NicolaLancellotti commented on a change in pull request #9384:
URL: https://github.com/apache/tvm/pull/9384#discussion_r740305018
##########
File path: tests/python/contrib/test_ethosu/test_legalize.py
##########
@@ -458,7 +458,102 @@ def verify(ext_func):
mod = partition_ethosu_by_table(mod, depthwise_pattern_table)
mod["tvmgen_default_ethosu_main_0"] = dataflow_pattern.rewrite(
- legalize.EthosuDepthwiseConv2DRewriter(),
mod["tvmgen_default_ethosu_main_0"]
+ legalize.DepthwiseConv2DRewriter(), mod["tvmgen_default_ethosu_main_0"]
+ )
+ verify(mod["tvmgen_default_ethosu_main_0"])
+
+
[email protected]("pooling_type", ["MAX", "AVG"])
[email protected]("ifm_shape", [[1, 9, 12, 3], [1, 10, 20, 2]])
[email protected]("strides", [[1, 2], [2, 3]])
[email protected]("pool_shape", [[1, 2], [2, 3]])
[email protected]("activation_function", ["NONE", "RELU"])
[email protected]("padding", ["SAME", "VALID"])
+def test_tflite_pool2d_legalize(
+ ifm_shape, pooling_type, strides, pool_shape, activation_function, padding
+):
+ dtype = "int8"
+
+ def create_tflite_graph():
+ class Model(tf.Module):
+ @tf.function
+ def tf_function(self, x):
+ if pooling_type == "MAX":
+ op = tf.nn.max_pool(x, pool_shape, strides, padding)
+ elif pooling_type == "AVG":
+ op = tf.nn.avg_pool(x, pool_shape, strides, padding)
+ if activation_function == "RELU":
+ op = tf.nn.relu(op)
+ return op
+
+ model = Model()
+ concrete_func = model.tf_function.get_concrete_function(
+ tf.TensorSpec(ifm_shape, dtype=tf.float32)
+ )
+
+ # Convert the model
+ def representative_dataset():
+ for _ in range(100):
+ data = np.random.rand(*tuple(ifm_shape))
+ yield [data.astype(np.float32)]
+
+ converter =
tf.lite.TFLiteConverter.from_concrete_functions([concrete_func])
+ converter.optimizations = [tf.lite.Optimize.DEFAULT]
+ converter.representative_dataset = representative_dataset
+ converter.target_spec.supported_ops =
[tf.lite.OpsSet.TFLITE_BUILTINS_INT8]
+ converter.inference_input_type = tf.int8
+ converter.inference_output_type = tf.int8
+ tflite_model = converter.convert()
+ return tflite_model
+
+ def verify(ext_func):
+ ofm_shape = infra.compute_ofm_shape(ifm_shape, padding, pool_shape,
strides)
+ op = ext_func.body
+ assert list(op.args[0].checked_type.shape) == ifm_shape
+ assert op.args[0].checked_type.dtype == dtype
+ assert list(op.checked_type.shape) == ofm_shape
+ assert op.checked_type.dtype == dtype
+ assert op.attrs.pooling_type == pooling_type
+ assert list(op.attrs.strides) == strides
+ assert list(op.attrs.padding) == infra.compute_padding_shape(
+ ifm_shape, ofm_shape, padding, pool_shape, strides
+ )
+ assert list(op.attrs.pool_shape) == pool_shape
+ assert op.attrs.ofm_channels == ifm_shape[3]
+ if activation_function == "RELU":
+ assert str(op.attrs.activation) == "CLIP"
+
+ if pooling_type == "MAX":
+ rewriter = legalize.MaxPoolingRewriter()
+ pattern_table = [
+ (
+ ethosu.MaxPool2DParams.composite_name,
+ ethosu.qnn_maxpool2d_pattern(),
+ lambda pat: ethosu.MaxPool2DParams(pat).is_valid(),
+ ),
+ ]
+ elif pooling_type == "AVG":
+ rewriter = legalize.AvgPoolingRewriter()
+ pattern_table = [
+ (
+ ethosu.AvgPool2DParams.composite_name,
+ ethosu.qnn_avgpool2d_pattern(),
+ lambda pat: ethosu.AvgPool2DParams(pat).is_valid(),
+ ),
+ ]
Review comment:
Thinking about it, it should be ok to have only one test, that because
`MAX` and `AVG` are the two possible values of the `pooling_type` attribute of
the Ethos(TM)-U Pooling operator. We are not testing two different operators in
one test, but different parameters of the same operator.
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
