diff --git a/.idea/deployment.xml b/.idea/deployment.xml
index 53a02f7..94f98bd 100644
--- a/.idea/deployment.xml
+++ b/.idea/deployment.xml
@@ -1,6 +1,6 @@
-
+
@@ -86,6 +86,20 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
@@ -142,6 +156,27 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
@@ -156,6 +191,13 @@
+
+
+
+
+
+
+
@@ -793,6 +835,48 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
@@ -801,5 +885,6 @@
+
\ No newline at end of file
diff --git a/.idea/generate_articles_directory.iml b/.idea/generate_articles_directory.iml
index 37901ac..12ca792 100644
--- a/.idea/generate_articles_directory.iml
+++ b/.idea/generate_articles_directory.iml
@@ -5,7 +5,7 @@
-
+
diff --git a/articles_directory_predict.py b/articles_directory_predict.py
index c3d4380..050f870 100644
--- a/articles_directory_predict.py
+++ b/articles_directory_predict.py
@@ -164,4 +164,4 @@ def articles_directory():
return jsonify(results) # 返回结果
if __name__ == "__main__":
- app.run(debug=False, host='0.0.0.0', port=18000)
\ No newline at end of file
+ app.run(debug=False, host='0.0.0.0', port=18000)
diff --git a/articles_directory_predict_word.py b/articles_directory_predict_word.py
new file mode 100644
index 0000000..86d6a0a
--- /dev/null
+++ b/articles_directory_predict_word.py
@@ -0,0 +1,240 @@
+from flask import Flask, jsonify
+from flask import request
+import time
+import concurrent.futures
+import requests
+import socket
+import os
+
+
+class log:
+ def __init__(self):
+ pass
+
+ def log(*args, **kwargs):
+ format = '%Y/%m/%d-%H:%M:%S'
+ format_h = '%Y-%m-%d'
+ value = time.localtime(int(time.time()))
+ dt = time.strftime(format, value)
+ dt_log_file = time.strftime(format_h, value)
+ log_file = 'log_file/access-%s' % dt_log_file + ".log"
+ if not os.path.exists(log_file):
+ with open(os.path.join(log_file), 'w', encoding='utf-8') as f:
+ print(dt, *args, file=f, **kwargs)
+ else:
+ with open(os.path.join(log_file), 'a+', encoding='utf-8') as f:
+ print(dt, *args, file=f, **kwargs)
+
+
+def get_host_ip():
+ """
+ 查询本机ip地址
+ :return: ip
+ """
+ try:
+ s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
+ s.connect(('8.8.8.8', 80))
+ ip = s.getsockname()[0]
+ finally:
+ s.close()
+
+ return ip
+
+chatgpt_url_predict = "http://{}:12000/predict".format(str(get_host_ip()))
+chatgpt_url_search = "http://{}:12000/search".format(str(get_host_ip()))
+
+prompt = {
+ "mulu_title_Level_2": "User:为论文题目“%s”生成中文目录，要求只有一级标题和二级标题，一级标题使用中文数字 例如一、xxx；二级标题使用阿拉伯数字 例如1.1 xxx；一级标题生成%s个；每个一级标题包含%s个二级标题\n\nAssistant:",
+ "mulu_title_Level_2_1": "User:为论文题目“%s”生成中文目录，要求只有一级标题和二级标题，一级标题使用中文数字 例如一、xxx；二级标题使用阿拉伯数字 例如1.1 xxx；一级标题生成%s个；每个一级标题包含%s-%s个二级标题\n\nAssistant:",
+ "mulu_title_Level_3": "User:为论文题目“%s”生成中文目录，要求只有一级标题，二级标题和三级标题，一级标题使用中文数字 例如一、xxx；二级标题使用阿拉伯数字 例如1.1 xxx；三级标题使用阿拉伯数字 例如1.1.2 xxx；一级标题生成%s个；每个一级标题包含%s个二级标题；三级标题个数适中\n\nAssistant:",
+ "mulu_title_Level_3_1": "User:为论文题目“%s”生成中文目录，要求只有一级标题，二级标题和三级标题，一级标题使用中文数字 例如一、xxx；二级标题使用阿拉伯数字 例如1.1 xxx；三级标题使用阿拉伯数字 例如1.1.2 xxx；一级标题生成%s个；每个一级标题包含%s-%s个二级标题；三级标题个数适中\n\nAssistant:",
+ "mulu_title_Level_3_2": "User:为论文题目“%s”生成中文目录，要求只有一级标题，二级标题和三级标题，一级标题使用中文数字 例如一、xxx；二级标题使用阿拉伯数字 例如1.1 xxx；三级标题使用阿拉伯数字 例如1.1.2 xxx；一级标题生成%s个；每个一级标题包含%s-%s个二级标题；三级标题个数越多越好\n\nAssistant:"
+}
+
+nums_word_dict = {
