, 224, 224, 3)
+ in_min, in_max = util.get_range_for_dtype_str(input_dtype)
+ input_data = np.random.randint(in_min, high=in_max, size=input_shape, dtype=input_dtype)
+ relay_mod, params = convert_to_relay(tflite_model_buf)
+ input_data = {input_tensor: input_data}
+ output_data = generate_ref_data(relay_mod, input_data)
+
+ mod = partition_for_ethosu(relay_mod, params)
+ compiled_models = infra.build_source(
+ mod, input_data, output_data, accel_type, output_tolerance=10
+ )
+ infra.verify_source(compiled_models, accel_type)
+
+
if __name__ == "__main__":
- test_forward_mobilenet_v1()
+ test_forward_mobilenet_v1(ACCEL_TYPES[0])
Review comment:
Agreed, that's probably better.
##########
File path: tests/python/contrib/test_ethosu/test_networks.py
##########
@@ -65,5 +65,31 @@ def test_forward_mobilenet_v1(accel_type):
infra.verify_source(compiled_models, accel_type)
+def test_forward_mobilenet_v2(accel_type="ethos-u55-256"):
+ """Test the Mobilenet V2 TF Lite model."""
+ np.random.seed(23)
+ tflite_model_file = tf_testing.get_workload_official(
+ "https://storage.googleapis.com/download.tensorflow.org/models/"
+ "tflite_11_05_08/mobilenet_v2_1.0_224_quant.tgz",
+ "mobilenet_v2_1.0_224_quant.tflite",
+ )
+ with open(tflite_model_file, "rb") as f:
+ tflite_model_buf = f.read()
+ input_tensor = "input"
+ input_dtype = "uint8"
+ input_shape = (1, 224, 224, 3)
+ in_min, in_max = util.get_range_for_dtype_str(input_dtype)
+ input_data = np.random.randint(in_min, high=in_max, size=input_shape, dtype=input_dtype)
+ relay_mod, params = convert_to_relay(tflite_model_buf)
+ input_data = {input_tensor: input_data}
+ output_data = generate_ref_data(relay_mod, input_data)
+
+ mod = partition_for_ethosu(relay_mod, params)
+ compiled_models = infra.build_source(
+ mod, input_data, output_data, accel_type, output_tolerance=10
Review comment:
I can give that a try.
##########
File path: tests/python/contrib/test_ethosu/test_networks.py
##########
@@ -65,5 +65,31 @@ def test_forward_mobilenet_v1(accel_type):
infra.verify_source(compiled_models, accel_type)
+def test_forward_mobilenet_v2(accel_type="ethos-u55-256"):
Review comment:
I cannot say I have a definitive answer for this. Probably not the most efficient way to improve test coverage relative to test time.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] jacobbohlin commented on pull request #10242: [microNPU] Add MobileNetv2 test to network tests
Posted by GitBox <gi...@apache.org>.
jacobbohlin commented on pull request #10242:
URL: https://github.com/apache/tvm/pull/10242#issuecomment-1039254125
@manupa-arm @mbaret @lhutton1 @ekalda @NicolaLancellotti @dchauhan-arm
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] lhutton1 commented on a change in pull request #10242: [microNPU] Add MobileNetv2 test to network tests
Posted by GitBox <gi...@apache.org>.
lhutton1 commented on a change in pull request #10242:
URL: https://github.com/apache/tvm/pull/10242#discussion_r806313919
##########
File path: tests/python/contrib/test_ethosu/test_networks.py
##########
@@ -65,5 +65,31 @@ def test_forward_mobilenet_v1(accel_type):
infra.verify_source(compiled_models, accel_type)
+def test_forward_mobilenet_v2(accel_type="ethos-u55-256"):
+ """Test the Mobilenet V2 TF Lite model."""
+ np.random.seed(23)
+ tflite_model_file = tf_testing.get_workload_official(
+ "https://storage.googleapis.com/download.tensorflow.org/models/"
+ "tflite_11_05_08/mobilenet_v2_1.0_224_quant.tgz",
+ "mobilenet_v2_1.0_224_quant.tflite",
+ )
+ with open(tflite_model_file, "rb") as f:
+ tflite_model_buf = f.read()
+ input_tensor = "input"
+ input_dtype = "uint8"
+ input_shape = (1, 224, 224, 3)
+ in_min, in_max = util.get_range_for_dtype_str(input_dtype)
+ input_data = np.random.randint(in_min, high=in_max, size=input_shape, dtype=input_dtype)
+ relay_mod, params = convert_to_relay(tflite_model_buf)
+ input_data = {input_tensor: input_data}
+ output_data = generate_ref_data(relay_mod, input_data)
+
+ mod = partition_for_ethosu(relay_mod, params)
+ compiled_models = infra.build_source(
+ mod, input_data, output_data, accel_type, output_tolerance=10
+ )
+ infra.verify_source(compiled_models, accel_type)
+
+
if __name__ == "__main__":
- test_forward_mobilenet_v1()
+ test_forward_mobilenet_v1(ACCEL_TYPES[0])
Review comment:
I think we could just remove this section completely?
