wehub-resource-sync
248 lines
9.7 KiB
Python
248 lines
9.7 KiB
Python
# Copyright (c) 2021 PaddlePaddle Authors. All Rights Reserved.
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
|
|
from __future__ import annotations
|
|
|
|
import unittest
|
|
from functools import partial
|
|
|
|
import numpy as np
|
|
from program_config import ProgramConfig, TensorConfig
|
|
from trt_layer_auto_scan_test import TrtLayerAutoScanTest
|
|
|
|
import paddle.inference as paddle_infer
|
|
|
|
|
|
class TrtConvertFillConstantTest(TrtLayerAutoScanTest):
|
|
def is_program_valid(self, program_config: ProgramConfig) -> bool:
|
|
return True
|
|
|
|
def sample_program_configs(self):
|
|
def generate_value_data():
|
|
return np.array([1]).astype(np.int32)
|
|
|
|
def generate_input_data():
|
|
return np.random.random([2, 5, 7]).astype(np.float32)
|
|
|
|
for dtype in [5, 2, 3]:
|
|
for str_value in ["2", "23", "-1"]:
|
|
value = float(str_value)
|
|
if np.random.choice([False, True]):
|
|
str_value = str_value
|
|
else:
|
|
str_value = ""
|
|
dics = [
|
|
{
|
|
"str_value": str_value,
|
|
"value": value,
|
|
"dtype": dtype,
|
|
"shape": [2, 3, 4],
|
|
},
|
|
{"axis": -1},
|
|
]
|
|
for mode in ["ShapeTensor", "ShapeTensorList"]:
|
|
self.mode = mode
|
|
if mode == "ValueTensor":
|
|
ops_config = [
|
|
{
|
|
"op_type": "fill_constant",
|
|
"op_inputs": {"ValueTensor": ["value_data"]},
|
|
"op_outputs": {
|
|
"Out": ["out_data"],
|
|
},
|
|
"op_attrs": dics[0],
|
|
},
|
|
]
|
|
elif mode == "ShapeTensor":
|
|
ops_config = [
|
|
{
|
|
"op_type": "shape",
|
|
"op_inputs": {
|
|
"Input": ["input_data"],
|
|
},
|
|
"op_outputs": {"Out": ["shape_data"]},
|
|
"op_attrs": {},
|
|
},
|
|
{
|
|
"op_type": "fill_constant",
|
|
"op_inputs": {
|
|
"ShapeTensor": ["shape_data"],
|
|
},
|
|
"op_outputs": {
|
|
"Out": ["out_data"],
|
|
},
|
|
"op_attrs": dics[0],
|
|
},
|
|
]
|
|
else:
|
|
ops_config = [
|
|
{
|
|
"op_type": "shape",
|
|
"op_inputs": {
|
|
"Input": ["input_data"],
|
|
},
|
|
"op_outputs": {"Out": ["shape_data"]},
|
|
"op_attrs": {},
|
|
},
|
|
{
|
|
"op_type": "slice",
|
|
"op_inputs": {
|
|
"Input": ["shape_data"],
|
|
},
|
|
"op_outputs": {"Out": ["split_shape_data_0"]},
|
|
"op_attrs": {
|
|
"axes": [0],
|
|
"starts": [0],
|
|
"ends": [1],
|
|
"decrease_axis": [],
|
|
},
|
|
},
|
|
{
|
|
"op_type": "slice",
|
|
"op_inputs": {
|
|
"Input": ["shape_data"],
|
|
},
|
|
"op_outputs": {"Out": ["split_shape_data_1"]},
|
|
"op_attrs": {
|
|
"axes": [0],
|
|
"starts": [1],
|
|
"ends": [2],
|
|
"decrease_axis": [],
|
|
},
|
|
},
|
|
{
|
|
"op_type": "slice",
|
|
"op_inputs": {
|
|
"Input": ["shape_data"],
|
|
},
|
|
"op_outputs": {"Out": ["split_shape_data_2"]},
|
|
"op_attrs": {
|
|
"axes": [0],
|
|
"starts": [2],
|
|
"ends": [3],
|
|
"decrease_axis": [],
|
|
},
|
|
},
|
|
{
|
|
"op_type": "fill_constant",