+ "3000": {
+ "prompt": prompt["mulu_title_Level_2"],
+ "title_1_nums": "3",
+ "title_2_nums": ["2"],
+ "title_small_nums": ["3", "2"]
+ },
+ "5000": {
+ "prompt": prompt["mulu_title_Level_2_1"],
+ "title_1_nums": "3",
+ "title_2_nums": ["2","3"],
+ "title_small_nums": ["3", "2","3"]
+ },
+ "8000": {
+ "prompt": prompt["mulu_title_Level_2_1"],
+ "title_1_nums": "4",
+ "title_2_nums": ["2", "4"],
+ "title_small_nums": ["4", "2", "4"],
+ },
+ "10000": {
+ "prompt": prompt["mulu_title_Level_2_1"],
+ "title_1_nums": "6",
+ "title_2_nums": ["3", "5"],
+ "title_small_nums": ["6", "3", "5"],
+ },
+ "15000": {
+ "prompt": prompt["mulu_title_Level_3_1"],
+ "title_1_nums": "5",
+ "title_2_nums": ["3", "5"],
+ "title_small_nums": ["5", "3", "5"],
+ },
+ "20000": {
+ "prompt": prompt["mulu_title_Level_3_2"],
+ "title_1_nums": "7",
+ "title_2_nums": ["3", "5"],
+ "title_small_nums": ["7", "3", "5"],
+ },
+ "30000": {
+ "prompt": prompt["mulu_title_Level_3_2"],
+ "title_1_nums": "8",
+ "title_2_nums": ["5", "8"],
+ "title_small_nums": ["8", "5", "8"],
+ },
+}
+
+app = Flask(__name__)
+app.config["JSON_AS_ASCII"] = False
+
+def dialog_line_parse(url, text):
+ """
+ 将数据输入模型进行分析并输出结果
+ :param url: 模型url
+ :param text: 进入模型的数据
+ :return: 模型返回结果
+ """
+
+ response = requests.post(
+ url,
+ json=text,
+ timeout=1000
+ )
+ if response.status_code == 200:
+ return response.json()
+ else:
+ # logger.error(
+ # "【{}】 Failed to get a proper response from remote "
+ # "server. Status Code: {}. Response: {}"
+ # "".format(url, response.status_code, response.text)
+ # )
+ print("【{}】 Failed to get a proper response from remote "
+ "server. Status Code: {}. Response: {}"
+ "".format(url, response.status_code, response.text))
+ print(text)
+ return []
+
+
+def request_api_chatgpt(prompt):
+ data = {
+ "texts": prompt
+ }
+ response = requests.post(
+ chatgpt_url_predict,
+ json=data,
+ timeout=100000
+ )
+ if response.status_code == 200:
+ return response.json()
+ else:
+ # logger.error(
+ # "【{}】 Failed to get a proper response from remote "
+ # "server. Status Code: {}. Response: {}"
+ # "".format(url, response.status_code, response.text)
+ # )
+ print("Failed to get a proper response from remote "
+ "server. Status Code: {}. Response: {}"
+ "".format(response.status_code, response.text))
+ return {}
+
+
+def uuid_search(uuid):
+ data = {
+ "id": uuid
+ }
+ response = requests.post(
+ chatgpt_url_search,
+ json=data,
+ timeout=100000
+ )
+ if response.status_code == 200:
+ return response.json()
+ else:
+ # logger.error(
+ # "【{}】 Failed to get a proper response from remote "
+ # "server. Status Code: {}. Response: {}"
+ # "".format(url, response.status_code, response.text)
+ # )
+ print("Failed to get a proper response from remote "
+ "server. Status Code: {}. Response: {}"
+ "".format(response.status_code, response.text))
+ return {}
+
+
+def uuid_search_mp(results):
+ results_list = [""] * len(results)
+ while True:
+ tiaochu_bool = True
+
+ for i in results_list:
+ if i == "":
+ tiaochu_bool = False
+ break
+
+ if tiaochu_bool == True:
+ break
+
+ for i in range(len(results)):
+ uuid = results[i]["texts"]["id"]
+
+ result = uuid_search(uuid)
+ if result["code"] == 200:
+ results_list[i] = result["text"]
+ time.sleep(3)
+ return results_list
