Merge pull request #1869 from LDOUBLEV/rare

upload rare code

configs/rec/rec_mv3_tps_bilstm_att.yml

@@ -0,0 +1,102 @@
+Global:
+  use_gpu: true
+  epoch_num: 72
+  log_smooth_window: 20
+  print_batch_step: 10
+  save_model_dir: ./output/rec/rec_mv3_tps_bilstm_att/
+  save_epoch_step: 3
+  # evaluation is run every 5000 iterations after the 4000th iteration
+  eval_batch_step: [0, 2000]
+  # if pretrained_model is saved in static mode, load_static_weights must set to True
+  cal_metric_during_train: True
+  pretrained_model:
+  checkpoints:
+  save_inference_dir:
+  use_visualdl: False
+  infer_img: doc/imgs_words/ch/word_1.jpg
+  # for data or label process
+  character_dict_path: 
+  character_type: en
+  max_text_length: 25
+  infer_mode: False
+  use_space_char: False
+
+
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.999
+  lr:
+    learning_rate: 0.0005
+  regularizer:
+    name: 'L2'
+    factor: 0.00001
+
+Architecture:
+  model_type: rec
+  algorithm: RARE
+  Transform:
+    name: TPS
+    num_fiducial: 20
+    loc_lr: 0.1
+    model_name: small
+  Backbone:
+    name: MobileNetV3
+    scale: 0.5
+    model_name: large
+  Neck:
+    name: SequenceEncoder
+    encoder_type: rnn 
+    hidden_size: 96
+  Head:
+    name: AttentionHead  
+    hidden_size: 96
+    
+
+Loss:
+  name: AttentionLoss
+
+PostProcess:
+  name: AttnLabelDecode
+
+Metric:
+  name: RecMetric
+  main_indicator: acc
+
+Train:
+  dataset:
+    name: LMDBDateSet
+    data_dir: ../training/
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - AttnLabelEncode: # Class handling label
+      - RecResizeImg:
+          image_shape: [3, 32, 100]
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
+  loader:
+    shuffle: True
+    batch_size_per_card: 256
+    drop_last: True
+    num_workers: 8
+
+Eval:
+  dataset:
+    name: LMDBDateSet
+    data_dir: ../validation/
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - AttnLabelEncode: # Class handling label
+      - RecResizeImg:
+          image_shape: [3, 32, 100]
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: 256
+    num_workers: 1

configs/rec/rec_r34_vd_tps_bilstm_att.yml

@@ -0,0 +1,101 @@
+Global:
+  use_gpu: true
+  epoch_num: 400
+  log_smooth_window: 20
+  print_batch_step: 10
+  save_model_dir: ./output/rec/b3_rare_r34_none_gru/
+  save_epoch_step: 3
+  # evaluation is run every 5000 iterations after the 4000th iteration
+  eval_batch_step: [0, 2000]
+  # if pretrained_model is saved in static mode, load_static_weights must set to True
+  cal_metric_during_train: True
+  pretrained_model:
+  checkpoints:
+  save_inference_dir:
+  use_visualdl: False
+  infer_img: doc/imgs_words/ch/word_1.jpg
+  # for data or label process
+  character_dict_path: 
+  character_type: en
+  max_text_length: 25
+  infer_mode: False
+  use_space_char: False
+
+
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.999
+  lr:
+    learning_rate: 0.0005
+  regularizer:
+    name: 'L2'
+    factor: 0.00000
+
+Architecture:
+  model_type: rec
+  algorithm: RARE
+  Transform:
+    name: TPS
+    num_fiducial: 20
+    loc_lr: 0.1
+    model_name: large
+  Backbone:
+    name: ResNet  
+    layers: 34
+  Neck:
+    name: SequenceEncoder
+    encoder_type: rnn 
+    hidden_size: 256 #96
+  Head:
+    name: AttentionHead  # AttentionHead
+    hidden_size: 256 #
+    l2_decay: 0.00001
+
+Loss:
+  name: AttentionLoss
+
+PostProcess:
+  name: AttnLabelDecode
+
+Metric:
+  name: RecMetric
+  main_indicator: acc
+
+Train:
+  dataset:
+    name: LMDBDateSet
+    data_dir: ../training/
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - AttnLabelEncode: # Class handling label
+      - RecResizeImg:
+          image_shape: [3, 32, 100]
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
+  loader:
+    shuffle: True
+    batch_size_per_card: 256
+    drop_last: True
+    num_workers: 8
+
+Eval:
+  dataset:
+    name: LMDBDateSet
+    data_dir: ../validation/
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - AttnLabelEncode: # Class handling label
+      - RecResizeImg:
+          image_shape: [3, 32, 100]
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: 256
+    num_workers: 8