|
|
"op_inputs": {
|
|
"ShapeTensorList": [
|
|
"split_shape_data_0",
|
|
"split_shape_data_1",
|
|
"split_shape_data_2",
|
|
],
|
|
},
|
|
"op_outputs": {
|
|
"Out": ["out_data"],
|
|
},
|
|
"op_attrs": dics[0],
|
|
},
|
|
]
|
|
ops = self.generate_op_config(ops_config)
|
|
if mode == "ValueTensor":
|
|
program_config = ProgramConfig(
|
|
ops=ops,
|
|
weights={},
|
|
inputs={
|
|
"value_data": TensorConfig(
|
|
data_gen=partial(generate_value_data)
|
|
),
|
|
},
|
|
outputs=["out_data"],
|
|
)
|
|
else:
|
|
program_config = ProgramConfig(
|
|
ops=ops,
|
|
weights={},
|
|
inputs={
|
|
"input_data": TensorConfig(
|
|
data_gen=partial(generate_input_data)
|
|
),
|
|
},
|
|
outputs=["out_data"],
|
|
)
|
|
yield program_config
|
|
|
|
def generate_dynamic_shape(self, attrs):
|
|
if self.mode == "ValueTensor":
|
|
self.input_shape = [1, 1]
|
|
max_shape = list(self.input_shape)
|
|
min_shape = list(self.input_shape)
|
|
opt_shape = list(self.input_shape)
|
|
for i in range(len(self.input_shape)):
|
|
max_shape[i] = max_shape[i] + 1
|
|
self.dynamic_shape.min_input_shape = {"value_data": min_shape}
|
|
self.dynamic_shape.max_input_shape = {"value_data": max_shape}
|
|
self.dynamic_shape.opt_input_shape = {"value_data": opt_shape}
|
|
else:
|
|
self.dynamic_shape.min_input_shape = {
|
|
"input_data": [2, 3, 7],
|
|
}
|
|
self.dynamic_shape.max_input_shape = {
|
|
"input_data": [2, 5, 7],
|
|
}
|
|
self.dynamic_shape.opt_input_shape = {
|
|
"input_data": [2, 4, 7],
|
|
}
|
|
return self.dynamic_shape
|
|
|
|
def sample_predictor_configs(
|
|
self, program_config, run_pir=False
|
|
) -> tuple[paddle_infer.Config, list[int], float]:
|
|
def clear_dynamic_shape():
|
|
self.dynamic_shape.min_input_shape = {}
|
|
self.dynamic_shape.max_input_shape = {}
|
|
self.dynamic_shape.opt_input_shape = {}
|
|
|
|
def generate_trt_nodes_num(attrs, dynamic_shape):
|
|
if self.mode == "ValueTensor":
|
|
return 0, 3
|
|
else:
|
|
ver = paddle_infer.get_trt_compile_version()
|
|
if ver[0] * 1000 + ver[1] * 100 + ver[2] * 10 < 8500:
|
|
return 1, 3
|
|
else:
|
|
return 1, 2
|
|
|
|
attrs = [
|
|
program_config.ops[i].attrs for i in range(len(program_config.ops))
|
|
]
|
|
# Don't test static shape
|
|
|
|
# for dynamic_shape
|
|
self.generate_dynamic_shape(attrs)
|
|
self.trt_param.precision = paddle_infer.PrecisionType.Float32
|
|
yield (
|
|
self.create_inference_config(),
|
|
generate_trt_nodes_num(attrs, True),
|
|
1e-5,
|
|
)
|
|
self.trt_param.precision = paddle_infer.PrecisionType.Half
|
|
yield (
|
|
self.create_inference_config(),
|
|
generate_trt_nodes_num(attrs, True),
|
|
1e-3,
|
|
)
|
|
|
|
def add_skip_trt_case(self):
|
|
pass
|
|
|
|
def test(self):
|
|
self.add_skip_trt_case()
|
|
self.run_test(run_pir=True)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
unittest.main()
|