07/Label semantic roles #5798
Merged
07/Label semantic roles #5798
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
… label_semantic_roles
… label_semantic_roles
lcy-seso
reviewed
Nov 21, 2017
| size=[mark_dict_len, mark_dim], | ||
| data_type='float32', | ||
| is_sparse=IS_SPARSE, | ||
| param_attr=std_0) |
There was a problem hiding this comment.
这个 embedding 参数是需要学习的，对吗，这里的 std_0 会将参数初始化成全零吗？但book里面似乎也是这样设置，这个好奇怪。
There was a problem hiding this comment.
book里的embedding层提供了初始化模型，不用学习~
There was a problem hiding this comment.
嗯，这个正在加，load之前训练好的embedding
| size=hidden_dim, | ||
| candidate_activation='relu', | ||
| gate_activation='sigmoid', | ||
| cell_activation='sigmoid', |
There was a problem hiding this comment.
这个激活的组合好奇怪，除去 gate 固定用 sigmoid 之外。
上面红色圈出来的是cell_activation，蓝色圈出来的是candidate_activation，我理解的对吗？
sigmoid激活域最小，relu激活输出无上限，连乘太多次之后容易激活多大（CRF 后面会对relu的激活值再求 exponential），一般都习惯性从 tanh 作为默认激活开始训练。- 不过 book 原来的配置就是这样的设置，先以（数值）稳定地训练为目标吧。
| mark_dim = 5 | ||
| hidden_dim = 512 | ||
| depth = 8 | ||
| default_std = 1 / math.sqrt(hidden_dim) / 3.0 |
There was a problem hiding this comment.
先去掉这些奇怪的初始化方式了，默认使用XavierInitializer
| layers.fc(input=input_tmp[1], | ||
| size=label_dict_len, | ||
| bias_attr=std_default, | ||
| param_attr=lstm_para_attr) |
There was a problem hiding this comment.
这个参数会被初始化成全零吗，只剩下 bias？看着好奇怪/危险。
| name='label', shape=[1], data_type='int32', main_program=program) | ||
| hidden = layers.fc(input=images, | ||
| size=128, | ||
| act='relu', |
Superjomn
reviewed
Nov 22, 2017
| candidate_activation='relu', | ||
| gate_activation='sigmoid', | ||
| cell_activation='sigmoid', | ||
| is_reverse=((i % 2) == 1), |
There was a problem hiding this comment.
change this argument to
is_reversed
or
reversed
better?
@qingqing01
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
project: #5813
fix: #5691
Tasks