+
+
+def get_multiple_urls(urls):
+ input_values = []
+
+ for i in urls:
+ input_values.append(i[1])
+ with concurrent.futures.ThreadPoolExecutor() as executor:
+ # 使用map方法并发地调用worker_function
+ results = list(executor.map(request_api_chatgpt, input_values))
+
+ with concurrent.futures.ThreadPoolExecutor() as executor:
+ # 使用map方法并发地调用worker_function
+ results = list(executor.map(uuid_search_mp, [results]))
+
+ return_list = []
+ for i in results[0]:
+ data = {
+ "code": 200,
+ "data": i
+ }
+ return_list.append(data)
+ return return_list
+
+
+@app.route("/articles_directory", methods=["POST"])
+def articles_directory():
+ title = request.json["title"] # 获取用户query中的文本 例如"I love you"
+ nums_catalogue = request.json["nums_catalogue"]
+ nums_word = request.json["nums_word"] # 获取用户query中的文本 例如"I love you"
+ nums_catalogue = int(nums_catalogue)
+ keyword = tuple([title] + nums_word_dict[nums_word]["title_small_nums"])
+ prompt = nums_word_dict[nums_word]["prompt"]%keyword
+ print(prompt)
+ input_data = []
+ for i in range(nums_catalogue):
+ input_data.append([i, prompt])
+
+ results = get_multiple_urls(input_data)
+ log.log('text:{},'.format(prompt))
+ return jsonify(results) # 返回结果
+
+if __name__ == "__main__":
+ app.run(debug=False, host='0.0.0.0', port=18001)
diff --git a/articles_directory_predict_word_online.py b/articles_directory_predict_word_online.py
new file mode 100644
index 0000000..87e9bca
--- /dev/null
+++ b/articles_directory_predict_word_online.py
@@ -0,0 +1,240 @@
+from flask import Flask, jsonify
+from flask import request
+import time
+import concurrent.futures
+import requests
+import socket
+import os
+
+
+class log:
+ def __init__(self):
+ pass
+
+ def log(*args, **kwargs):
+ format = '%Y/%m/%d-%H:%M:%S'
+ format_h = '%Y-%m-%d'
+ value = time.localtime(int(time.time()))
+ dt = time.strftime(format, value)
+ dt_log_file = time.strftime(format_h, value)
+ log_file = 'log_file/access-%s' % dt_log_file + ".log"
+ if not os.path.exists(log_file):
+ with open(os.path.join(log_file), 'w', encoding='utf-8') as f:
+ print(dt, *args, file=f, **kwargs)
+ else:
+ with open(os.path.join(log_file), 'a+', encoding='utf-8') as f:
+ print(dt, *args, file=f, **kwargs)
+
+
+def get_host_ip():
+ """
+ 查询本机ip地址
+ :return: ip
+ """
+ try:
+ s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
+ s.connect(('8.8.8.8', 80))
+ ip = s.getsockname()[0]
+ finally:
+ s.close()
+
+ return ip
+
+chatgpt_url_predict = "http://{}:12001/predict".format(str(get_host_ip()))
+chatgpt_url_search = "http://{}:12001/search".format(str(get_host_ip()))
+
+prompt = {
+ "mulu_title_Level_2": "User:为论文题目“%s”生成中文目录，要求只有一级标题和二级标题，一级标题使用中文数字 例如一、xxx；二级标题使用阿拉伯数字 例如1.1 xxx；一级标题生成%s个；每个一级标题包含%s个二级标题\n\nAssistant:",
+ "mulu_title_Level_2_1": "User:为论文题目“%s”生成中文目录，要求只有一级标题和二级标题，一级标题使用中文数字 例如一、xxx；二级标题使用阿拉伯数字 例如1.1 xxx；一级标题生成%s个；每个一级标题包含%s-%s个二级标题\n\nAssistant:",
+ "mulu_title_Level_3": "User:为论文题目“%s”生成中文目录，要求只有一级标题，二级标题和三级标题，一级标题使用中文数字 例如一、xxx；二级标题使用阿拉伯数字 例如1.1 xxx；三级标题使用阿拉伯数字 例如1.1.2 xxx；一级标题生成%s个；每个一级标题包含%s个二级标题；三级标题个数适中\n\nAssistant:",
+ "mulu_title_Level_3_1": "User:为论文题目“%s”生成中文目录，要求只有一级标题，二级标题和三级标题，一级标题使用中文数字 例如一、xxx；二级标题使用阿拉伯数字 例如1.1 xxx；三级标题使用阿拉伯数字 例如1.1.2 xxx；一级标题生成%s个；每个一级标题包含%s-%s个二级标题；三级标题个数适中\n\nAssistant:",
