# 存 GeneralBrowserActionInstance
if save_doc:
general_browser_action_doc = GeneralBrowserActionInstance(
uuid=action_uuid,
related_articles=related_articles,
new_found_articles=new_found_articles,
pdf_page_snapshot=BinaryAttachment(uuid=orig_pdf_bin_uuid),
img_page_snapshot=BinaryAttachment(uuid=orig_png_bin_uuid))
upsert_document(general_browser_action_doc, False)
if __name__ == "__main__":
from gs_research_workflow.common.mongo_resource import mongo_db_conn, used_db_position, db_nlp
mongo_db_conn(used_db_position, db_nlp)
pd.set_option("display.max_rows", 20)
pd.set_option("display.max_columns", None)
# GeneralBrowserBackendProcess.process_action_result(
# "/tmp/laigen/debug_data/general_desktop_browser_backend_action/xueqiu", "", False)
GeneralBrowserBackendProcess.process_action_result(
"/tmp/laigen/debug_data/general_desktop_browser_backend_action/google_news_tab",
"debug_batch_uuid", "debug_action_uuid_1", True)
# import dateparser
#
# ls_s = ["3 hours ago", "22-Jun-20", "05-29 23:24 ", "Thu, Mar. 19", "Fri, May 22", "Aug. 19, 2019",
# "1分钟前", "3天前", "5小时前", "57m", "5h", "Yesterday, 2:07 PM", "Today, 12:58 AM", "5 years ago"]
# for s in ls_s:
"pred_val": df_y_pred_orig_val.iloc[-1]
}).dropna()
df_pred_summary[
"delta_v"] = df_pred_summary["true_val"] - df_pred_summary["pred_val"]
df_pred_summary["delta_percentage"] = (df_pred_summary["true_val"] - df_pred_summary["pred_val"]) * 100. / \
df_pred_summary["true_val"]
df_pred_zscore = pd.DataFrame({
"true_val": df_zscore.iloc[-1],
"pred_val": df_y_pred.iloc[-1]
}).dropna()
print(df_pred_summary)
mongo_db_conn("intranet", db_nlp)
import pandas as pd
news_objs = NewsIndex.objects(from_keyword__in=SearchKeyword.objects(
agent__in=AgentWithTag.objects(tags="govt_central")))
df = pd.DataFrame(data=[{
"title": news.title,
"pub_time": news.publish_time,
"abstract": news.abstract,
"url": news.url,
"from_kw": news.from_keyword.search_keyword
} for news in news_objs])
print(df)
# print(news_objs)
# news_objs = NewsIndex.objects(from_keyword__in=SearchKeyword.objects(agent="Trump"))