doc/doc_ch/algorithm_overview.md

@@ -40,7 +40,7 @@ PaddleOCR基于动态图开源的文本识别算法列表：
 - [x]  CRNN([paper](https://arxiv.org/abs/1507.05717))[7]（ppocr推荐）
 - [x]  Rosetta([paper](https://arxiv.org/abs/1910.05085))[10]
 - [x]  STAR-Net([paper](http://www.bmva.org/bmvc/2016/papers/paper043/index.html))[11]
-- [ ]  RARE([paper](https://arxiv.org/abs/1603.03915v1))[12] coming soon
+- [x]  RARE([paper](https://arxiv.org/abs/1603.03915v1))[12]
 - [x]  SRN([paper](https://arxiv.org/abs/2003.12294))[5]
 
 参考[DTRB][3](https://arxiv.org/abs/1904.01906)文字识别训练和评估流程，使用MJSynth和SynthText两个文字识别数据集训练，在IIIT, SVT, IC03, IC13, IC15, SVTP, CUTE数据集上进行评估，算法效果如下：
@@ -53,6 +53,9 @@ PaddleOCR基于动态图开源的文本识别算法列表：
 |CRNN|MobileNetV3|79.97%|rec_mv3_none_bilstm_ctc|[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_none_bilstm_ctc_v2.0_train.tar)|
 |StarNet|Resnet34_vd|84.44%|rec_r34_vd_tps_bilstm_ctc|[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_tps_bilstm_ctc_v2.0_train.tar)|
 |StarNet|MobileNetV3|81.42%|rec_mv3_tps_bilstm_ctc|[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_tps_bilstm_ctc_v2.0_train.tar)|
+|RARE|MobileNetV3|82.5|rec_mv3_tps_bilstm_att||[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_tps_bilstm_att_v2.0_train.tar)|
+|RARE|Resnet34_vd|83.6|rec_r34_vd_tps_bilstm_att||[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_tps_bilstm_att_v2.0_train.tar)|
 |SRN|Resnet50_vd_fpn| 88.52% | rec_r50fpn_vd_none_srn | [下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r50_vd_srn_train.tar) |
 
+
 PaddleOCR文本识别算法的训练和使用请参考文档教程中[模型训练/评估中的文本识别部分](./recognition.md)。

doc/doc_ch/recognition.md

@@ -201,6 +201,8 @@ PaddleOCR支持训练和评估交替进行, 可以在 `configs/rec/rec_icdar15_t
 | rec_mv3_none_none_ctc.yml |  Rosetta |   Mobilenet_v3 large 0.5 |  None   |  None |  ctc  |
 | rec_r34_vd_none_bilstm_ctc.yml |  CRNN |   Resnet34_vd |  None   |  BiLSTM |  ctc  |
 | rec_r34_vd_none_none_ctc.yml |  Rosetta |   Resnet34_vd |  None   |  None |  ctc  |
+| rec_mv3_tps_bilstm_att.yml |  CRNN |   Mobilenet_v3 |  TPS   |  BiLSTM |  att  |
+| rec_r34_vd_tps_bilstm_att.yml |  CRNN |   Resnet34_vd |  TPS   |  BiLSTM |  att  |
 | rec_r50fpn_vd_none_srn.yml    | SRN | Resnet50_fpn_vd    | None    | rnn | srn |
 
 训练中文数据，推荐使用[rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml)，如您希望尝试其他算法在中文数据集上的效果，请参考下列说明修改配置文件：

doc/doc_en/algorithm_overview_en.md

@@ -42,7 +42,7 @@ PaddleOCR open-source text recognition algorithms list:
 - [x]  CRNN([paper](https://arxiv.org/abs/1507.05717))[7]
 - [x]  Rosetta([paper](https://arxiv.org/abs/1910.05085))[10]
 - [x]  STAR-Net([paper](http://www.bmva.org/bmvc/2016/papers/paper043/index.html))[11]
-- [ ]  RARE([paper](https://arxiv.org/abs/1603.03915v1))[12] coming soon
+- [x]  RARE([paper](https://arxiv.org/abs/1603.03915v1))[12]
 - [x]  SRN([paper](https://arxiv.org/abs/2003.12294))[5]
 