+ "mulu_title_Level_3_2": "User:为论文题目“%s”生成中文目录，要求只有一级标题，二级标题和三级标题，一级标题使用中文数字 例如一、xxx；二级标题使用阿拉伯数字 例如1.1 xxx；三级标题使用阿拉伯数字 例如1.1.2 xxx；一级标题生成%s个；每个一级标题包含%s-%s个二级标题；三级标题个数越多越好\n\nAssistant:"
+}
+
+nums_word_dict = {
+ "3000": {
+ "prompt": prompt["mulu_title_Level_2"],
+ "title_1_nums": "3",
+ "title_2_nums": ["2"],
+ "title_small_nums": ["3", "2"]
+ },
+ "5000": {
+ "prompt": prompt["mulu_title_Level_2_1"],
+ "title_1_nums": "3",
+ "title_2_nums": ["2","3"],
+ "title_small_nums": ["3", "2","3"]
+ },
+ "8000": {
+ "prompt": prompt["mulu_title_Level_2_1"],
+ "title_1_nums": "4",
+ "title_2_nums": ["2", "4"],
+ "title_small_nums": ["4", "2", "4"],
+ },
+ "10000": {
+ "prompt": prompt["mulu_title_Level_2_1"],
+ "title_1_nums": "6",
+ "title_2_nums": ["3", "5"],
+ "title_small_nums": ["6", "3", "5"],
+ },
+ "15000": {
+ "prompt": prompt["mulu_title_Level_3_1"],
+ "title_1_nums": "5",
+ "title_2_nums": ["3", "5"],
+ "title_small_nums": ["5", "3", "5"],
+ },
+ "20000": {
+ "prompt": prompt["mulu_title_Level_3_2"],
+ "title_1_nums": "7",
+ "title_2_nums": ["3", "5"],
+ "title_small_nums": ["7", "3", "5"],
+ },
+ "30000": {
+ "prompt": prompt["mulu_title_Level_3_2"],
+ "title_1_nums": "8",
+ "title_2_nums": ["5", "8"],
+ "title_small_nums": ["8", "5", "8"],
+ },
+}
+
+app = Flask(__name__)
+app.config["JSON_AS_ASCII"] = False
+
+def dialog_line_parse(url, text):
+ """
+ 将数据输入模型进行分析并输出结果
+ :param url: 模型url
+ :param text: 进入模型的数据
+ :return: 模型返回结果
+ """
+
+ response = requests.post(
+ url,
+ json=text,
+ timeout=1000
+ )
+ if response.status_code == 200:
+ return response.json()
+ else:
+ # logger.error(
+ # "【{}】 Failed to get a proper response from remote "
+ # "server. Status Code: {}. Response: {}"
+ # "".format(url, response.status_code, response.text)
+ # )
+ print("【{}】 Failed to get a proper response from remote "
+ "server. Status Code: {}. Response: {}"
+ "".format(url, response.status_code, response.text))
+ print(text)
+ return []
+
+
+def request_api_chatgpt(prompt):
+ data = {
+ "texts": prompt
+ }
+ response = requests.post(
+ chatgpt_url_predict,
+ json=data,
+ timeout=100000
+ )
+ if response.status_code == 200:
+ return response.json()
+ else:
+ # logger.error(
+ # "【{}】 Failed to get a proper response from remote "
+ # "server. Status Code: {}. Response: {}"
+ # "".format(url, response.status_code, response.text)
+ # )
+ print("Failed to get a proper response from remote "
+ "server. Status Code: {}. Response: {}"
+ "".format(response.status_code, response.text))
+ return {}
+
+
+def uuid_search(uuid):
+ data = {
+ "id": uuid
+ }
+ response = requests.post(
+ chatgpt_url_search,
+ json=data,
+ timeout=100000
+ )
+ if response.status_code == 200:
+ return response.json()
+ else:
+ # logger.error(
+ # "【{}】 Failed to get a proper response from remote "
+ # "server. Status Code: {}. Response: {}"
+ # "".format(url, response.status_code, response.text)
+ # )
+ print("Failed to get a proper response from remote "
+ "server. Status Code: {}. Response: {}"
+ "".format(response.status_code, response.text))
+ return {}
+
+
+def uuid_search_mp(results):
+ results_list = [""] * len(results)
+ while True:
+ tiaochu_bool = True
+
+ for i in results_list:
+ if i == "":
+ tiaochu_bool = False
+ break
+
+ if tiaochu_bool == True:
+ break
+
+ for i in range(len(results)):
+ uuid = results[i]["texts"]["id"]
+
+ result = uuid_search(uuid)
+ if result["code"] == 200:
+ results_list[i] = result["text"]
+ time.sleep(3)
+ return results_list
+
+
+def get_multiple_urls(urls):
+ input_values = []
+
+ for i in urls:
+ input_values.append(i[1])
+ with concurrent.futures.ThreadPoolExecutor() as executor:
+ # 使用map方法并发地调用worker_function
+ results = list(executor.map(request_api_chatgpt, input_values))
+
+ with concurrent.futures.ThreadPoolExecutor() as executor:
+ # 使用map方法并发地调用worker_function
+ results = list(executor.map(uuid_search_mp, [results]))
+
+ return_list = []
+ for i in results[0]:
+ data = {
+ "code": 200,
+ "data": i
+ }
+ return_list.append(data)
+ return return_list
+
+
+@app.route("/articles_directory", methods=["POST"])
+def articles_directory():
+ title = request.json["title"] # 获取用户query中的文本 例如"I love you"
+ nums_catalogue = request.json["nums_catalogue"]
+ nums_word = request.json["nums_word"] # 获取用户query中的文本 例如"I love you"
+ nums_catalogue = int(nums_catalogue)
+ keyword = tuple([title] + nums_word_dict[nums_word]["title_small_nums"])
+ prompt = nums_word_dict[nums_word]["prompt"]%keyword
+ print(prompt)
+ input_data = []
+ for i in range(nums_catalogue):
+ input_data.append([i, prompt])
+
+ results = get_multiple_urls(input_data)
+ log.log('text:{},'.format(prompt))
+ return jsonify(results) # 返回结果
+
+if __name__ == "__main__":
+ app.run(debug=False, host='0.0.0.0', port=18002)
diff --git a/gunicorn_config_word.py b/gunicorn_config_word.py
new file mode 100644
index 0000000..b19db78
--- /dev/null
+++ b/gunicorn_config_word.py
@@ -0,0 +1,21 @@
+# 并行工作线程数
+workers = 2
+# 监听内网端口5000【按需要更改】
+bind = '0.0.0.0:18001'
+
+loglevel = 'debug'
+
+worker_class = "gevent"
+# 设置守护进程【关闭连接时，程序仍在运行】
+daemon = True
+# 设置超时时间120s，默认为30s。按自己的需求进行设置
+timeout = 120
+# 设置访问日志和错误信息日志路径
+accesslog = './logs/acess_word.log'
+errorlog = './logs/error_word.log'
+# access_log_format = '%(h) - %(t)s - %(u)s - %(s)s %(H)s'
+# errorlog = '-' # 记录到标准输出
+
+
+# 设置最大并发量
+worker_connections = 20000
diff --git a/gunicorn_config_word_online.py b/gunicorn_config_word_online.py
new file mode 100644
index 0000000..d69be7b
--- /dev/null
+++ b/gunicorn_config_word_online.py
@@ -0,0 +1,21 @@
+# 并行工作线程数
+workers = 2
+# 监听内网端口5000【按需要更改】
+bind = '0.0.0.0:18002'
+
+loglevel = 'debug'
+
+worker_class = "gevent"
+# 设置守护进程【关闭连接时，程序仍在运行】
+daemon = True
+# 设置超时时间120s，默认为30s。按自己的需求进行设置
+timeout = 120
+# 设置访问日志和错误信息日志路径
+accesslog = './logs/acess_word_online.log'
+errorlog = './logs/error_word_online.log'
+# access_log_format = '%(h) - %(t)s - %(u)s - %(s)s %(H)s'
+# errorlog = '-' # 记录到标准输出
+
+
+# 设置最大并发量
+worker_connections = 20000
diff --git a/run_api_gunicorn_word.sh b/run_api_gunicorn_word.sh
new file mode 100644
index 0000000..b35abd7
--- /dev/null
+++ b/run_api_gunicorn_word.sh
@@ -0,0 +1 @@
+gunicorn articles_directory_predict_word:app -c gunicorn_config_word.py
diff --git a/run_api_gunicorn_word_online.sh b/run_api_gunicorn_word_online.sh
new file mode 100644
index 0000000..9cdbf74
--- /dev/null
+++ b/run_api_gunicorn_word_online.sh
@@ -0,0 +1 @@
+gunicorn articles_directory_predict_word_online:app -c gunicorn_config_word_online.py
\ No newline at end of file