commit
3c4d4065ae
@ -1,125 +0,0 @@
|
||||
#edit-mode: -*- python -*-
|
||||
# Copyright (c) 2016 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.
|
||||
|
||||
#Todo(luotao02) This config is only used for unitest. It is out of date now, and will be updated later.
|
||||
|
||||
TrainData(ProtoData(
|
||||
files = 'trainer/tests/train_files.txt',
|
||||
usage_ratio = 1.0,
|
||||
))
|
||||
|
||||
TestData(ProtoData(
|
||||
files = 'trainer/tests/test_files.txt'
|
||||
))
|
||||
|
||||
default_initial_std(1)
|
||||
default_decay_rate(4e-4)
|
||||
default_device(0)
|
||||
|
||||
Inputs("features", "word", "pos", "chunk")
|
||||
|
||||
Outputs("crf")
|
||||
|
||||
Layer(
|
||||
name = "features",
|
||||
type = "data",
|
||||
size = 4339,
|
||||
)
|
||||
|
||||
Layer(
|
||||
name = "word",
|
||||
type = "data",
|
||||
size = 478,
|
||||
)
|
||||
|
||||
Layer(
|
||||
name = "pos",
|
||||
type = "data",
|
||||
size = 45
|
||||
)
|
||||
|
||||
Layer(
|
||||
name = "chunk",
|
||||
type = "data",
|
||||
size = 23
|
||||
)
|
||||
|
||||
Layer(
|
||||
name = "output",
|
||||
type = "mixed",
|
||||
size = 23,
|
||||
bias = False,
|
||||
device = -1,
|
||||
inputs = [
|
||||
FullMatrixProjection("features", parameter_name="feature_weights"),
|
||||
# TableProjection("word"),
|
||||
# TableProjection("pos"),
|
||||
],
|
||||
)
|
||||
|
||||
Layer(
|
||||
name = "crf",
|
||||
type = "crf",
|
||||
size = 23,
|
||||
device = -1,
|
||||
inputs = [
|
||||
Input("output", parameter_name="crfw"),
|
||||
"chunk"
|
||||
]
|
||||
)
|
||||
|
||||
Layer(
|
||||
name = "crf_decoding",
|
||||
type = "crf_decoding",
|
||||
size = 23,
|
||||
device = -1,
|
||||
inputs = [
|
||||
Input("output", parameter_name="crfw"),
|
||||
"chunk"
|
||||
]
|
||||
)
|
||||
|
||||
Evaluator(
|
||||
name = "error",
|
||||
type = "sum",
|
||||
inputs = "crf_decoding",
|
||||
)
|
||||
|
||||
'''
|
||||
# chuck evaluator cannot be used for GPU training
|
||||
Evaluator(
|
||||
name = "chunk_f1",
|
||||
type = "chunk",
|
||||
inputs = ["crf_decoding", "chunk"],
|
||||
chunk_scheme = "IOB",
|
||||
num_chunk_types = 11,
|
||||
)
|
||||
'''
|
||||
|
||||
Settings(
|
||||
algorithm = 'sgd',
|
||||
batch_size = 100,
|
||||
average_window = 0.5,
|
||||
max_average_window = 2500,
|
||||
learning_rate = 1e-1,
|
||||
learning_rate_decay_a = 5e-7,
|
||||
learning_rate_decay_b = 0.75,
|
||||
l1weight = 0,
|
||||
l2weight = 1,
|
||||
c1 = 0.0001,
|
||||
backoff = 0.5,
|
||||
owlqn_steps = 100,
|
||||
max_backoff = 5,
|
||||
)
|
Binary file not shown.
File diff suppressed because one or more lines are too long
File diff suppressed because it is too large
Load Diff
File diff suppressed because it is too large
Load Diff
@ -1 +0,0 @@
|
||||
trainer/tests/test_proto.bin
|
@ -1 +0,0 @@
|
||||
trainer/tests/data_bin_part
|
File diff suppressed because it is too large
Load Diff
@ -1 +0,0 @@
|
||||
trainer/tests/train_proto.bin
|
@ -1 +0,0 @@
|
||||
trainer/tests/compare_sparse_data
|
Loading…
Reference in new issue