 Refer to [DTRB](https://arxiv.org/abs/1904.01906), the training and evaluation result of these above text recognition (using MJSynth and SynthText for training, evaluate on IIIT, SVT, IC03, IC13, IC15, SVTP, CUTE) is as follow:
@@ -55,6 +55,8 @@ Refer to [DTRB](https://arxiv.org/abs/1904.01906), the training and evaluation r
 |CRNN|MobileNetV3|79.97%|rec_mv3_none_bilstm_ctc|[Download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_none_bilstm_ctc_v2.0_train.tar)|
 |StarNet|Resnet34_vd|84.44%|rec_r34_vd_tps_bilstm_ctc|[Download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_tps_bilstm_ctc_v2.0_train.tar)|
 |StarNet|MobileNetV3|81.42%|rec_mv3_tps_bilstm_ctc|[Download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_tps_bilstm_ctc_v2.0_train.tar)|
+|RARE|MobileNetV3|82.5|rec_mv3_tps_bilstm_att||[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_tps_bilstm_att_v2.0_train.tar)|
+|RARE|Resnet34_vd|83.6|rec_r34_vd_tps_bilstm_att||[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_tps_bilstm_att_v2.0_train.tar)|
 |SRN|Resnet50_vd_fpn| 88.52% | rec_r50fpn_vd_none_srn |[Download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r50_vd_srn_train.tar)|
 
 Please refer to the document for training guide and use of PaddleOCR text recognition algorithms [Text recognition model training/evaluation/prediction](./recognition_en.md)

doc/doc_en/recognition_en.md

@@ -195,8 +195,11 @@ If the evaluation set is large, the test will be time-consuming. It is recommend
 | rec_mv3_none_none_ctc.yml |  Rosetta |   Mobilenet_v3 large 0.5 |  None   |  None |  ctc  |
 | rec_r34_vd_none_bilstm_ctc.yml |  CRNN |   Resnet34_vd |  None   |  BiLSTM |  ctc  |
 | rec_r34_vd_none_none_ctc.yml |  Rosetta |   Resnet34_vd |  None   |  None |  ctc  |
+| rec_mv3_tps_bilstm_att.yml |  CRNN |   Mobilenet_v3 |  TPS   |  BiLSTM |  att  |
+| rec_r34_vd_tps_bilstm_att.yml |  CRNN |   Resnet34_vd |  TPS   |  BiLSTM |  att  |
 | rec_r50fpn_vd_none_srn.yml    | SRN | Resnet50_fpn_vd    | None    | rnn | srn |
 
+
 For training Chinese data, it is recommended to use
 [rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml). If you want to try the result of other algorithms on the Chinese data set, please refer to the following instructions to modify the configuration file:
 co

ppocr/data/imaug/label_ops.py

@@ -199,16 +199,30 @@ class AttnLabelEncode(BaseRecLabelEncode):
         super(AttnLabelEncode,
               self).__init__(max_text_length, character_dict_path,
                              character_type, use_space_char)
-        self.beg_str = "sos"
-        self.end_str = "eos"
 
     def add_special_char(self, dict_character):
-        dict_character = [self.beg_str, self.end_str] + dict_character
+        self.beg_str = "sos"
+        self.end_str = "eos"
+        dict_character = [self.beg_str] + dict_character + [self.end_str]
         return dict_character
 
-    def __call__(self, text):
+    def __call__(self, data):
+        text = data['label']
         text = self.encode(text)
-        return text
+        if text is None:
+            return None
+        if len(text) >= self.max_text_len:
+            return None
+        data['length'] = np.array(len(text))
+        text = [0] + text + [len(self.character) - 1] + [0] * (self.max_text_len
+                                                               - len(text) - 1)
+        data['label'] = np.array(text)
+        return data
+
+    def get_ignored_tokens(self):
+        beg_idx = self.get_beg_end_flag_idx("beg")
+        end_idx = self.get_beg_end_flag_idx("end")
+        return [beg_idx, end_idx]
 
     def get_beg_end_flag_idx(self, beg_or_end):
         if beg_or_end == "beg":

ppocr/losses/__init__.py

@@ -23,13 +23,15 @@ def build_loss(config):
 
