Fix encoding format (#42)

* Fix the encoding format

demo/elmo/elmo_finetune.py

+#coding:utf-8
 import argparse
 import ast
 import io

demo/image-classification/img_classifier.py

+#coding:utf-8
 import argparse
 import os
 

demo/image-classification/predict.py

+#coding:utf-8
 import argparse
 import os
 

demo/lac/lac_demo.py

+#coding:utf-8
+from __future__ import print_function
+
+import json
 import os
+import six
+
 import paddlehub as hub
 
 if __name__ == "__main__":
@@ -12,5 +18,11 @@ if __name__ == "__main__":
     # execute predict and print the result
     results = lac.lexical_analysis(data=inputs)
     for result in results:
-        print(result['word'])
-        print(result['tag'])
+        if six.PY2:
+            print(json.dumps(
+                result['word'], encoding="utf8", ensure_ascii=False))
+            print(json.dumps(
+                result['tag'], encoding="utf8", ensure_ascii=False))
+        else:
+            print(result['word'])
+            print(result['tag'])

demo/senta/predict.py

+#coding:utf-8
 from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function

demo/senta/senta_demo.py

-# coding: utf-8
+#coding:utf-8
+from __future__ import print_function
+
+import json
 import os
+import six
+
 import paddlehub as hub
 
 if __name__ == "__main__":
@@ -11,5 +16,12 @@ if __name__ == "__main__":
     input_dict = {"text": test_text}
 
     results = senta.sentiment_classify(data=input_dict)
+
+    for index, text in enumerate(test_text):
+        results[index]["text"] = text
     for index, result in enumerate(results):
-        print(test_text[index], result['sentiment_key'])
+        if six.PY2:
+            print(json.dumps(
+                results[index], encoding="utf8", ensure_ascii=False))
+        else:
+            print(results[index])

demo/senta/senta_finetune.py

+#coding:utf-8
 import argparse
 import ast
 

demo/sequence-labeling/predict.py

+#coding:utf-8
 #   Copyright (c) 2019 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");

demo/sequence-labeling/sequence_label.py

+#coding:utf-8
 #   Copyright (c) 2019 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");

demo/ssd/ssd_demo.py

+#coding:utf-8
 import os
 import paddlehub as hub
 

demo/text-classification/predict.py

+#coding:utf-8
 #   Copyright (c) 2019 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");

demo/text-classification/simple_demo.py

+#coding:utf-8
 import paddle.fluid as fluid
 import paddlehub as hub
 

demo/text-classification/text_classifier.py

+#coding:utf-8
 #   Copyright (c) 2019 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");

paddlehub/__init__.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"
@@ -12,10 +13,13 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-#coding:utf-8
-
 import six
 
+if six.PY2:
+    import sys
+    reload(sys)
+    sys.setdefaultencoding("UTF-8")
+
 from . import module
 from . import common
 from . import io
@@ -47,8 +51,3 @@ from .finetune.config import RunConfig
 from .finetune.strategy import AdamWeightDecayStrategy
 from .finetune.strategy import DefaultStrategy
 from .finetune.strategy import DefaultFinetuneStrategy
-
-if six.PY2:
-    import sys
-    reload(sys)
-    sys.setdefaultencoding("UTF-8")

paddlehub/commands/__init__.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/commands/base_command.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/commands/clear.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/commands/cml_utils.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/commands/download.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/commands/help.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/commands/hub.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"
@@ -16,6 +17,7 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 
+import six
 import sys
 
 from paddlehub.common.logger import logger
@@ -60,4 +62,10 @@ def main():
 
 
 if __name__ == "__main__":
-    command.execute(sys.argv[1:])
+    argv = []
+    for item in sys.argv:
+        if six.PY2:
+            argv.append(item.decode(sys.stdin.encoding).decode("utf8"))
+        else:
+            argv.append(item)
+    command.execute(argv[1:])

paddlehub/commands/install.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/commands/list.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/commands/run.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"
@@ -17,6 +18,7 @@ from __future__ import division
 from __future__ import print_function
 
 import argparse
+import json
 import os
 import sys
 
@@ -172,7 +174,7 @@ class RunCommand(BaseCommand):
         results = module(
             sign_name=self.args.signature, data=input_data, **config)
         if six.PY2:
-            print(repr(results).decode('string_escape'))
+            print(json.dumps(results, encoding="utf8", ensure_ascii=False))
         else:
             print(results)
 

paddlehub/commands/search.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/commands/show.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/commands/uninstall.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/commands/version.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/common/__init__.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/common/arg_helper.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/common/dir.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/common/downloader.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/common/hub_server.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/common/logger.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/common/paddle_helper.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/common/utils.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/dataset/__init__.py

