forked from PaddlePaddle/PaddleRec
-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
1 parent
923f599
commit 388beb7
Showing
15 changed files
with
230 additions
and
723 deletions.
There are no files selected for viewing
This file was deleted.
Oops, something went wrong.
This file contains 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
This file contains 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
This file contains 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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,113 @@ | ||
| # Copyright (c) 2020 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 paddle | ||
| import paddle.nn as nn | ||
| import paddle.nn.functional as F | ||
| import math | ||
|
|
||
| import net | ||
|
|
||
|
|
||
| class DygraphModel(): | ||
| # define model | ||
| def create_model(self, config): | ||
| vocab_text_size = config.get("hyper_parameters.vocab_text_size") | ||
| vocab_tag_size = config.get("hyper_parameters.vocab_tag_size") | ||
| emb_dim = config.get("hyper_parameters.emb_dim") | ||
| hid_dim = config.get("hyper_parameters.hid_dim") | ||
| win_size = config.get("hyper_parameters.win_size") | ||
| margin = config.get("hyper_parameters.margin") | ||
| neg_size = config.get("hyper_parameters.neg_size") | ||
| text_len = config.get("hyper_parameters.text_len") | ||
|
|
||
| tagspace_model = net.TagspaceLayer(vocab_text_size, vocab_tag_size, | ||
| emb_dim, hid_dim, win_size, margin, | ||
| neg_size, text_len) | ||
| return tagspace_model | ||
|
|
||
| # define feeds which convert numpy of batch data to paddle.tensor | ||
| def create_feeds(self, batch_data, text_len, neg_size): | ||
| text = paddle.to_tensor(batch_data[0].numpy().astype('int64').reshape( | ||
| -1, text_len)) | ||
| pos_tag = paddle.to_tensor(batch_data[1].numpy().astype('int64') | ||
| .reshape(-1, 1)) | ||
| neg_tag = paddle.to_tensor(batch_data[2].numpy().astype('int64') | ||
| .reshape(-1, neg_size)) | ||
| return [text, pos_tag, neg_tag] | ||
|
|
||
| # define loss function by predicts and label | ||
| def create_loss(self, batch_size, margin, cos_pos, cos_neg): | ||
| loss_part1 = paddle.subtract( | ||
| paddle.full( | ||
| shape=[batch_size, 1], fill_value=margin, dtype='float32'), | ||
| cos_pos) | ||
| loss_part2 = paddle.add(loss_part1, cos_neg) | ||
| loss_part3 = paddle.maximum( | ||
| paddle.full( | ||
| shape=[batch_size, 1], fill_value=0.0, dtype='float32'), | ||
| loss_part2) | ||
| avg_cost = paddle.mean(loss_part3) | ||
| return avg_cost | ||
|
|
||
| # define optimizer | ||
| def create_optimizer(self, dy_model, config): | ||
| lr = config.get("hyper_parameters.optimizer.learning_rate", 0.001) | ||
| optimizer = paddle.optimizer.Adagrad( | ||
| learning_rate=lr, parameters=dy_model.parameters()) | ||
| return optimizer | ||
|
|
||
| # define metrics such as auc/acc | ||
| # multi-task need to define multi metric | ||
|
|
||
| def get_acc(self, x, y, batch_size): | ||
| less = paddle.cast(paddle.less_than(x, y), dtype='float32') | ||
| label_ones = paddle.full( | ||
| dtype='float32', shape=[batch_size, 1], fill_value=1.0) | ||
| correct = paddle.sum(less) | ||
| total = paddle.sum(label_ones) | ||
| acc = paddle.divide(correct, total) | ||
| return acc | ||
|
|
||
| def create_metrics(self): | ||
| metrics_list_name = [] | ||
| metrics_list = [] | ||
| return metrics_list, metrics_list_name | ||
|
|
||
| # construct train forward phase | ||
| def train_forward(self, dy_model, metrics_list, batch_data, config): | ||
| neg_size = config.get("hyper_parameters.neg_size") | ||
| text_len = config.get("hyper_parameters.text_len") | ||
| margin = config.get("hyper_parameters.margin") | ||
| batch_size = config.get("runner.train_batch_size", 128) | ||
| inputs = self.create_feeds(batch_data, text_len, neg_size) | ||
|
|
||
| cos_pos, cos_neg = dy_model.forward(inputs) | ||
| loss = self.create_loss(batch_size, margin, cos_pos, cos_neg) | ||
| # update metrics | ||
| acc = self.get_acc(cos_neg, cos_pos, batch_size) | ||
| print_dict = {"loss": loss, "ACC": acc} | ||
| return loss, metrics_list, print_dict | ||
|
|
||
| def infer_forward(self, dy_model, metrics_list, batch_data, config): | ||
| neg_size = config.get("hyper_parameters.neg_size") | ||
| text_len = config.get("hyper_parameters.text_len") | ||
| batch_size = config.get("runner.infer_batch_size", 128) | ||
| inputs = self.create_feeds(batch_data, text_len, neg_size) | ||
|
|
||
| cos_pos, cos_neg = dy_model.forward(inputs) | ||
| # update metrics | ||
| acc = self.get_acc(cos_neg, cos_pos, batch_size) | ||
| print_dict = {"ACC": acc} | ||
| return metrics_list, print_dict |
Oops, something went wrong.