     # rec loss
     from .rec_ctc_loss import CTCLoss
+    from .rec_att_loss import AttentionLoss
     from .rec_srn_loss import SRNLoss
 
     # cls loss
     from .cls_loss import ClsLoss
 
     support_dict = [
-        'DBLoss', 'EASTLoss', 'SASTLoss', 'CTCLoss', 'ClsLoss', 'SRNLoss'
+        'DBLoss', 'EASTLoss', 'SASTLoss', 'CTCLoss', 'ClsLoss', 'AttentionLoss',
+        'SRNLoss'
     ]
 
     config = copy.deepcopy(config)

ppocr/losses/rec_att_loss.py

@@ -0,0 +1,39 @@
+# copyright (c) 2021 PaddlePaddle Authors. All Rights Reserve.
+#
+# 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 absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import paddle
+from paddle import nn
+
+
+class AttentionLoss(nn.Layer):
+    def __init__(self, **kwargs):
+        super(AttentionLoss, self).__init__()
+        self.loss_func = nn.CrossEntropyLoss(weight=None, reduction='none')
+
+    def forward(self, predicts, batch):
+        targets = batch[1].astype("int64")
+        label_lengths = batch[2].astype('int64')
+        batch_size, num_steps, num_classes = predicts.shape[0], predicts.shape[
+            1], predicts.shape[2]
+        assert len(targets.shape) == len(list(predicts.shape)) - 1, \
+            "The target's shape and inputs's shape is [N, d] and [N, num_steps]"
+
+        inputs = paddle.reshape(predicts, [-1, predicts.shape[-1]])
+        targets = paddle.reshape(targets, [-1])
+
+        return {'loss': paddle.sum(self.loss_func(inputs, targets))}

ppocr/modeling/heads/__init__.py

@@ -23,12 +23,14 @@ def build_head(config):
 
     # rec head
     from .rec_ctc_head import CTCHead
+    from .rec_att_head import AttentionHead
     from .rec_srn_head import SRNHead
 
     # cls head
     from .cls_head import ClsHead
     support_dict = [
-        'DBHead', 'EASTHead', 'SASTHead', 'CTCHead', 'ClsHead', 'SRNHead'
+        'DBHead', 'EASTHead', 'SASTHead', 'CTCHead', 'ClsHead', 'AttentionHead',
+        'SRNHead'
     ]
 
     module_name = config.pop('name')