+#coding:utf-8
 #   Copyright (c) 2019 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");

paddlehub/dataset/base_cv_dataset.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/dataset/chnsenticorp.py

+#coding:utf-8
 #   Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/dataset/dataset.py

+#coding:utf-8
 #   Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/dataset/dogcat.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/dataset/flowers.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/dataset/food101.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/dataset/indoor67.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/dataset/lcqmc.py

+#coding:utf-8
 #   Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/dataset/msra_ner.py

+#coding:utf-8
 #   Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"
@@ -20,7 +21,6 @@ import os
 import codecs
 import csv
 import json
-import six
 from collections import namedtuple
 
 from paddlehub.dataset import InputExample, HubDataset

paddlehub/dataset/nlpcc_dbqa.py

+#coding:utf-8
 #   Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/dataset/stanford_dogs.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/finetune/__init__.py

+#coding:utf-8
 #   Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/finetune/checkpoint.py

+#coding:utf-8
 #   Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/finetune/checkpoint_pb2.py

+#coding:utf-8
 # Generated by the protocol buffer compiler.  DO NOT EDIT!
 # source: checkpoint.proto
 

paddlehub/finetune/config.py

+#coding:utf-8
 #   Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/finetune/evaluate.py

+#coding:utf-8
 #   Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/finetune/finetune.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/finetune/optimization.py

+#coding:utf-8
 #   Copyright (c) 2019 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");

paddlehub/finetune/regularizer.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/finetune/strategy.py

+#coding:utf-8
 #   Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/finetune/task.py

+#coding:utf-8
 #  Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/io/__init__.py

+#coding:utf-8

paddlehub/io/augmentation.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/io/parser.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/io/type.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/module/__init__.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/module/base_processor.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/module/check_info_pb2.py

+#coding:utf-8
 # Generated by the protocol buffer compiler.  DO NOT EDIT!
 # source: check_info.proto
 

paddlehub/module/checker.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/module/manager.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/module/module.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/module/module_desc_pb2.py

+#coding:utf-8
 # Generated by the protocol buffer compiler.  DO NOT EDIT!
 # source: module_desc.proto
 

paddlehub/module/signature.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/reader/__init__.py

+#coding:utf-8
 #   Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/reader/batching.py

+#coding:utf-8
 #   Copyright (c) 2019 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");

paddlehub/reader/cv_reader.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

paddlehub/reader/nlp_reader.py

+#coding:utf-8
 #   Copyright (c) 2019 PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
@@ -20,6 +21,7 @@ import csv
 import json
 import platform
 import six
+import sys
 from collections import namedtuple
 
 import paddle
@@ -31,12 +33,6 @@ from .batching import pad_batch_data
 import paddlehub as hub
 
 
-def get_encoding():
-    if platform.platform().lower().startswith("windows"):
-        return "gbk"
-    return "utf8"
-
-
 class BaseReader(object):
     def __init__(self,
                  dataset,
@@ -426,15 +422,13 @@ class LACClassifyReader(object):
         def preprocess(text):
             data_dict = {self.feed_key: [text]}
             processed = self.lac.lexical_analysis(data=data_dict)
-            for data in processed:
-                for index, word in enumerate(data['word']):
-                    if six.PY2 and type(word) == str:
-                        data['word'][index] = word.decode(get_encoding())
             processed = [
                 self.vocab[word] for word in processed[0]['word']
                 if word in self.vocab
             ]
             if len(processed) == 0:
+                if six.PY2:
+                    text = text.encode(sys.stdout.encoding)
                 logger.warning(
                     "The words in text %s can't be found in the vocabulary." %
                     (text))

paddlehub/version.py

+#coding:utf-8
 # Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"

requirements.txt

@@ -6,4 +6,3 @@ pyyaml
 numpy >= 1.12.0
 Pillow
 six >= 1.10.0
-chardet == 3.0.4

setup.py

+#coding:utf-8
 #   Copyright (c) 2019  PaddlePaddle Authors. All Rights Reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"
@@ -30,8 +31,12 @@ def python_version():
 max_version, mid_version, min_version = python_version()
 
 REQUIRED_PACKAGES = [
-    'numpy >= 1.12.0', 'six >= 1.10.0', 'protobuf >= 3.1.0', 'pyyaml', 'Pillow',
-    "visualdl >= 1.3.0", "chardet == 3.0.4"
+    'numpy >= 1.12.0',
+    'six >= 1.10.0',
+    'protobuf >= 3.1.0',
+    'pyyaml',
+    'Pillow',
+    "visualdl >= 1.3.0",
 ]
 
 if max_version < 3: