diff --git a/notebook/DL_nlp_bert_ner/img/notebook1.png b/notebook/DL_nlp_bert_ner/img/notebook1.png new file mode 100644 index 000000000..6322347c2 Binary files /dev/null and b/notebook/DL_nlp_bert_ner/img/notebook1.png differ diff --git a/notebook/DL_nlp_bert_ner/img/notebook2.png b/notebook/DL_nlp_bert_ner/img/notebook2.png new file mode 100644 index 000000000..fa11512d9 Binary files /dev/null and b/notebook/DL_nlp_bert_ner/img/notebook2.png differ diff --git a/notebook/DL_nlp_bert_ner/img/notebook3.png b/notebook/DL_nlp_bert_ner/img/notebook3.png new file mode 100644 index 000000000..33478571e Binary files /dev/null and b/notebook/DL_nlp_bert_ner/img/notebook3.png differ diff --git a/notebook/DL_nlp_bert_ner/img/notebook4.png b/notebook/DL_nlp_bert_ner/img/notebook4.png new file mode 100644 index 000000000..ce6d7677e Binary files /dev/null and b/notebook/DL_nlp_bert_ner/img/notebook4.png differ diff --git a/notebook/DL_nlp_bert_ner/ner.ipynb b/notebook/DL_nlp_bert_ner/ner.ipynb index aadb84c0e..c3747a4c1 100644 --- a/notebook/DL_nlp_bert_ner/ner.ipynb +++ b/notebook/DL_nlp_bert_ner/ner.ipynb @@ -41,19 +41,10 @@ "\n", "第二步：填写notebook所需的参数：\n", "\n", - "| 参数 | 说明 |\n", - "| - - - - - | - - - - - |\n", - "| 计费方式 | 按需计费 |\n", - "| 名称 | Notebook实例名称 |\n", - "| 工作环境 | Python3 |\n", - "| 资源池 | 选择\"公共资源池\"即可 |\n", - "| 类型 | 选择\"GPU\" |\n", - "| 规格 | 选择\"8核 | 64GiB | 1*p100\" |\n", - "| 存储配置 | 选择EVS，磁盘规格5GB |\n", + "![jupyter](./img/notebook1.png)\n", "\n", "第三步：配置好notebook参数后，点击下一步，进入notebook信息预览。确认无误后，点击“立即创建”\n", - "\n", - "![create_nb_creation_summary](./img/create_nb_creation_summary.png)\n", + "![jupyter](./img/notebook2.png)\n", "\n", "第四步：创建完成后，返回开发环境主界面，等待Notebook创建完毕后，打开Notebook，进行下一步操作。\n", "![modelarts_notebook_index](./img/modelarts_notebook_index.png)\n", @@ -62,13 +53,14 @@ "\n", "接下来，我们创建一个实际的开发环境，用于后续的实验步骤。\n", "\n", - "第一步：点击下图所示的“打开”按钮，进入刚刚创建的Notebook\n", - "![inter_dev_env](img/enter_dev_env.png)\n", + "第一步：点击下图所示的“启动”按钮，加载后“打开”按钮变从灰色变为蓝色后点击“打开”进入刚刚创建的Notebook\n", + "![jupyter](./img/notebook3.png)\n", + "![jupyter](./img/notebook4.png)\n", "\n", - "第二步：创建一个Python3环境的的Notebook。点击右上角的\"New\"，然后创建TensorFlow 1.13.1开发环境。\n", "\n", - "第三步：点击左上方的文件名\"Untitled\"，并输入一个与本实验相关的名称\n", + "第二步：创建一个Python3环境的的Notebook。点击右上角的\"New\"，然后选择TensorFlow 1.13.1开发环境。\n", "\n", + "第三步：点击左上方的文件名\"Untitled\"，并输入一个与本实验相关的名称，如\"ner\"\n", "![notebook_untitled_filename](./img/notebook_untitled_filename.png)\n", "![notebook_name_the_ipynb](./img/notebook_name_the_ipynb.png)\n", "\n", @@ -96,14 +88,18 @@ "name": "stdout", "output_type": "stream", "text": [ - "Successfully download file modelarts-labs/notebook/DL_nlp_ner/ner.tar.gz from OBS to local ./ner.tar.gz\n", - "total 375220\r\n", - "drwxrwsrwx 4 ma-user ma-group 4096 Sep 6 13:34 .\r\n", - "drwsrwsr-x 22 ma-user ma-group 4096 Sep 6 13:03 ..\r\n", - "drwxr-s--- 2 ma-user ma-group 4096 Sep 6 13:33 .ipynb_checkpoints\r\n", - "-rw-r----- 1 ma-user ma-group 45114 Sep 6 13:33 ner.ipynb\r\n", - "-rw-r----- 1 ma-user ma-group 384157325 Sep 6 13:35 ner.tar.gz\r\n", - "drwx--S--- 2 ma-user ma-group 4096 Sep 6 13:03 .Trash-1000\r\n" + "Successfully download file modelarts-labs-bj4/notebook/DL_nlp_ner/ner.tar.gz from OBS to local ./ner.tar.gz\n", + "total 375356\r\n", + "drwxr-x--- 5 ma-user ma-group 4096 Nov 19 18:03 .\r\n", + "drwxrwxrwx 3 ma-user ma-group 4096 Nov 19 17:25 ..\r\n", + "drwxr-x--- 2 ma-user ma-group 4096 Nov 19 17:15 img\r\n", + "drwxr-x--- 2 ma-user ma-group 4096 Nov 19 18:00 .ipynb_checkpoints\r\n", + "drwxr-x--- 7 ma-user ma-group 4096 Sep 9 2019 ner\r\n", + "-rw-r----- 1 ma-user ma-group 56192 Nov 19 18:02 ner.ipynb\r\n", + "-rw-r----- 1 ma-user ma-group 384141687 Nov 19 18:03 ner.tar.gz\r\n", + "-rw-r----- 1 ma-user ma-group 68329 Nov 19 18:00 nlp_ner-Copy1.ipynb\r\n", + "-rw-r----- 1 ma-user ma-group 68329 Nov 19 18:00 nlp_ner.ipynb\r\n", + "-rw-r----- 1 ma-user ma-group 431 Nov 18 10:18 README.md\r\n" ] } ], @@ -140,13 +136,16 @@ "name": "stdout", "output_type": "stream", "text": [ - "total 68\r\n", - "drwxrwsrwx 5 ma-user ma-group 4096 Sep 6 13:35 .\r\n", - "drwsrwsr-x 22 ma-user ma-group 4096 Sep 6 13:03 ..\r\n", - "drwxr-s--- 2 ma-user ma-group 4096 Sep 6 13:33 .ipynb_checkpoints\r\n", - "drwxr-s--- 8 ma-user ma-group 4096 Sep 6 00:24 ner\r\n", - "-rw-r----- 1 ma-user ma-group 45114 Sep 6 13:33 ner.ipynb\r\n", - "drwx--S--- 2 ma-user ma-group 4096 Sep 6 13:03 .Trash-1000\r\n" + "total 216\r\n", + "drwxr-x--- 5 ma-user ma-group 4096 Nov 19 18:03 .\r\n", + "drwxrwxrwx 3 ma-user ma-group 4096 Nov 19 17:25 ..\r\n", + "drwxr-x--- 2 ma-user ma-group 4096 Nov 19 17:15 img\r\n", + "drwxr-x--- 2 ma-user ma-group 4096 Nov 19 18:00 .ipynb_checkpoints\r\n", + "drwxr-x--- 7 ma-user ma-group 4096 Sep 9 2019 ner\r\n", + "-rw-r----- 1 ma-user ma-group 56192 Nov 19 18:02 ner.ipynb\r\n", + "-rw-r----- 1 ma-user ma-group 68329 Nov 19 18:00 nlp_ner-Copy1.ipynb\r\n", + "-rw-r----- 1 ma-user ma-group 68329 Nov 19 18:00 nlp_ner.ipynb\r\n", + "-rw-r----- 1 ma-user ma-group 431 Nov 18 10:18 README.md\r\n" ] } ], @@ -316,43 +315,64 @@ "name": "stdout", "output_type": "stream", "text": [ - "Requirement already satisfied: tensorflow==1.11.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages\n", - "Requirement already satisfied: absl-py>=0.1.6 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow==1.11.0)\n", - "Requirement already satisfied: wheel>=0.26 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow==1.11.0)\n", - "Requirement already satisfied: gast>=0.2.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow==1.11.0)\n", - "Requirement already satisfied: six>=1.10.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow==1.11.0)\n", - "Requirement already satisfied: setuptools<=39.1.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow==1.11.0)\n", + "Collecting tensorflow==1.11.0\n", + " Downloading http://repo.myhuaweicloud.com/repository/pypi/packages/ce/d5/38cd4543401708e64c9ee6afa664b936860f4630dd93a49ab863f9998cd2/tensorflow-1.11.0-cp36-cp36m-manylinux1_x86_64.whl (63.0MB)\n", + "\u001b[K 100% |████████████████████████████████| 63.0MB 21.2MB/s ta 0:00:011��███▏ | 22.0MB 105.8MB/s eta 0:00:01\n", + "\u001b[?25hRequirement already satisfied: six>=1.10.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow==1.11.0)\n", "Requirement already satisfied: termcolor>=1.1.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow==1.11.0)\n", - "Requirement already satisfied: numpy>=1.13.3 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow==1.11.0)\n", - "Requirement already satisfied: tensorboard<1.12.0,>=1.11.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow==1.11.0)\n", - "Requirement already satisfied: grpcio>=1.8.6 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow==1.11.0)\n", + "Requirement already satisfied: absl-py>=0.1.6 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow==1.11.0)\n", "Requirement already satisfied: keras-applications>=1.0.5 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow==1.11.0)\n", + "Requirement already satisfied: gast>=0.2.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow==1.11.0)\n", "Requirement already satisfied: keras-preprocessing>=1.0.3 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow==1.11.0)\n", "Requirement already satisfied: protobuf>=3.6.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow==1.11.0)\n", "Requirement already satisfied: astor>=0.6.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow==1.11.0)\n", + "Collecting setuptools<=39.1.0 (from tensorflow==1.11.0)\n", + " Downloading http://repo.myhuaweicloud.com/repository/pypi/packages/8c/10/79282747f9169f21c053c562a0baa21815a8c7879be97abd930dbcf862e8/setuptools-39.1.0-py2.py3-none-any.whl (566kB)\n", + "\u001b[K 100% |████████████████████████████████| 573kB 39.9MB/s ta 0:00:01\n", + "\u001b[?25hRequirement already satisfied: wheel>=0.26 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow==1.11.0)\n", + "Collecting tensorboard<1.12.0,>=1.11.0 (from tensorflow==1.11.0)\n", + " Downloading http://repo.myhuaweicloud.com/repository/pypi/packages/9b/2f/4d788919b1feef04624d63ed6ea45a49d1d1c834199ec50716edb5d310f4/tensorboard-1.11.0-py3-none-any.whl (3.0MB)\n", + "\u001b[K 100% |████████████████████████████████| 3.0MB 44.7MB/s ta 0:00:01\n", + "\u001b[?25hRequirement already satisfied: numpy>=1.13.3 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow==1.11.0)\n", + "Requirement already satisfied: grpcio>=1.8.6 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow==1.11.0)\n", + "Requirement already satisfied: h5py in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from keras-applications>=1.0.5->tensorflow==1.11.0)\n", "Requirement already satisfied: markdown>=2.6.8 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorboard<1.12.0,>=1.11.0->tensorflow==1.11.0)\n", "Requirement already satisfied: werkzeug>=0.11.10 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorboard<1.12.0,>=1.11.0->tensorflow==1.11.0)\n", - "Requirement already satisfied: h5py in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from keras-applications>=1.0.5->tensorflow==1.11.0)\n", - "\u001b[33mYou are using pip version 9.0.1, however version 19.2.3 is available.\n", + "Installing collected packages: setuptools, tensorboard, tensorflow\n", + " Found existing installation: setuptools 41.0.0\n", + " Uninstalling setuptools-41.0.0:\n", + " Successfully uninstalled setuptools-41.0.0\n", + " Found existing installation: tensorboard 1.13.1\n", + " Uninstalling tensorboard-1.13.1:\n", + " Successfully uninstalled tensorboard-1.13.1\n", + "Successfully installed setuptools-39.1.0 tensorboard-1.11.0 tensorflow-1.11.0\n", + "\u001b[33mYou are using pip version 9.0.1, however version 20.2.4 is available.\n", "You should consider upgrading via the 'pip install --upgrade pip' command.\u001b[0m\n", - "Requirement already satisfied: tensorflow-gpu==1.11.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages\n", - "Requirement already satisfied: grpcio>=1.8.6 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", - "Requirement already satisfied: numpy>=1.13.3 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", - "Requirement already satisfied: absl-py>=0.1.6 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", - "Requirement already satisfied: keras-applications>=1.0.5 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", + "Collecting tensorflow-gpu==1.11.0\n", + " Downloading http://repo.myhuaweicloud.com/repository/pypi/packages/25/52/01438b81806765936eee690709edc2a975472c4e9d8d465a01840869c691/tensorflow_gpu-1.11.0-cp36-cp36m-manylinux1_x86_64.whl (258.8MB)\n", + "\u001b[K 100% |████████████████████████████████| 258.8MB 101.2MB/s ta 0:00:01MB/s eta 0:00:02 | 19.9MB 120.7MB/s eta 0:00:02 | 24.0MB 90.4MB/s eta 0:00:03 | 36.3MB 114.7MB/s eta 0:00:026% |█████▎ | 42.6MB 117.9MB/s eta 0:00:02��██▊ | 46.4MB 87.6MB/s eta 0:00:03 | 51.5MB 133.7MB/s eta 0:00:02████▏ | 57.6MB 108.7MB/s eta 0:00:0224% |███████▉ | 63.2MB 126.7MB/s eta 0:00:02020:02██████ | 250.0MB 102.2MB/s eta 0:00:01\n", + "\u001b[?25hRequirement already satisfied: keras-preprocessing>=1.0.3 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", + "Requirement already satisfied: wheel>=0.26 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", "Requirement already satisfied: setuptools<=39.1.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", - "Requirement already satisfied: tensorboard<1.12.0,>=1.11.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", - "Requirement already satisfied: termcolor>=1.1.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", - "Requirement already satisfied: gast>=0.2.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", "Requirement already satisfied: six>=1.10.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", + "Requirement already satisfied: numpy>=1.13.3 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", + "Requirement already satisfied: grpcio>=1.8.6 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", + "Requirement already satisfied: termcolor>=1.1.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", + "Requirement already satisfied: absl-py>=0.1.6 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", "Requirement already satisfied: astor>=0.6.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", - "Requirement already satisfied: wheel>=0.26 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", - "Requirement already satisfied: keras-preprocessing>=1.0.3 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", + "Requirement already satisfied: gast>=0.2.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", "Requirement already satisfied: protobuf>=3.6.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", + "Requirement already satisfied: keras-applications>=1.0.5 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", + "Requirement already satisfied: tensorboard<1.12.0,>=1.11.0 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorflow-gpu==1.11.0)\n", "Requirement already satisfied: h5py in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from keras-applications>=1.0.5->tensorflow-gpu==1.11.0)\n", "Requirement already satisfied: werkzeug>=0.11.10 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorboard<1.12.0,>=1.11.0->tensorflow-gpu==1.11.0)\n", "Requirement already satisfied: markdown>=2.6.8 in /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages (from tensorboard<1.12.0,>=1.11.0->tensorflow-gpu==1.11.0)\n", - "\u001b[33mYou are using pip version 9.0.1, however version 19.2.3 is available.\n", + "Installing collected packages: tensorflow-gpu\n", + " Found existing installation: tensorflow-gpu 1.13.1\n", + " Uninstalling tensorflow-gpu-1.13.1:\n", + " Successfully uninstalled tensorflow-gpu-1.13.1\n", + "Successfully installed tensorflow-gpu-1.11.0\n", + "\u001b[33mYou are using pip version 9.0.1, however version 20.2.4 is available.\n", "You should consider upgrading via the 'pip install --upgrade pip' command.\u001b[0m\n" ] } @@ -374,7 +394,26 @@ "cell_type": "code", "execution_count": 4, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:523: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:524: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:525: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:526: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:527: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:532: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\n" + ] + } + ], "source": [ "import os\n", "import json\n", @@ -752,91 +791,14 @@ "INFO:tensorflow: Batch size = 64\n", "INFO:tensorflow: Num steps = 1630\n", "INFO:tensorflow:Using config: {'_model_dir': './ner/output', '_tf_random_seed': None, '_save_summary_steps': 1000, '_save_checkpoints_steps': 1000, '_save_checkpoints_secs': None, '_session_config': allow_soft_placement: true\n", - ", '_keep_checkpoint_max': 5, '_keep_checkpoint_every_n_hours': 10000, '_log_step_count_steps': 100, '_train_distribute': None, '_device_fn': None, '_protocol': None, '_eval_distribute': None, '_experimental_distribute': None, '_service': None, '_cluster_spec': , '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1}\n", - "INFO:tensorflow:Calling model_fn.\n", - "INFO:tensorflow:*** Features ***\n", - "INFO:tensorflow: name = input_ids, shape = (32, 128)\n", - "INFO:tensorflow: name = input_mask, shape = (32, 128)\n", - "INFO:tensorflow: name = label_ids, shape = (32, 128)\n", - "INFO:tensorflow: name = segment_ids, shape = (32, 128)\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "shape of input_ids (32, 128)\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "INFO:tensorflow:Done calling model_fn.\n", - "INFO:tensorflow:Create CheckpointSaverHook.\n", - "INFO:tensorflow:Graph was finalized.\n", - "INFO:tensorflow:Running local_init_op.\n", - "INFO:tensorflow:Done running local_init_op.\n", - "INFO:tensorflow:Saving checkpoints for 0 into ./ner/output/model.ckpt.\n", - "INFO:tensorflow:loss = 159.27417, step = 0\n", - "INFO:tensorflow:global_steps = 0, loss = 159.27417\n", - "INFO:tensorflow:global_step/sec: 1.43142\n", - "INFO:tensorflow:loss = 52.92234, step = 100 (69.862 sec)\n", - "INFO:tensorflow:global_steps = 100, loss = 52.92234 (69.862 sec)\n", - "INFO:tensorflow:global_step/sec: 1.81746\n", - "INFO:tensorflow:loss = 45.81129, step = 200 (55.022 sec)\n", - "INFO:tensorflow:global_steps = 200, loss = 45.81129 (55.022 sec)\n", - "INFO:tensorflow:global_step/sec: 1.82186\n", - "INFO:tensorflow:loss = 48.826424, step = 300 (54.890 sec)\n", - "INFO:tensorflow:global_steps = 300, loss = 48.826424 (54.890 sec)\n", - "INFO:tensorflow:global_step/sec: 1.82109\n", - "INFO:tensorflow:loss = 44.61993, step = 400 (54.910 sec)\n", - "INFO:tensorflow:global_steps = 400, loss = 44.61993 (54.911 sec)\n", - "INFO:tensorflow:global_step/sec: 1.82003\n", - "INFO:tensorflow:loss = 44.479523, step = 500 (54.944 sec)\n", - "INFO:tensorflow:global_steps = 500, loss = 44.479523 (54.944 sec)\n", - "INFO:tensorflow:global_step/sec: 1.81887\n", - "INFO:tensorflow:loss = 43.29698, step = 600 (54.979 sec)\n", - "INFO:tensorflow:global_steps = 600, loss = 43.29698 (54.981 sec)\n", - "INFO:tensorflow:global_step/sec: 1.82034\n", - "INFO:tensorflow:loss = 52.886993, step = 700 (54.938 sec)\n", - "INFO:tensorflow:global_steps = 700, loss = 52.886993 (54.936 sec)\n", - "INFO:tensorflow:global_step/sec: 1.81812\n", - "INFO:tensorflow:loss = 46.30081, step = 800 (54.999 sec)\n", - "INFO:tensorflow:global_steps = 800, loss = 46.30081 (54.999 sec)\n", - "INFO:tensorflow:global_step/sec: 1.824\n", - "INFO:tensorflow:loss = 47.800972, step = 900 (54.824 sec)\n", - "INFO:tensorflow:global_steps = 900, loss = 47.800972 (54.824 sec)\n", - "INFO:tensorflow:Saving checkpoints for 1000 into ./ner/output/model.ckpt.\n", - "INFO:tensorflow:global_step/sec: 1.53438\n", - "INFO:tensorflow:loss = 45.00616, step = 1000 (65.173 sec)\n", - "INFO:tensorflow:global_steps = 1000, loss = 45.00616 (65.173 sec)\n", - "INFO:tensorflow:global_step/sec: 1.81754\n", - "INFO:tensorflow:loss = 41.210243, step = 1100 (55.020 sec)\n", - "INFO:tensorflow:global_steps = 1100, loss = 41.210243 (55.020 sec)\n", - "INFO:tensorflow:global_step/sec: 1.81626\n", - "INFO:tensorflow:loss = 48.78985, step = 1200 (55.058 sec)\n", - "INFO:tensorflow:global_steps = 1200, loss = 48.78985 (55.058 sec)\n", - "INFO:tensorflow:global_step/sec: 1.81829\n", - "INFO:tensorflow:loss = 48.236824, step = 1300 (54.997 sec)\n", - "INFO:tensorflow:global_steps = 1300, loss = 48.236824 (55.000 sec)\n", - "INFO:tensorflow:global_step/sec: 1.81966\n", - "INFO:tensorflow:loss = 49.241657, step = 1400 (54.956 sec)\n", - "INFO:tensorflow:global_steps = 1400, loss = 49.241657 (54.952 sec)\n", - "INFO:tensorflow:global_step/sec: 1.81475\n", - "INFO:tensorflow:loss = 49.225563, step = 1500 (55.104 sec)\n", - "INFO:tensorflow:global_steps = 1500, loss = 49.225563 (55.104 sec)\n", - "INFO:tensorflow:global_step/sec: 1.81869\n", - "INFO:tensorflow:loss = 43.25946, step = 1600 (54.988 sec)\n", - "INFO:tensorflow:global_steps = 1600, loss = 43.25946 (54.988 sec)\n", - "INFO:tensorflow:Saving checkpoints for 1630 into ./ner/output/model.ckpt.\n", - "INFO:tensorflow:Loss for final step: 50.242386.\n" + ", '_keep_checkpoint_max': 5, '_keep_checkpoint_every_n_hours': 10000, '_log_step_count_steps': 100, '_train_distribute': None, '_device_fn': None, '_protocol': None, '_eval_distribute': None, '_experimental_distribute': None, '_service': None, '_cluster_spec': , '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1}\n", + "INFO:tensorflow:Skipping training since max_steps has already saved.\n" ] }, { "data": { "text/plain": [ - "" + "" ] }, "execution_count": 10, @@ -952,18 +914,18 @@ "output_type": "stream", "text": [ "INFO:tensorflow:Done calling model_fn.\n", - "INFO:tensorflow:Starting evaluation at 2019-09-06-05:53:54\n", + "INFO:tensorflow:Starting evaluation at 2020-11-19-10:44:30\n", "INFO:tensorflow:Graph was finalized.\n", "INFO:tensorflow:Restoring parameters from ./ner/output/model.ckpt-1630\n", "INFO:tensorflow:Running local_init_op.\n", "INFO:tensorflow:Done running local_init_op.\n", - "INFO:tensorflow:Finished evaluation at 2019-09-06-05:54:35\n", - "INFO:tensorflow:Saving dict for global step 1630: eval_loss = 0.042222254, global_step = 1630, loss = 35.88692\n", + "INFO:tensorflow:Finished evaluation at 2020-11-19-10:45:01\n", + "INFO:tensorflow:Saving dict for global step 1630: eval_loss = 0.1202376, global_step = 1630, loss = 34.488083\n", "INFO:tensorflow:Saving 'checkpoint_path' summary for global step 1630: ./ner/output/model.ckpt-1630\n", "INFO:tensorflow:***** Eval results *****\n", - "INFO:tensorflow: eval_loss = 0.042222254\n", + "INFO:tensorflow: eval_loss = 0.1202376\n", "INFO:tensorflow: global_step = 1630\n", - "INFO:tensorflow: loss = 35.88692\n" + "INFO:tensorflow: loss = 34.488083\n" ] } ], @@ -1041,18 +1003,18 @@ "output_type": "stream", "text": [ "INFO:tensorflow:Done calling model_fn.\n", - "INFO:tensorflow:Starting evaluation at 2019-09-06-05:55:20\n", + "INFO:tensorflow:Starting evaluation at 2020-11-19-10:45:05\n", "INFO:tensorflow:Graph was finalized.\n", "INFO:tensorflow:Restoring parameters from ./ner/output/model.ckpt-1630\n", "INFO:tensorflow:Running local_init_op.\n", "INFO:tensorflow:Done running local_init_op.\n", - "INFO:tensorflow:Finished evaluation at 2019-09-06-05:55:24\n", - "INFO:tensorflow:Saving dict for global step 1630: eval_loss = 0.017003676, global_step = 1630, loss = 34.176826\n", + "INFO:tensorflow:Finished evaluation at 2020-11-19-10:45:07\n", + "INFO:tensorflow:Saving dict for global step 1630: eval_loss = 0.053423714, global_step = 1630, loss = 33.130444\n", "INFO:tensorflow:Saving 'checkpoint_path' summary for global step 1630: ./ner/output/model.ckpt-1630\n", "INFO:tensorflow:***** Predict results *****\n", - "INFO:tensorflow: eval_loss = 0.017003676\n", + "INFO:tensorflow: eval_loss = 0.053423714\n", "INFO:tensorflow: global_step = 1630\n", - "INFO:tensorflow: loss = 34.176826\n", + "INFO:tensorflow: loss = 33.130444\n", "INFO:tensorflow:Calling model_fn.\n", "INFO:tensorflow:*** Features ***\n", "INFO:tensorflow: name = input_ids, shape = (?, 128)\n", @@ -1076,22 +1038,202 @@ "INFO:tensorflow:Graph was finalized.\n", "INFO:tensorflow:Restoring parameters from ./ner/output/model.ckpt-1630\n", "INFO:tensorflow:Running local_init_op.\n", - "INFO:tensorflow:Done running local_init_op.\n" + "INFO:tensorflow:Done running local_init_op.\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:在 香 港 回 归 前 的 最 后 阶 段 ， 中 共 中 央 举 办 《 “ 一 国 两 制 ” 与 香 港 基 本 法 》 讲 座 ， 中 央 领 导 同 志 认 真 听 讲 ， 虚 心 学 习 ， 很 有 意 义 。\n", + "INFO:tensorflow:O B-LOC I-LOC O O O O O O O O O B-ORG I-ORG I-ORG I-ORG O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:这 表 明 ， 以 江 泽 民 同 志 为 核 心 的 党 中 央 坚 定 不 移 地 贯 彻 邓 小 平 同 志 “ 一 国 两 制 ” 的 伟 大 构 想 ， 不 折 不 扣 地 执 行 基 本 法 。\n", + "INFO:tensorflow:O O O O O B-PER I-PER I-PER O O O O O O B-ORG I-ORG I-ORG O O O O O O O B-PER I-PER I-PER O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:“ 一 国 两 制 ” 是 邓 小 平 同 志 的 一 个 伟 大 构 想 ， 《 中 华 人 民 共 和 国 香 港 特 别 行 政 区 基 本 法 》 是 贯 彻 落 实 “ 一 国 两 制 ” 伟 大 构 想 的 一 部 全 国 性 法 律 ， 是 一 部 有 鲜 明 中 国 特 色 的 法 律 。\n", + "INFO:tensorflow:O O O O O O O B-PER I-PER I-PER O O O O O O O O O O O B-LOC I-LOC I-LOC I-LOC I-LOC I-LOC I-LOC B-LOC I-LOC I-LOC I-LOC I-LOC I-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:看 包 公 断 案 的 戏 ， 看 他 威 风 凛 凛 坐 公 堂 拍 桌 子 动 刑 具 ， 多 少 还 有 一 点 担 心 ， 总 怕 靠 这 一 套 办 法 弄 出 错 案 来 ， 放 过 了 真 正 的 坏 人 ；\n", + "INFO:tensorflow:O B-PER I-PER O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:可 看 《 包 公 赶 驴 》 这 出 戏 ， 心 里 就 很 踏 实 ： 这 样 是 一 断 一 个 准 的 。\n", + "INFO:tensorflow:O O O B-PER I-PER O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:譬 如 看 《 施 公 案 》 ， 施 大 人 坐 公 堂 问 案 子 不 得 要 领 ， 总 是 扮 成 普 通 百 姓 深 入 民 间 暗 中 查 访 ， 结 果 就 屡 破 奇 案 了 。\n", + "INFO:tensorflow:O O O O B-PER O O O O B-PER O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:如 果 有 人 问 我 ： “ 你 看 过 许 多 包 公 戏 ， 哪 一 出 最 好 ？ ”\n", + "INFO:tensorflow:O O O O O O O O O O O O O B-PER I-PER O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:我 要 毫 不 犹 豫 地 回 答 道 ： “ 自 然 是 《 包 公 赶 驴 》 啦 ！\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O B-PER I-PER O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:学 习 基 本 法 顺 利 迎 回 归\n", + "INFO:tensorflow:O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:本 报 评 论 员\n", + "INFO:tensorflow:O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:再 过 5 5 天 ， 我 国 政 府 将 对 香 港 恢 复 行 使 主 权 。\n", + "INFO:tensorflow:O O O O O O O O O O O O B-LOC I-LOC O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:它 把 中 央 对 解 决 香 港 问 题 的 基 本 方 针 政 策 具 体 化 、 法 律 化 ， 成 为 国 家 意 志 。\n", + "INFO:tensorflow:O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:学 习 基 本 法 ， 顺 利 迎 回 归 ， 是 一 项 迫 切 的 任 务 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:要 学 好 基 本 法 ， 首 先 要 认 识 到 基 本 法 的 意 义 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:说 国 际 意 义 ， 不 只 对 第 三 世 界 ， 而 且 对 全 人 类 都 具 有 长 远 意 义 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:这 是 一 个 具 有 创 造 性 的 杰 作 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:“ 基 本 法 不 仅 为 确 保 香 港 平 稳 过 渡 发 挥 重 要 作 用 ， 也 为 确 保 香 港 长 期 繁 荣 稳 定 发 挥 重 要 作 用 ；\n", + "INFO:tensorflow:O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:不 仅 为 当 前 解 决 香 港 问 题 发 挥 作 用 ， 也 为 在 不 远 的 将 来 解 决 澳 门 问 题 和 最 终 解 决 台 湾 问 题 ， 实 现 祖 国 完 全 统 一 发 挥 重 要 作 用 。\n", + "INFO:tensorflow:O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:基 本 法 的 主 要 特 征 ， 是 把 “ 一 国 ” 与 “ 两 制 ” 紧 密 结 合 ， 维 护 国 家 的 主 权 、 统 一 和 领 土 完 整 与 授 权 香 港 特 别 行 政 区 实 行 高 度 自 治 紧 密 结 合 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O B-LOC I-LOC I-LOC I-LOC I-LOC I-LOC I-LOC O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:在 一 个 统 一 的 中 华 人 民 共 和 国 ， 可 以 实 行 社 会 主 义 和 资 本 主 义 两 种 制 度 ， 这 是 为 了 民 族 、 国 家 的 根 本 利 益 。\n", + "INFO:tensorflow:O O O O O O B-LOC I-LOC I-LOC I-LOC I-LOC I-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:只 有 认 真 学 习 ， 才 能 理 解 意 义 ， 认 识 特 征 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:制 定 一 部 好 法 律 ， 很 不 容 易 ；\n", + "INFO:tensorflow:O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:遵 守 法 律 ， 执 行 法 律 ， 也 很 不 容 易 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:必 须 重 申 ， 有 法 必 依 ， 执 法 必 严 ， 违 法 必 究 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:基 本 法 作 为 一 部 全 国 性 的 法 律 ， 不 仅 香 港 要 严 格 遵 守 ， 各 省 、 自 治 区 、 直 辖 市 都 要 严 格 遵 守 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:从 中 共 中 央 举 办 这 个 讲 座 ， 可 以 看 出 ， 党 和 政 府 正 在 努 力 加 强 法 制 建 设 ， 坚 持 依 法 治 国 。\n", + "INFO:tensorflow:O B-ORG I-ORG I-ORG I-ORG O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:有 了 法 律 ， 有 了 制 度 ， 就 有 了 保 证 ， 就 使 “ 一 国 两 制 ” 的 伟 大 构 想 以 法 律 的 形 式 固 定 下 来 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:全 国 人 民 特 别 是 香 港 同 胞 也 从 中 再 一 次 看 到 ， 中 国 共 产 党 和 人 民 政 府 是 高 度 负 责 任 的 党 和 政 府 ， 一 切 从 人 民 的 利 益 出 发 ， 一 切 为 了 祖 国 的 繁 荣 富 强 ， 香 港 的 明 天 将 更 美 好 。\n", + "INFO:tensorflow:O O O O O O O B-LOC I-LOC O O O O O O O O O O O B-ORG I-ORG I-ORG I-ORG I-ORG O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:学 习 基 本 法 ， 中 央 领 导 带 了 个 好 头 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:全 党 和 全 国 人 民 特 别 是 各 级 党 政 领 导 干 部 ， 都 要 重 视 学 习 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:只 有 学 习 好 ， 才 能 贯 彻 好 。\n", + "INFO:tensorflow:O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:为 了 迎 接 香 港 顺 利 回 归 祖 国 这 一 中 华 民 族 的 盛 事 ， 首 先 要 有 一 个 扎 实 的 思 想 准 备 和 良 好 的 精 神 状 态 。\n", + "INFO:tensorflow:O O O O B-LOC I-LOC O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:基 本 法 连 着 你 我 他\n", + "INFO:tensorflow:O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:叶 秋\n", + "INFO:tensorflow:B-PER I-PER\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:赠 书 想 来 是 香 港 同 胞 的 一 种 文 明 礼 仪 。\n", + "INFO:tensorflow:O O O O O B-LOC I-LOC O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:抵 港 仅 数 日 ， 就 收 到 厚 厚 几 摞 书 。\n", + "INFO:tensorflow:O B-LOC O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:匆 匆 翻 阅 一 遍 ， 发 现 各 种 版 本 的 《 中 华 人 民 共 和 国 香 港 特 别 行 政 区 基 本 法 》 竟 有 六 册 之 多 ， 推 介 普 及 基 本 法 的 书 籍 还 要 多 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O B-LOC I-LOC I-LOC I-LOC I-LOC I-LOC I-LOC B-LOC I-LOC I-LOC I-LOC I-LOC I-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:应 约 去 湾 仔 道 谈 事 ， 路 过 一 个 名 为 “ 艺 美 ” 的 书 店 ， 看 到 摆 放 在 最 抢 眼 位 置 的 也 是 基 本 法 及 其 推 介 图 书 。\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:tensorflow:O O O B-LOC I-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:由 此 可 见 ， 在 法 制 观 念 很 强 的 港 人 心 目 中 ， 基 本 法 具 有 极 大 的 权 威 和 尊 严 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O B-LOC O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:行 政 官 员 表 示 ： “ 香 港 继 续 繁 荣 稳 定 、 实 现 香 港 梦 的 成 功 要 素 ， 在 基 本 法 中 得 到 了 充 分 保 证 。 ”\n", + "INFO:tensorflow:O O O O O O O O B-LOC I-LOC O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:法 律 界 人 士 认 为 ： “ 法 治 精 神 能 否 继 续 保 持 ， 基 本 法 已 作 了 明 确 规 定 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:只 要 恪 守 广 大 港 人 认 受 的 香 港 法 律 体 系 中 的 这 个 总 纲 纪 、 总 章 程 ， 香 港 将 健 步 迈 向 新 世 纪 。 ”\n", + "INFO:tensorflow:O O O O O O B-LOC O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:劳 工 界 的 成 员 说 ， 涉 及 保 障 劳 工 合 法 权 益 的 条 款 ， “ 香 港 现 在 有 的 ， 基 本 法 都 保 持 了 ；\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:大 家 因 此 吃 了 定 心 丸 。 ”\n", + "INFO:tensorflow:O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:基 本 法 受 到 港 人 的 普 遍 欢 迎 和 高 度 重 视 是 势 所 必 然 。\n", + "INFO:tensorflow:O O O O O B-LOC O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:历 时 四 年 零 八 个 月 、 凝 聚 了 香 港 和 内 地 无 数 人 的 智 慧 而 制 定 的 基 本 法 ， 将 邓 小 平 同 志 倡 导 的 “ 一 国 两 制 ” 伟 大 构 想 以 法 律 形 式 固 定 下 来 ， 成 为 国 家 和 人 民 的 意 志 。\n", + "INFO:tensorflow:O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O B-PER I-PER I-PER O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:基 本 法 既 是 香 港 回 归 后 特 区 一 切 运 作 的 法 律 基 础 ， 更 是 保 持 香 港 长 期 稳 定 繁 荣 的 法 律 保 证 。\n", + "INFO:tensorflow:O O O O O B-LOC I-LOC O O O B-LOC I-LOC O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:实 践 已 经 并 将 继 续 证 明 这 一 点 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:说 来 也 巧 ， 姬 鹏 飞 同 志 1 9 9 0 年 4 月 在 邓 小 平 同 志 题 写 书 名 的 《 基 本 法 的 诞 生 》 一 书 序 言 中 也 写 了 同 样 的 话 。\n", + "INFO:tensorflow:O O O O O B-PER I-PER I-PER O O O O O O O O O O B-PER I-PER I-PER O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:真 可 谓 仁 者 智 者 所 见 略 同 。\n", + "INFO:tensorflow:O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:基 本 法 是 一 部 具 有 普 遍 约 束 力 的 重 要 法 律 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:7 月 1 日 ， 这 部 重 要 法 律 即 开 始 正 式 实 施 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:基 本 法 不 仅 体 现 了 香 港 同 胞 的 意 志 和 利 益 ， 也 体 现 了 全 国 人 民 的 意 志 和 利 益 。\n", + "INFO:tensorflow:O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:正 因 为 如 此 ， 江 泽 民 同 志 强 调 ： 香 港 基 本 法 是 一 部 全 国 性 的 法 律 ， 不 仅 香 港 要 严 格 遵 守 ， 各 省 、 自 治 区 、 直 辖 市 都 要 严 格 遵 守 。\n", + "INFO:tensorflow:O O O O O O B-PER I-PER I-PER O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:还 表 示 ， 不 仅 我 要 遵 守 ， 我 希 望 香 港 同 胞 和 全 国 1 2 亿 人 民 也 要 遵 守 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:学 习 、 贯 彻 基 本 法 的 过 程 ， 无 疑 是 增 强 法 制 观 念 、 推 进 法 制 建 设 的 过 程 ， 无 疑 是 内 地 和 香 港 在 新 的 征 途 上 并 肩 同 行 、 共 创 辉 煌 的 过 程 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:法 律 一 旦 为 人 民 群 众 所 掌 握 ， 就 会 变 成 伟 大 的 力 量 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:行 文 至 此 ， 我 对 “ 基 本 法 连 着 你 我 他 ” 有 了 更 深 刻 、 更 真 切 的 理 解 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ - "processed 2270 tokens with 78 phrases; found: 82 phrases; correct: 76.\n", + "processed 370 tokens with 19 phrases; found: 18 phrases; correct: 16.