ppocr/modeling/heads/rec_att_head.py

@@ -0,0 +1,199 @@
+# copyright (c) 2021 PaddlePaddle Authors. All Rights Reserve.
+#
+# 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 absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import paddle
+import paddle.nn as nn
+import paddle.nn.functional as F
+import numpy as np
+
+
+class AttentionHead(nn.Layer):
+    def __init__(self, in_channels, out_channels, hidden_size, **kwargs):
+        super(AttentionHead, self).__init__()
+        self.input_size = in_channels
+        self.hidden_size = hidden_size
+        self.num_classes = out_channels
+
+        self.attention_cell = AttentionGRUCell(
+            in_channels, hidden_size, out_channels, use_gru=False)
+        self.generator = nn.Linear(hidden_size, out_channels)
+
+    def _char_to_onehot(self, input_char, onehot_dim):
+        input_ont_hot = F.one_hot(input_char, onehot_dim)
+        return input_ont_hot
+
+    def forward(self, inputs, targets=None, batch_max_length=25):
+        batch_size = inputs.shape[0]
+        num_steps = batch_max_length
+
+        hidden = paddle.zeros((batch_size, self.hidden_size))
+        output_hiddens = []
+
+        if targets is not None:
+            for i in range(num_steps):
+                char_onehots = self._char_to_onehot(
+                    targets[:, i], onehot_dim=self.num_classes)
+                (outputs, hidden), alpha = self.attention_cell(hidden, inputs,
+                                                               char_onehots)
+                output_hiddens.append(paddle.unsqueeze(outputs, axis=1))
+            output = paddle.concat(output_hiddens, axis=1)
+            probs = self.generator(output)
+
+        else:
+            targets = paddle.zeros(shape=[batch_size], dtype="int32")
+            probs = None
+
+            for i in range(num_steps):
+                char_onehots = self._char_to_onehot(
+                    targets, onehot_dim=self.num_classes)
+                (outputs, hidden), alpha = self.attention_cell(hidden, inputs,
+                                                               char_onehots)
+                probs_step = self.generator(outputs)
+                if probs is None:
+                    probs = paddle.unsqueeze(probs_step, axis=1)
+                else:
+                    probs = paddle.concat(
+                        [probs, paddle.unsqueeze(
+                            probs_step, axis=1)], axis=1)
+                next_input = probs_step.argmax(axis=1)
+                targets = next_input
+
+        return probs
+
+
+class AttentionGRUCell(nn.Layer):
+    def __init__(self, input_size, hidden_size, num_embeddings, use_gru=False):
+        super(AttentionGRUCell, self).__init__()
+        self.i2h = nn.Linear(input_size, hidden_size, bias_attr=False)
+        self.h2h = nn.Linear(hidden_size, hidden_size)
+        self.score = nn.Linear(hidden_size, 1, bias_attr=False)
+
+        self.rnn = nn.GRUCell(
+            input_size=input_size + num_embeddings, hidden_size=hidden_size)
+
+        self.hidden_size = hidden_size
+
+    def forward(self, prev_hidden, batch_H, char_onehots):
+
+        batch_H_proj = self.i2h(batch_H)
+        prev_hidden_proj = paddle.unsqueeze(self.h2h(prev_hidden), axis=1)
+
+        res = paddle.add(batch_H_proj, prev_hidden_proj)
+        res = paddle.tanh(res)
+        e = self.score(res)
+
+        alpha = F.softmax(e, axis=1)
+        alpha = paddle.transpose(alpha, [0, 2, 1])
+        context = paddle.squeeze(paddle.mm(alpha, batch_H), axis=1)
+        concat_context = paddle.concat([context, char_onehots], 1)
+
+        cur_hidden = self.rnn(concat_context, prev_hidden)
+
+        return cur_hidden, alpha
+
+
+class AttentionLSTM(nn.Layer):
+    def __init__(self, in_channels, out_channels, hidden_size, **kwargs):
+        super(AttentionLSTM, self).__init__()
+        self.input_size = in_channels
+        self.hidden_size = hidden_size
+        self.num_classes = out_channels
+
+        self.attention_cell = AttentionLSTMCell(
+            in_channels, hidden_size, out_channels, use_gru=False)
+        self.generator = nn.Linear(hidden_size, out_channels)
+
+    def _char_to_onehot(self, input_char, onehot_dim):
+        input_ont_hot = F.one_hot(input_char, onehot_dim)
+        return input_ont_hot
+
+    def forward(self, inputs, targets=None, batch_max_length=25):
+        batch_size = inputs.shape[0]
+        num_steps = batch_max_length
+
+        hidden = (paddle.zeros((batch_size, self.hidden_size)), paddle.zeros(
+            (batch_size, self.hidden_size)))
+        output_hiddens = []
+
+        if targets is not None:
+            for i in range(num_steps):
+                # one-hot vectors for a i-th char
+                char_onehots = self._char_to_onehot(
+                    targets[:, i], onehot_dim=self.num_classes)
+                hidden, alpha = self.attention_cell(hidden, inputs,
+                                                    char_onehots)
+
+                hidden = (hidden[1][0], hidden[1][1])
+                output_hiddens.append(paddle.unsqueeze(hidden[0], axis=1))
+            output = paddle.concat(output_hiddens, axis=1)
+            probs = self.generator(output)
+
+        else:
+            targets = paddle.zeros(shape=[batch_size], dtype="int32")
+            probs = None
+
+            for i in range(num_steps):
+                char_onehots = self._char_to_onehot(
+                    targets, onehot_dim=self.num_classes)
+                hidden, alpha = self.attention_cell(hidden, inputs,
+                                                    char_onehots)
+                probs_step = self.generator(hidden[0])
+                hidden = (hidden[1][0], hidden[1][1])
+                if probs is None:
+                    probs = paddle.unsqueeze(probs_step, axis=1)
+                else:
+                    probs = paddle.concat(
+                        [probs, paddle.unsqueeze(
+                            probs_step, axis=1)], axis=1)
+
+                next_input = probs_step.argmax(axis=1)
+
+                targets = next_input
+
+        return probs
+
+
+class AttentionLSTMCell(nn.Layer):
+    def __init__(self, input_size, hidden_size, num_embeddings, use_gru=False):
+        super(AttentionLSTMCell, self).__init__()
+        self.i2h = nn.Linear(input_size, hidden_size, bias_attr=False)
+        self.h2h = nn.Linear(hidden_size, hidden_size)
+        self.score = nn.Linear(hidden_size, 1, bias_attr=False)
+        if not use_gru:
+            self.rnn = nn.LSTMCell(
+                input_size=input_size + num_embeddings, hidden_size=hidden_size)
+        else:
+            self.rnn = nn.GRUCell(
+                input_size=input_size + num_embeddings, hidden_size=hidden_size)
+
+        self.hidden_size = hidden_size
+
+    def forward(self, prev_hidden, batch_H, char_onehots):
+        batch_H_proj = self.i2h(batch_H)
+        prev_hidden_proj = paddle.unsqueeze(self.h2h(prev_hidden[0]), axis=1)
+        res = paddle.add(batch_H_proj, prev_hidden_proj)
+        res = paddle.tanh(res)
+        e = self.score(res)
+
+        alpha = F.softmax(e, axis=1)
+        alpha = paddle.transpose(alpha, [0, 2, 1])
+        context = paddle.squeeze(paddle.mm(alpha, batch_H), axis=1)
+        concat_context = paddle.concat([context, char_onehots], 1)
+        cur_hidden = self.rnn(concat_context, prev_hidden)
+
+        return cur_hidden, alpha