##########
File path: tests/python/contrib/test_ethosu/test_networks.py
##########
@@ -65,5 +65,31 @@ def test_forward_mobilenet_v1(accel_type):
infra.verify_source(compiled_models, accel_type)
+def test_forward_mobilenet_v2(accel_type="ethos-u55-256"):
Review comment:
Is it worth parameterising this network like the above?
##########
File path: tests/python/contrib/test_ethosu/test_networks.py
##########
@@ -65,5 +65,31 @@ def test_forward_mobilenet_v1(accel_type):
infra.verify_source(compiled_models, accel_type)
+def test_forward_mobilenet_v2(accel_type="ethos-u55-256"):
+ """Test the Mobilenet V2 TF Lite model."""
+ np.random.seed(23)
+ tflite_model_file = tf_testing.get_workload_official(
+ "https://storage.googleapis.com/download.tensorflow.org/models/"
+ "tflite_11_05_08/mobilenet_v2_1.0_224_quant.tgz",
+ "mobilenet_v2_1.0_224_quant.tflite",
+ )
+ with open(tflite_model_file, "rb") as f:
+ tflite_model_buf = f.read()
+ input_tensor = "input"
+ input_dtype = "uint8"
+ input_shape = (1, 224, 224, 3)
+ in_min, in_max = util.get_range_for_dtype_str(input_dtype)
+ input_data = np.random.randint(in_min, high=in_max, size=input_shape, dtype=input_dtype)
+ relay_mod, params = convert_to_relay(tflite_model_buf)
+ input_data = {input_tensor: input_data}
+ output_data = generate_ref_data(relay_mod, input_data)
+
+ mod = partition_for_ethosu(relay_mod, params)
+ compiled_models = infra.build_source(
+ mod, input_data, output_data, accel_type, output_tolerance=10
Review comment:
It would be interesting to see if the output tolerance could be lowered now that we support more operators
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] jacobbohlin commented on pull request #10242: [microNPU] Add MobileNetv2 test to network tests
Posted by GitBox <gi...@apache.org>.
jacobbohlin commented on pull request #10242:
URL: https://github.com/apache/tvm/pull/10242#issuecomment-1040118155
Thanks @manupa-arm. That is a much better approach, I will update according to your suggestions.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] jacobbohlin commented on pull request #10242: [microNPU] Add MobileNetv2 test to network tests
Posted by GitBox <gi...@apache.org>.
jacobbohlin commented on pull request #10242:
URL: https://github.com/apache/tvm/pull/10242#issuecomment-1039254125
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] manupa-arm commented on pull request #10242: [microNPU] Add MobileNetv2 test to network tests
Posted by GitBox <gi...@apache.org>.
manupa-arm commented on pull request #10242:
URL: https://github.com/apache/tvm/pull/10242#issuecomment-1040092717
Thanks @jacobbohlin
I think we could refactor the tests here to be a single one with stimulus being the model url and accelerator.
Moreover, lets write out a tuples of model url and accelerator variants for each tests -- therefore we have more control/visibility of what is being tested as opposed default parametric explosion of the test space.
Some references that we have done this at other places, also feel free to migrate any utility that you feel we can re-use to aot_test_utils
https://github.com/apache/tvm/blob/0009a308d82a321f2399923ab14b4c088461b4f2/tests/python/relay/aot/test_crt_aot_usmp.py#L205-L219
https://github.com/apache/tvm/blob/0009a308d82a321f2399923ab14b4c088461b4f2/tests/python/relay/aot/test_crt_aot_usmp.py#L232-L240
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] manupa-arm commented on pull request #10242: [microNPU] Add MobileNetv2 test to network tests
Posted by GitBox <gi...@apache.org>.
manupa-arm commented on pull request #10242:
URL: https://github.com/apache/tvm/pull/10242#issuecomment-1040092717
Thanks @jacobbohlin
I think we could refactor the tests here to be a single one with stimulus being the model url and accelerator.
Moreover, lets write out a tuples of model url and accelerator variants for each tests -- therefore we have more control/visibility of what is being tested as opposed default parametric explosion of the test space.
Some references that we have done this at other places, also feel free to migrate any utility that you feel we can re-use to aot_test_utils
https://github.com/apache/tvm/blob/0009a308d82a321f2399923ab14b4c088461b4f2/tests/python/relay/aot/test_crt_aot_usmp.py#L205-L219
https://github.com/apache/tvm/blob/0009a308d82a321f2399923ab14b4c088461b4f2/tests/python/relay/aot/test_crt_aot_usmp.py#L232-L240
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] manupa-arm edited a comment on pull request #10242: [microNPU] Add MobileNetv2 test to network tests
Posted by GitBox <gi...@apache.org>.
manupa-arm edited a comment on pull request #10242:
URL: https://github.com/apache/tvm/pull/10242#issuecomment-1040092717
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org