\n", "\n", - "accuracy: 99.56%; precision: 92.68%; recall: 97.44%; FB1: 95.00\n", + "accuracy: 98.92%; precision: 88.89%; recall: 84.21%; FB1: 86.49\n", "\n", - " LOC: precision: 97.83%; recall: 100.00%; FB1: 98.90 46\n", + " LOC: precision: 100.00%; recall: 100.00%; FB1: 100.00 4\n", "\n", - " ORG: precision: 66.67%; recall: 100.00%; FB1: 80.00 12\n", + " ORG: precision: 100.00%; recall: 100.00%; FB1: 100.00 4\n", "\n", - " PER: precision: 95.83%; recall: 92.00%; FB1: 93.88 24\n", + " PER: precision: 80.00%; recall: 72.73%; FB1: 76.19 10\n", "\n" ] } @@ -1190,6 +1332,24 @@ "execution_count": 1, "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:523: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:524: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:525: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:526: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:527: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:532: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\n" + ] + }, { "name": "stdout", "output_type": "stream", @@ -1199,26 +1359,19 @@ "INFO:tensorflow:Restoring parameters from ./ner/output/model.ckpt-1630\n", "{1: 'O', 2: 'B-PER', 3: 'I-PER', 4: 'B-ORG', 5: 'I-ORG', 6: 'B-LOC', 7: 'I-LOC', 8: 'X', 9: '[CLS]', 10: '[SEP]'}\n", "输入句子:\n", - "中国男篮与委内瑞拉队在北京五棵松体育馆展开小组赛最后一场比赛的争夺，赵继伟12分4助攻3抢断、易建联11分8篮板、周琦8分7篮板2盖帽。\n", - "[['B-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'O', 'B-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'O', 'B-LOC', 'I-LOC', 'B-LOC', 'I-LOC', 'I-LOC', 'I-LOC', 'I-LOC', 'I-LOC', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'B-PER', 'I-PER', 'I-PER', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'B-PER', 'I-PER', 'I-PER', 'O', 'O', 'O', 'O', 'O', 'O', 'B-PER', 'I-PER', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O']]\n", - "LOC, 北京, 五棵松体育馆\n", - "PER, 赵继伟, 易建联, 周琦\n", - "ORG, 中国男篮, 委内瑞拉队\n", - "time used: 0.908481 sec\n", - "输入句子:\n", "周杰伦（Jay Chou），1979年1月18日出生于台湾省新北市，毕业于淡江中学，中国台湾流行乐男歌手。\n", "[['B-PER', 'I-PER', 'I-PER', 'O', 'B-PER', 'I-PER', 'I-PER', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'B-LOC', 'I-LOC', 'I-LOC', 'B-LOC', 'I-LOC', 'I-LOC', 'O', 'O', 'O', 'O', 'B-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'O', 'B-LOC', 'I-LOC', 'B-LOC', 'I-LOC', 'O', 'O', 'O', 'O', 'O', 'O', 'O']]\n", "LOC, 台湾省, 新北市, 中国, 台湾\n", "PER, 周杰伦, jaycho##u\n", "ORG, 淡江中学\n", - "time used: 0.058148 sec\n", + "time used: 0.885981 sec\n", "输入句子:\n", "马云，1964年9月10日生于浙江省杭州市，1988年毕业于杭州师范学院外语系，同年担任杭州电子工业学院英文及国际贸易教师。\n", "[['B-PER', 'I-PER', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'B-LOC', 'I-LOC', 'I-LOC', 'B-LOC', 'I-LOC', 'I-LOC', 'O', 'O', 'O', 'O', 'O', 'O', 'B-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'O', 'O', 'O', 'O', 'O', 'B-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O']]\n", "LOC, 浙江省, 杭州市\n", "PER, 马云\n", "ORG, 杭州师范学院外语系, 杭州电子工业学院\n", - "time used: 0.065471 sec\n", + "time used: 0.041288 sec\n", "输入句子:\n", "再见\n", "\n", @@ -1231,11 +1384,11 @@ ] }, { - "cell_type": "code", - "execution_count": null, + "cell_type": "markdown", "metadata": {}, - "outputs": [], - "source": [] + "source": [ + "## 本案例到此结束" + ] } ], "metadata": { @@ -1254,7 +1407,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.7.3" + "version": "3.8.3" } }, "nbformat": 4, diff --git a/notebook/DL_nlp_bert_ner/nlp_ner.ipynb b/notebook/DL_nlp_bert_ner/nlp_ner.ipynb index 7a7bcd990..a7b75c34e 100644 --- a/notebook/DL_nlp_bert_ner/nlp_ner.ipynb +++ b/notebook/DL_nlp_bert_ner/nlp_ner.ipynb @@ -28,18 +28,10 @@ "\n", "第二步：填写notebook所需的参数：\n", "\n", - "| 参数 | 说明 |\n", - "| - - - - - | - - - - - |\n", - "| 计费方式 | 按需计费 |\n", - "| 名称 | 自定义名称 |\n", - "| 工作环境 | Python3 |\n", - "| 资源池 | 公共资源池 |\n", - "| 类型 | GPU |\n", - "| 规格 | [限时免费]体验规格GPU版 |\n", - "| 存储配置 | EVS |\n", - "| 磁盘规格 | 5GB |\n", + "![jupyter](./img/notebook1.png)\n", "\n", "第三步：配置好notebook参数后，点击下一步，进入notebook信息预览。确认无误后，点击“立即创建”\n", + "![jupyter](./img/notebook2.png)\n", "\n", "第四步：创建完成后，返回开发环境主界面，等待Notebook创建完毕后，打开Notebook，进行下一步操作。\n", "![modelarts_notebook_index](./img/modelarts_notebook_index.png)\n", @@ -48,13 +40,14 @@ "\n", "接下来，我们创建一个实际的开发环境，用于后续的实验步骤。\n", "\n", - "第一步：点击下图所示的“打开”按钮，进入刚刚创建的Notebook\n", - "![inter_dev_env](img/enter_dev_env.png)\n", + "第一步：点击下图所示的“启动”按钮，加载后“打开”按钮变从灰色变为蓝色后点击“打开”进入刚刚创建的Notebook\n", + "![jupyter](./img/notebook3.png)\n", + "![jupyter](./img/notebook4.png)\n", "\n", - "第二步：创建一个Python3环境的的Notebook。点击右上角的\"New\"，然后创建TensorFlow 1.13.1开发环境。\n", "\n", - "第三步：点击左上方的文件名\"Untitled\"，并输入一个与本实验相关的名称\n", + "第二步：创建一个Python3环境的的Notebook。点击右上角的\"New\"，然后选择TensorFlow 1.13.1开发环境。\n", "\n", + "第三步：点击左上方的文件名\"Untitled\"，并输入一个与本实验相关的名称，如\"facial_expression\"\n", "![notebook_untitled_filename](./img/notebook_untitled_filename.png)\n", "![notebook_name_the_ipynb](./img/notebook_name_the_ipynb.png)\n", "\n", @@ -68,6 +61,7 @@ "开发环境准备好啦，接下来可以愉快地写代码啦！\n", "\n", "\n", + "\n", "### 准备源代码和数据\n", "\n", "准备案例所需的源代码和数据，相关资源已经保存在 OBS 中，我们通过 Moxing 将资源下载到本地。" @@ -82,8 +76,8 @@ "name": "stderr", "output_type": "stream", "text": [ - "INFO:root:Using MoXing-v1.15.1-3fc51aac\n", - "INFO:root:Using OBS-Python-SDK-3.1.2\n" + "INFO:root:Using MoXing-v1.17.3-\n", + "INFO:root:Using OBS-Python-SDK-3.20.7\n" ] }, { @@ -130,7 +124,26 @@ "cell_type": "code", "execution_count": 2, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:526: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:527: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:528: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:529: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:530: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:535: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\n" + ] + } + ], "source": [ "import os\n", "import json\n", @@ -522,7 +535,7 @@ "INFO:tensorflow: Batch size = 64\n", "INFO:tensorflow: Num steps = 1630\n", "INFO:tensorflow:Using config: {'_model_dir': './ner/output', '_tf_random_seed': None, '_save_summary_steps': 1000, '_save_checkpoints_steps': 1000, '_save_checkpoints_secs': None, '_session_config': allow_soft_placement: true\n", - ", '_keep_checkpoint_max': 5, '_keep_checkpoint_every_n_hours': 10000, '_log_step_count_steps': 100, '_train_distribute': None, '_device_fn': None, '_protocol': None, '_eval_distribute': None, '_experimental_distribute': None, '_service': None, '_cluster_spec': , '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1}\n", + ", '_keep_checkpoint_max': 5, '_keep_checkpoint_every_n_hours': 10000, '_log_step_count_steps': 100, '_train_distribute': None, '_device_fn': None, '_protocol': None, '_eval_distribute': None, '_experimental_distribute': None, '_service': None, '_cluster_spec': , '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1}\n", "WARNING:tensorflow:From /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n", "Instructions for updating:\n", "Colocations handled automatically by placer.\n", @@ -538,10 +551,10 @@ "INFO:tensorflow: name = input_mask, shape = (32, 128)\n", "INFO:tensorflow: name = label_ids, shape = (32, 128)\n", "INFO:tensorflow: name = segment_ids, shape = (32, 128)\n", - "WARNING:tensorflow:From /home/ma-user/work/test_course/7_2/ner/bert/modeling.py:358: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\n", + "WARNING:tensorflow:From /home/ma-user/work/DL_nlp_bert_ner/ner/bert/modeling.py:358: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\n", "Instructions for updating:\n", "Please use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n", - "WARNING:tensorflow:From /home/ma-user/work/test_course/7_2/ner/bert/modeling.py:671: dense (from tensorflow.python.layers.core) is deprecated and will be removed in a future version.\n", + "WARNING:tensorflow:From /home/ma-user/work/DL_nlp_bert_ner/ner/bert/modeling.py:671: dense (from tensorflow.python.layers.core) is deprecated and will be removed in a future version.\n", "Instructions for updating:\n", "Use keras.layers.dense instead.\n" ] @@ -569,65 +582,65 @@ "INFO:tensorflow:Running local_init_op.\n", "INFO:tensorflow:Done running local_init_op.