ppocr/postprocess/rec_postprocess.py

@@ -135,16 +135,62 @@ class AttnLabelDecode(BaseRecLabelDecode):
                  **kwargs):
         super(AttnLabelDecode, self).__init__(character_dict_path,
                                               character_type, use_space_char)
-        self.beg_str = "sos"
-        self.end_str = "eos"
 
     def add_special_char(self, dict_character):
-        dict_character = [self.beg_str, self.end_str] + dict_character
+        self.beg_str = "sos"
+        self.end_str = "eos"
+        dict_character = dict_character
+        dict_character = [self.beg_str] + dict_character + [self.end_str]
         return dict_character
 
-    def __call__(self, text):
+    def decode(self, text_index, text_prob=None, is_remove_duplicate=False):
+        """ convert text-index into text-label. """
+        result_list = []
+        ignored_tokens = self.get_ignored_tokens()
+        [beg_idx, end_idx] = self.get_ignored_tokens()
+        batch_size = len(text_index)
+        for batch_idx in range(batch_size):
+            char_list = []
+            conf_list = []
+            for idx in range(len(text_index[batch_idx])):
+                if text_index[batch_idx][idx] in ignored_tokens:
+                    continue
+                if int(text_index[batch_idx][idx]) == int(end_idx):
+                    break
+                if is_remove_duplicate:
+                    # only for predict
+                    if idx > 0 and text_index[batch_idx][idx - 1] == text_index[
+                            batch_idx][idx]:
+                        continue
+                char_list.append(self.character[int(text_index[batch_idx][
+                    idx])])
+                if text_prob is not None:
+                    conf_list.append(text_prob[batch_idx][idx])
+                else:
+                    conf_list.append(1)
+            text = ''.join(char_list)
+            result_list.append((text, np.mean(conf_list)))
+        return result_list
+
+    def __call__(self, preds, label=None, *args, **kwargs):
+        """
         text = self.decode(text)
-        return text
+        if label is None:
+            return text
+        else:
+            label = self.decode(label, is_remove_duplicate=False)
+            return text, label
+        """
+        if isinstance(preds, paddle.Tensor):
+            preds = preds.numpy()
+
+        preds_idx = preds.argmax(axis=2)
+        preds_prob = preds.max(axis=2)
+        text = self.decode(preds_idx, preds_prob, is_remove_duplicate=False)
+        if label is None:
+            return text
+        label = self.decode(label, is_remove_duplicate=False)
+        return text, label
 
     def get_ignored_tokens(self):
         beg_idx = self.get_beg_end_flag_idx("beg")