81 lines
2.4 KiB
Python
81 lines
2.4 KiB
Python
# Copyright (c) 2024 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.
|
|
|
|
import unittest
|
|
|
|
import numpy as np
|
|
|
|
import paddle
|
|
from paddle.nn.functional.input import embedding_renorm_
|
|
|
|
|
|
def ref_embedding_renorm_(x, weight, max_norm, norm_type=2.0):
|
|
weight = weight.copy()
|
|
x = np.reshape(x, (-1,))
|
|
x = np.unique(x)
|
|
x = np.sort(x)
|
|
for i in range(len(x)):
|
|
norm = np.linalg.norm(
|
|
weight[int(x[i])], ord=norm_type, axis=0, keepdims=False
|
|
)
|
|
if norm > max_norm:
|
|
weight[int(x[i])] *= max_norm / (norm + 1e-7)
|
|
return weight
|
|
|
|
|
|
class TestEmbeddingRenormOp(unittest.TestCase):
|
|
def setUp(self):
|
|
self._init_attr()
|
|
self.dtype = self._init_dtype()
|
|
x = np.array([[2, 1, 3], [4, 5, 6]]).astype("int64")
|
|
weight = np.random.random((10, 4)).astype(self.dtype) * 10
|
|
y_ref = ref_embedding_renorm_(x, weight, self.max_norm, self.norm_type)
|
|
self.inputs = {'X': x, 'Weight': weight}
|
|
self.outputs = {'Out': y_ref}
|
|
self.attrs = {'max_norm': self.max_norm, 'norm_type': self.norm_type}
|
|
|
|
def _init_dtype(self):
|
|
return "float32"
|
|
|
|
def _init_attr(self):
|
|
self.max_norm = 1.0
|
|
self.norm_type = 2.0
|
|
|
|
def test_check_output(self):
|
|
paddle_result = embedding_renorm_(
|
|
paddle.to_tensor(self.inputs['X']),
|
|
paddle.to_tensor(self.inputs['Weight']),
|
|
self.max_norm,
|
|
self.norm_type,
|
|
)
|
|
np.testing.assert_allclose(
|
|
paddle_result.numpy(), self.outputs['Out'], atol=1e-5
|
|
)
|
|
|
|
|
|
class TestEmbeddingRenormOp1(TestEmbeddingRenormOp):
|
|
def _init_attr(self):
|
|
self.max_norm = 1.0
|
|
self.norm_type = 1.0
|
|
|
|
|
|
class TestEmbeddingRenormOp2(TestEmbeddingRenormOp):
|
|
def _init_attr(self):
|
|
self.max_norm = 1.0
|
|
self.norm_type = 3.0
|
|
|
|
|
|
if __name__ == '__main__':
|
|
unittest.main()
|