\n", "INFO:tensorflow:Saving checkpoints for 0 into ./ner/output/model.ckpt.\n", - "INFO:tensorflow:loss = 107.43357, step = 0\n", - "INFO:tensorflow:global_steps = 0, loss = 107.43357\n", - "INFO:tensorflow:global_step/sec: 2.13493\n", - "INFO:tensorflow:loss = 40.935146, step = 100 (46.840 sec)\n", - "INFO:tensorflow:global_steps = 100, loss = 40.935146 (46.840 sec)\n", - "INFO:tensorflow:global_step/sec: 2.81105\n", - "INFO:tensorflow:loss = 54.911964, step = 200 (35.574 sec)\n", - "INFO:tensorflow:global_steps = 200, loss = 54.911964 (35.574 sec)\n", - "INFO:tensorflow:global_step/sec: 2.82329\n", - "INFO:tensorflow:loss = 44.035446, step = 300 (35.420 sec)\n", - "INFO:tensorflow:global_steps = 300, loss = 44.035446 (35.420 sec)\n", - "INFO:tensorflow:global_step/sec: 2.82064\n", - "INFO:tensorflow:loss = 45.9765, step = 400 (35.453 sec)\n", - "INFO:tensorflow:global_steps = 400, loss = 45.9765 (35.453 sec)\n", - "INFO:tensorflow:global_step/sec: 2.79905\n", - "INFO:tensorflow:loss = 47.315308, step = 500 (35.726 sec)\n", - "INFO:tensorflow:global_steps = 500, loss = 47.315308 (35.726 sec)\n", - "INFO:tensorflow:global_step/sec: 2.81567\n", - "INFO:tensorflow:loss = 43.322674, step = 600 (35.516 sec)\n", - "INFO:tensorflow:global_steps = 600, loss = 43.322674 (35.515 sec)\n", - "INFO:tensorflow:global_step/sec: 2.8103\n", - "INFO:tensorflow:loss = 46.60252, step = 700 (35.583 sec)\n", - "INFO:tensorflow:global_steps = 700, loss = 46.60252 (35.583 sec)\n", - "INFO:tensorflow:global_step/sec: 2.80791\n", - "INFO:tensorflow:loss = 40.147606, step = 800 (35.614 sec)\n", - "INFO:tensorflow:global_steps = 800, loss = 40.147606 (35.614 sec)\n", - "INFO:tensorflow:global_step/sec: 2.81373\n", - "INFO:tensorflow:loss = 35.02433, step = 900 (35.540 sec)\n", - "INFO:tensorflow:global_steps = 900, loss = 35.02433 (35.540 sec)\n", + "INFO:tensorflow:loss = 101.12476, step = 0\n", + "INFO:tensorflow:global_steps = 0, loss = 101.12476\n", + "INFO:tensorflow:global_step/sec: 2.11756\n", + "INFO:tensorflow:loss = 44.08002, step = 100 (47.225 sec)\n", + "INFO:tensorflow:global_steps = 100, loss = 44.08002 (47.224 sec)\n", + "INFO:tensorflow:global_step/sec: 2.81244\n", + "INFO:tensorflow:loss = 37.894424, step = 200 (35.556 sec)\n", + "INFO:tensorflow:global_steps = 200, loss = 37.894424 (35.556 sec)\n", + "INFO:tensorflow:global_step/sec: 2.82339\n", + "INFO:tensorflow:loss = 41.6473, step = 300 (35.418 sec)\n", + "INFO:tensorflow:global_steps = 300, loss = 41.6473 (35.418 sec)\n", + "INFO:tensorflow:global_step/sec: 2.81859\n", + "INFO:tensorflow:loss = 43.790276, step = 400 (35.479 sec)\n", + "INFO:tensorflow:global_steps = 400, loss = 43.790276 (35.479 sec)\n", + "INFO:tensorflow:global_step/sec: 2.81616\n", + "INFO:tensorflow:loss = 39.31551, step = 500 (35.509 sec)\n", + "INFO:tensorflow:global_steps = 500, loss = 39.31551 (35.509 sec)\n", + "INFO:tensorflow:global_step/sec: 2.81752\n", + "INFO:tensorflow:loss = 40.44882, step = 600 (35.492 sec)\n", + "INFO:tensorflow:global_steps = 600, loss = 40.44882 (35.492 sec)\n", + "INFO:tensorflow:global_step/sec: 2.82517\n", + "INFO:tensorflow:loss = 40.503124, step = 700 (35.396 sec)\n", + "INFO:tensorflow:global_steps = 700, loss = 40.503124 (35.396 sec)\n", + "INFO:tensorflow:global_step/sec: 2.81547\n", + "INFO:tensorflow:loss = 38.893593, step = 800 (35.518 sec)\n", + "INFO:tensorflow:global_steps = 800, loss = 38.893593 (35.518 sec)\n", + "INFO:tensorflow:global_step/sec: 2.81628\n", + "INFO:tensorflow:loss = 43.374577, step = 900 (35.508 sec)\n", + "INFO:tensorflow:global_steps = 900, loss = 43.374577 (35.508 sec)\n", "INFO:tensorflow:Saving checkpoints for 1000 into ./ner/output/model.ckpt.\n", - "INFO:tensorflow:global_step/sec: 2.30958\n", - "INFO:tensorflow:loss = 33.424232, step = 1000 (43.298 sec)\n", - "INFO:tensorflow:global_steps = 1000, loss = 33.424232 (43.298 sec)\n", - "INFO:tensorflow:global_step/sec: 2.80963\n", - "INFO:tensorflow:loss = 30.33161, step = 1100 (35.592 sec)\n", - "INFO:tensorflow:global_steps = 1100, loss = 30.33161 (35.592 sec)\n", - "INFO:tensorflow:global_step/sec: 2.80574\n", - "INFO:tensorflow:loss = 29.228725, step = 1200 (35.641 sec)\n", - "INFO:tensorflow:global_steps = 1200, loss = 29.228725 (35.641 sec)\n", - "INFO:tensorflow:global_step/sec: 2.80621\n", - "INFO:tensorflow:loss = 30.76773, step = 1300 (35.635 sec)\n", - "INFO:tensorflow:global_steps = 1300, loss = 30.76773 (35.635 sec)\n", - "INFO:tensorflow:global_step/sec: 2.80966\n", - "INFO:tensorflow:loss = 34.498055, step = 1400 (35.592 sec)\n", - "INFO:tensorflow:global_steps = 1400, loss = 34.498055 (35.592 sec)\n", - "INFO:tensorflow:global_step/sec: 2.80278\n", - "INFO:tensorflow:loss = 33.688343, step = 1500 (35.679 sec)\n", - "INFO:tensorflow:global_steps = 1500, loss = 33.688343 (35.679 sec)\n", - "INFO:tensorflow:global_step/sec: 2.81239\n", - "INFO:tensorflow:loss = 27.509579, step = 1600 (35.557 sec)\n", - "INFO:tensorflow:global_steps = 1600, loss = 27.509579 (35.557 sec)\n", + "INFO:tensorflow:global_step/sec: 2.30028\n", + "INFO:tensorflow:loss = 44.156574, step = 1000 (43.473 sec)\n", + "INFO:tensorflow:global_steps = 1000, loss = 44.156574 (43.473 sec)\n", + "INFO:tensorflow:global_step/sec: 2.8118\n", + "INFO:tensorflow:loss = 41.555115, step = 1100 (35.565 sec)\n", + "INFO:tensorflow:global_steps = 1100, loss = 41.555115 (35.565 sec)\n", + "INFO:tensorflow:global_step/sec: 2.8125\n", + "INFO:tensorflow:loss = 42.96095, step = 1200 (35.555 sec)\n", + "INFO:tensorflow:global_steps = 1200, loss = 42.96095 (35.556 sec)\n", + "INFO:tensorflow:global_step/sec: 2.81514\n", + "INFO:tensorflow:loss = 42.284943, step = 1300 (35.522 sec)\n", + "INFO:tensorflow:global_steps = 1300, loss = 42.284943 (35.522 sec)\n", + "INFO:tensorflow:global_step/sec: 2.81755\n", + "INFO:tensorflow:loss = 41.85956, step = 1400 (35.492 sec)\n", + "INFO:tensorflow:global_steps = 1400, loss = 41.85956 (35.492 sec)\n", + "INFO:tensorflow:global_step/sec: 2.81149\n", + "INFO:tensorflow:loss = 39.010487, step = 1500 (35.568 sec)\n", + "INFO:tensorflow:global_steps = 1500, loss = 39.010487 (35.568 sec)\n", + "INFO:tensorflow:global_step/sec: 2.81946\n", + "INFO:tensorflow:loss = 39.25674, step = 1600 (35.468 sec)\n", + "INFO:tensorflow:global_steps = 1600, loss = 39.25674 (35.468 sec)\n", "INFO:tensorflow:Saving checkpoints for 1630 into ./ner/output/model.ckpt.\n", - "INFO:tensorflow:Loss for final step: 30.562416.\n" + "INFO:tensorflow:Loss for final step: 45.543602.\n" ] }, { "data": { "text/plain": [ - "" + "" ] }, "execution_count": 8, @@ -712,7 +725,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 10, "metadata": {}, "outputs": [ { @@ -742,25 +755,19 @@ "name": "stderr", "output_type": "stream", "text": [ - "WARNING:tensorflow:From /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/ops/metrics_impl.py:363: to_float (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n", - "Instructions for updating:\n", - "Use tf.cast instead.\n", "INFO:tensorflow:Done calling model_fn.\n", - "INFO:tensorflow:Starting evaluation at 2020-08-09T13:19:33Z\n", + "INFO:tensorflow:Starting evaluation at 2020-11-19T09:46:43Z\n", "INFO:tensorflow:Graph was finalized.\n", - "WARNING:tensorflow:From /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/training/saver.py:1266: checkpoint_exists (from tensorflow.python.training.checkpoint_management) is deprecated and will be removed in a future version.\n", - "Instructions for updating:\n", - "Use standard file APIs to check for files with this prefix.\n", "INFO:tensorflow:Restoring parameters from ./ner/output/model.ckpt-1630\n", "INFO:tensorflow:Running local_init_op.\n", "INFO:tensorflow:Done running local_init_op.\n", - "INFO:tensorflow:Finished evaluation at 2020-08-09-13:20:00\n", - "INFO:tensorflow:Saving dict for global step 1630: eval_loss = 0.04342846, global_step = 1630, loss = 27.266157\n", + "INFO:tensorflow:Finished evaluation at 2020-11-19-09:47:09\n", + "INFO:tensorflow:Saving dict for global step 1630: eval_loss = 0.1202376, global_step = 1630, loss = 34.488083\n", "INFO:tensorflow:Saving 'checkpoint_path' summary for global step 1630: ./ner/output/model.ckpt-1630\n", "INFO:tensorflow:***** Eval results *****\n", - "INFO:tensorflow: eval_loss = 0.04342846\n", + "INFO:tensorflow: eval_loss = 0.1202376\n", "INFO:tensorflow: global_step = 1630\n", - "INFO:tensorflow: loss = 27.266157\n" + "INFO:tensorflow: loss = 34.488083\n" ] } ], @@ -807,7 +814,7 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 11, "metadata": {}, "outputs": [ { @@ -838,18 +845,18 @@ "output_type": "stream", "text": [ "INFO:tensorflow:Done calling model_fn.\n", - "INFO:tensorflow:Starting evaluation at 2020-08-09T13:20:17Z\n", + "INFO:tensorflow:Starting evaluation at 2020-11-19T09:48:19Z\n", "INFO:tensorflow:Graph was finalized.\n", "INFO:tensorflow:Restoring parameters from ./ner/output/model.ckpt-1630\n", "INFO:tensorflow:Running local_init_op.\n", "INFO:tensorflow:Done running local_init_op.\n", - "INFO:tensorflow:Finished evaluation at 2020-08-09-13:20:19\n", - "INFO:tensorflow:Saving dict for global step 1630: eval_loss = 0.013442095, global_step = 1630, loss = 26.094973\n", + "INFO:tensorflow:Finished evaluation at 2020-11-19-09:48:22\n", + "INFO:tensorflow:Saving dict for global step 1630: eval_loss = 0.053423714, global_step = 1630, loss = 33.130444\n", "INFO:tensorflow:Saving 'checkpoint_path' summary for global step 1630: ./ner/output/model.ckpt-1630\n", "INFO:tensorflow:***** Predict results *****\n", - "INFO:tensorflow: eval_loss = 0.013442095\n", + "INFO:tensorflow: eval_loss = 0.053423714\n", "INFO:tensorflow: global_step = 1630\n", - "INFO:tensorflow: loss = 26.094973\n", + "INFO:tensorflow: loss = 33.130444\n", "INFO:tensorflow:Calling model_fn.\n", "INFO:tensorflow:*** Features ***\n", "INFO:tensorflow: name = input_ids, shape = (?, 128)\n", @@ -873,22 +880,202 @@ "INFO:tensorflow:Graph was finalized.\n", "INFO:tensorflow:Restoring parameters from ./ner/output/model.ckpt-1630\n", "INFO:tensorflow:Running local_init_op.\n", - "INFO:tensorflow:Done running local_init_op.\n" + "INFO:tensorflow:Done running local_init_op.\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:在 香 港 回 归 前 的 最 后 阶 段 ， 中 共 中 央 举 办 《 “ 一 国 两 制 ” 与 香 港 基 本 法 》 讲 座 ， 中 央 领 导 同 志 认 真 听 讲 ， 虚 心 学 习 ， 很 有 意 义 。\n", + "INFO:tensorflow:O B-LOC I-LOC O O O O O O O O O B-ORG I-ORG I-ORG I-ORG O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:这 表 明 ， 以 江 泽 民 同 志 为 核 心 的 党 中 央 坚 定 不 移 地 贯 彻 邓 小 平 同 志 “ 一 国 两 制 ” 的 伟 大 构 想 ， 不 折 不 扣 地 执 行 基 本 法 。\n", + "INFO:tensorflow:O O O O O B-PER I-PER I-PER O O O O O O B-ORG I-ORG I-ORG O O O O O O O B-PER I-PER I-PER O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:“ 一 国 两 制 ” 是 邓 小 平 同 志 的 一 个 伟 大 构 想 ， 《 中 华 人 民 共 和 国 香 港 特 别 行 政 区 基 本 法 》 是 贯 彻 落 实 “ 一 国 两 制 ” 伟 大 构 想 的 一 部 全 国 性 法 律 ， 是 一 部 有 鲜 明 中 国 特 色 的 法 律 。\n", + "INFO:tensorflow:O O O O O O O B-PER I-PER I-PER O O O O O O O O O O O B-LOC I-LOC I-LOC I-LOC I-LOC I-LOC I-LOC B-LOC I-LOC I-LOC I-LOC I-LOC I-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:看 包 公 断 案 的 戏 ， 看 他 威 风 凛 凛 坐 公 堂 拍 桌 子 动 刑 具 ， 多 少 还 有 一 点 担 心 ， 总 怕 靠 这 一 套 办 法 弄 出 错 案 来 ， 放 过 了 真 正 的 坏 人 ；\n", + "INFO:tensorflow:O B-PER I-PER O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:可 看 《 包 公 赶 驴 》 这 出 戏 ， 心 里 就 很 踏 实 ： 这 样 是 一 断 一 个 准 的 。\n", + "INFO:tensorflow:O O O B-PER I-PER O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:譬 如 看 《 施 公 案 》 ， 施 大 人 坐 公 堂 问 案 子 不 得 要 领 ， 总 是 扮 成 普 通 百 姓 深 入 民 间 暗 中 查 访 ， 结 果 就 屡 破 奇 案 了 。\n", + "INFO:tensorflow:O O O O B-PER O O O O B-PER O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:如 果 有 人 问 我 ： “ 你 看 过 许 多 包 公 戏 ， 哪 一 出 最 好 ？ ”\n", + "INFO:tensorflow:O O O O O O O O O O O O O B-PER I-PER O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:我 要 毫 不 犹 豫 地 回 答 道 ： “ 自 然 是 《 包 公 赶 驴 》 啦 ！\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O B-PER I-PER O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:学 习 基 本 法 顺 利 迎 回 归\n", + "INFO:tensorflow:O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:本 报 评 论 员\n", + "INFO:tensorflow:O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:再 过 5 5 天 ， 我 国 政 府 将 对 香 港 恢 复 行 使 主 权 。\n", + "INFO:tensorflow:O O O O O O O O O O O O B-LOC I-LOC O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:它 把 中 央 对 解 决 香 港 问 题 的 基 本 方 针 政 策 具 体 化 、 法 律 化 ， 成 为 国 家 意 志 。\n", + "INFO:tensorflow:O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:学 习 基 本 法 ， 顺 利 迎 回 归 ， 是 一 项 迫 切 的 任 务 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:要 学 好 基 本 法 ， 首 先 要 认 识 到 基 本 法 的 意 义 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:说 国 际 意 义 ， 不 只 对 第 三 世 界 ， 而 且 对 全 人 类 都 具 有 长 远 意 义 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:这 是 一 个 具 有 创 造 性 的 杰 作 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:“ 基 本 法 不 仅 为 确 保 香 港 平 稳 过 渡 发 挥 重 要 作 用 ， 也 为 确 保 香 港 长 期 繁 荣 稳 定 发 挥 重 要 作 用 ；\n", + "INFO:tensorflow:O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:不 仅 为 当 前 解 决 香 港 问 题 发 挥 作 用 ， 也 为 在 不 远 的 将 来 解 决 澳 门 问 题 和 最 终 解 决 台 湾 问 题 ， 实 现 祖 国 完 全 统 一 发 挥 重 要 作 用 。\n", + "INFO:tensorflow:O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:基 本 法 的 主 要 特 征 ， 是 把 “ 一 国 ” 与 “ 两 制 ” 紧 密 结 合 ， 维 护 国 家 的 主 权 、 统 一 和 领 土 完 整 与 授 权 香 港 特 别 行 政 区 实 行 高 度 自 治 紧 密 结 合 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O B-LOC I-LOC I-LOC I-LOC I-LOC I-LOC I-LOC O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:在 一 个 统 一 的 中 华 人 民 共 和 国 ， 可 以 实 行 社 会 主 义 和 资 本 主 义 两 种 制 度 ， 这 是 为 了 民 族 、 国 家 的 根 本 利 益 。\n", + "INFO:tensorflow:O O O O O O B-LOC I-LOC I-LOC I-LOC I-LOC I-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:只 有 认 真 学 习 ， 才 能 理 解 意 义 ， 认 识 特 征 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:制 定 一 部 好 法 律 ， 很 不 容 易 ；\n", + "INFO:tensorflow:O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:遵 守 法 律 ， 执 行 法 律 ， 也 很 不 容 易 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:必 须 重 申 ， 有 法 必 依 ， 执 法 必 严 ， 违 法 必 究 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:基 本 法 作 为 一 部 全 国 性 的 法 律 ， 不 仅 香 港 要 严 格 遵 守 ， 各 省 、 自 治 区 、 直 辖 市 都 要 严 格 遵 守 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:从 中 共 中 央 举 办 这 个 讲 座 ， 可 以 看 出 ， 党 和 政 府 正 在 努 力 加 强 法 制 建 设 ， 坚 持 依 法 治 国 。\n", + "INFO:tensorflow:O B-ORG I-ORG I-ORG I-ORG O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:有 了 法 律 ， 有 了 制 度 ， 就 有 了 保 证 ， 就 使 “ 一 国 两 制 ” 的 伟 大 构 想 以 法 律 的 形 式 固 定 下 来 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:全 国 人 民 特 别 是 香 港 同 胞 也 从 中 再 一 次 看 到 ， 中 国 共 产 党 和 人 民 政 府 是 高 度 负 责 任 的 党 和 政 府 ， 一 切 从 人 民 的 利 益 出 发 ， 一 切 为 了 祖 国 的 繁 荣 富 强 ， 香 港 的 明 天 将 更 美 好 。\n", + "INFO:tensorflow:O O O O O O O B-LOC I-LOC O O O O O O O O O O O B-ORG I-ORG I-ORG I-ORG I-ORG O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:学 习 基 本 法 ， 中 央 领 导 带 了 个 好 头 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:全 党 和 全 国 人 民 特 别 是 各 级 党 政 领 导 干 部 ， 都 要 重 视 学 习 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:只 有 学 习 好 ， 才 能 贯 彻 好 。\n", + "INFO:tensorflow:O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:为 了 迎 接 香 港 顺 利 回 归 祖 国 这 一 中 华 民 族 的 盛 事 ， 首 先 要 有 一 个 扎 实 的 思 想 准 备 和 良 好 的 精 神 状 态 。\n", + "INFO:tensorflow:O O O O B-LOC I-LOC O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:基 本 法 连 着 你 我 他\n", + "INFO:tensorflow:O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:叶 秋\n", + "INFO:tensorflow:B-PER I-PER\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:赠 书 想 来 是 香 港 同 胞 的 一 种 文 明 礼 仪 。\n", + "INFO:tensorflow:O O O O O B-LOC I-LOC O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:抵 港 仅 数 日 ， 就 收 到 厚 厚 几 摞 书 。\n", + "INFO:tensorflow:O B-LOC O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:匆 匆 翻 阅 一 遍 ， 发 现 各 种 版 本 的 《 中 华 人 民 共 和 国 香 港 特 别 行 政 区 基 本 法 》 竟 有 六 册 之 多 ， 推 介 普 及 基 本 法 的 书 籍 还 要 多 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O B-LOC I-LOC I-LOC I-LOC I-LOC I-LOC I-LOC B-LOC I-LOC I-LOC I-LOC I-LOC I-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:应 约 去 湾 仔 道 谈 事 ， 路 过 一 个 名 为 “ 艺 美 ” 的 书 店 ， 看 到 摆 放 在 最 抢 眼 位 置 的 也 是 基 本 法 及 其 推 介 图 书 。\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:tensorflow:O O O B-LOC I-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:由 此 可 见 ， 在 法 制 观 念 很 强 的 港 人 心 目 中 ， 基 本 法 具 有 极 大 的 权 威 和 尊 严 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O B-LOC O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:行 政 官 员 表 示 ： “ 香 港 继 续 繁 荣 稳 定 、 实 现 香 港 梦 的 成 功 要 素 ， 在 基 本 法 中 得 到 了 充 分 保 证 。 ”\n", + "INFO:tensorflow:O O O O O O O O B-LOC I-LOC O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:法 律 界 人 士 认 为 ： “ 法 治 精 神 能 否 继 续 保 持 ， 基 本 法 已 作 了 明 确 规 定 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:只 要 恪 守 广 大 港 人 认 受 的 香 港 法 律 体 系 中 的 这 个 总 纲 纪 、 总 章 程 ， 香 港 将 健 步 迈 向 新 世 纪 。 ”\n", + "INFO:tensorflow:O O O O O O B-LOC O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:劳 工 界 的 成 员 说 ， 涉 及 保 障 劳 工 合 法 权 益 的 条 款 ， “ 香 港 现 在 有 的 ， 基 本 法 都 保 持 了 ；\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:大 家 因 此 吃 了 定 心 丸 。 ”\n", + "INFO:tensorflow:O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:基 本 法 受 到 港 人 的 普 遍 欢 迎 和 高 度 重 视 是 势 所 必 然 。\n", + "INFO:tensorflow:O O O O O B-LOC O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:历 时 四 年 零 八 个 月 、 凝 聚 了 香 港 和 内 地 无 数 人 的 智 慧 而 制 定 的 基 本 法 ， 将 邓 小 平 同 志 倡 导 的 “ 一 国 两 制 ” 伟 大 构 想 以 法 律 形 式 固 定 下 来 ， 成 为 国 家 和 人 民 的 意 志 。\n", + "INFO:tensorflow:O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O B-PER I-PER I-PER O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:基 本 法 既 是 香 港 回 归 后 特 区 一 切 运 作 的 法 律 基 础 ， 更 是 保 持 香 港 长 期 稳 定 繁 荣 的 法 律 保 证 。\n", + "INFO:tensorflow:O O O O O B-LOC I-LOC O O O B-LOC I-LOC O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:实 践 已 经 并 将 继 续 证 明 这 一 点 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:说 来 也 巧 ， 姬 鹏 飞 同 志 1 9 9 0 年 4 月 在 邓 小 平 同 志 题 写 书 名 的 《 基 本 法 的 诞 生 》 一 书 序 言 中 也 写 了 同 样 的 话 。\n", + "INFO:tensorflow:O O O O O B-PER I-PER I-PER O O O O O O O O O O B-PER I-PER I-PER O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:真 可 谓 仁 者 智 者 所 见 略 同 。\n", + "INFO:tensorflow:O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:基 本 法 是 一 部 具 有 普 遍 约 束 力 的 重 要 法 律 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:7 月 1 日 ， 这 部 重 要 法 律 即 开 始 正 式 实 施 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:基 本 法 不 仅 体 现 了 香 港 同 胞 的 意 志 和 利 益 ， 也 体 现 了 全 国 人 民 的 意 志 和 利 益 。\n", + "INFO:tensorflow:O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:正 因 为 如 此 ， 江 泽 民 同 志 强 调 ： 香 港 基 本 法 是 一 部 全 国 性 的 法 律 ， 不 仅 香 港 要 严 格 遵 守 ， 各 省 、 自 治 区 、 直 辖 市 都 要 严 格 遵 守 。\n", + "INFO:tensorflow:O O O O O O B-PER I-PER I-PER O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:还 表 示 ， 不 仅 我 要 遵 守 ， 我 希 望 香 港 同 胞 和 全 国 1 2 亿 人 民 也 要 遵 守 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:学 习 、 贯 彻 基 本 法 的 过 程 ， 无 疑 是 增 强 法 制 观 念 、 推 进 法 制 建 设 的 过 程 ， 无 疑 是 内 地 和 香 港 在 新 的 征 途 上 并 肩 同 行 、 共 创 辉 煌 的 过 程 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O B-LOC I-LOC O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:法 律 一 旦 为 人 民 群 众 所 掌 握 ， 就 会 变 成 伟 大 的 力 量 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O O O\n", + "INFO:tensorflow:list index out of range\n", + "INFO:tensorflow:行 文 至 此 ， 我 对 “ 基 本 法 连 着 你 我 他 ” 有 了 更 深 刻 、 更 真 切 的 理 解 。\n", + "INFO:tensorflow:O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ - "processed 2270 tokens with 78 phrases; found: 75 phrases; correct: 68.\n", + "processed 370 tokens with 19 phrases; found: 18 phrases; correct: 16.\n", "\n", - "accuracy: 99.38%; precision: 90.67%; recall: 87.18%; FB1: 88.89\n", + "accuracy: 98.92%; precision: 88.89%; recall: 84.21%; FB1: 86.49\n", "\n", - " LOC: precision: 97.78%; recall: 97.78%; FB1: 97.78 45\n", + " LOC: precision: 100.00%; recall: 100.00%; FB1: 100.00 4\n", "\n", - " ORG: precision: 77.78%; recall: 87.50%; FB1: 82.35 9\n", + " ORG: precision: 100.00%; recall: 100.00%; FB1: 100.00 4\n", "\n", - " PER: precision: 80.95%; recall: 68.00%; FB1: 73.91 21\n", + " PER: precision: 80.00%; recall: 72.73%; FB1: 76.19 10\n", "\n" ] } @@ -985,6 +1172,24 @@ "execution_count": 1, "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:526: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:527: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:528: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:529: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:530: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n", + "/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:535: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n", + " np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\n" + ] + }, { "name": "stdout", "output_type": "stream", @@ -994,7 +1199,7 @@ "WARNING:tensorflow:From /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n", "Instructions for updating:\n", "Colocations handled automatically by placer.\n", - "WARNING:tensorflow:From /home/ma-user/work/test_course/7_2/ner/bert/modeling.py:671: dense (from tensorflow.python.layers.core) is deprecated and will be removed in a future version.\n", + "WARNING:tensorflow:From /home/ma-user/work/DL_nlp_bert_ner/ner/bert/modeling.py:671: dense (from tensorflow.python.layers.core) is deprecated and will be removed in a future version.\n", "Instructions for updating:\n", "Use keras.layers.dense instead.\n", "WARNING:tensorflow:From /home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/tensorflow/contrib/crf/python/ops/crf.py:567: dynamic_rnn (from tensorflow.python.ops.rnn) is deprecated and will be removed in a future version.\n", @@ -1009,26 +1214,19 @@ "INFO:tensorflow:Restoring parameters from ./ner/output/model.ckpt-1630\n", "{1: 'O', 2: 'B-PER', 3: 'I-PER', 4: 'B-ORG', 5: 'I-ORG', 6: 'B-LOC', 7: 'I-LOC', 8: 'X', 9: '[CLS]', 10: '[SEP]'}\n", "输入句子:\n", - "中国男篮与委内瑞拉队在北京五棵松体育馆展开小组赛最后一场比赛的争夺，赵继伟12分4助攻3抢断、易建联11分8篮板、周琦8分7篮板2盖帽。\n", - "[['B-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'O', 'B-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'O', 'B-LOC', 'I-LOC', 'B-LOC', 'I-LOC', 'I-LOC', 'I-LOC', 'I-LOC', 'I-LOC', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'B-PER', 'I-PER', 'I-PER', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'B-PER', 'I-PER', 'I-PER', 'O', 'O', 'O', 'O', 'O', 'O', 'B-PER', 'I-PER', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O']]\n", - "LOC, 北京, 五棵松体育馆\n", - "PER, 赵继伟, 易建联, 周琦\n", - "ORG, 中国男篮, 委内瑞拉队\n", - "time used: 0.817237 sec\n", - "输入句子:\n", "周杰伦（Jay Chou），1979年1月18日出生于台湾省新北市，毕业于淡江中学，中国台湾流行乐男歌手。\n", "[['B-PER', 'I-PER', 'I-PER', 'O', 'B-PER', 'I-PER', 'I-PER', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'B-LOC', 'I-LOC', 'I-LOC', 'B-LOC', 'I-LOC', 'I-LOC', 'O', 'O', 'O', 'O', 'B-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'O', 'B-LOC', 'I-LOC', 'B-LOC', 'I-LOC', 'O', 'O', 'O', 'O', 'O', 'O', 'O']]\n", "LOC, 台湾省, 新北市, 中国, 台湾\n", "PER, 周杰伦, jaycho##u\n", "ORG, 淡江中学\n", - "time used: 0.033615 sec\n", + "time used: 0.826048 sec\n", "输入句子:\n", "马云，1964年9月10日生于浙江省杭州市，1988年毕业于杭州师范学院外语系，同年担任杭州电子工业学院英文及国际贸易教师。\n", "[['B-PER', 'I-PER', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'B-LOC', 'I-LOC', 'I-LOC', 'B-LOC', 'I-LOC', 'I-LOC', 'O', 'O', 'O', 'O', 'O', 'O', 'B-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'O', 'O', 'O', 'O', 'O', 'B-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'I-ORG', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O']]\n", "LOC, 浙江省, 杭州市\n", "PER, 马云\n", "ORG, 杭州师范学院外语系, 杭州电子工业学院\n", - "time used: 0.041776 sec\n", + "time used: 0.034984 sec\n", "输入句子:\n", "再见\n", "\n", @@ -1041,11 +1239,11 @@ ] }, { - "cell_type": "code", - "execution_count": null, + "cell_type": "markdown", "metadata": {}, - "outputs": [], - "source": [] + "source": [ + "## 本案例到此结束" + ] } ], "metadata": { @@ -1064,7 +1262,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.6.4" + "version": "3.8.3" } }, "nbformat": 4,