复制链接
克隆策略
In [12]:
df = m11.data.read()
#分箱
col_name = 'pe_ttm_bins'
def cal_bins(df):
    bins=5
    df[col_name] = np.array(pd.qcut(df.pe_ttm_0, bins, labels=range(0, bins)))
    return df
df = df.groupby('date').apply(cal_bins)
display("分箱结果:",df[['date','instrument','pe_ttm_0','pe_ttm_bins']])
#换成one-hot编码
df = pd.get_dummies(df,columns=[col_name])
display("one-hot编码结果:",df[['date','instrument','pe_ttm_0','pe_ttm_bins_0','pe_ttm_bins_1','pe_ttm_bins_2','pe_ttm_bins_3']])
'分箱结果:'
date instrument pe_ttm_0 pe_ttm_bins
0 2010-01-04 000001.SZA 78.793991 3
0 2011-01-04 000001.SZA 9.111742 0
0 2012-01-04 000001.SZA 8.408521 0
0 2013-01-04 000001.SZA 6.385547 0
0 2014-01-02 000001.SZA 6.746183 0
... ... ... ... ...
564577 2012-12-25 603993.SHA 35.913765 3
564578 2012-12-26 603993.SHA 37.037422 3
564579 2012-12-27 603993.SHA 35.568024 3
564580 2012-12-28 603993.SHA 35.654461 3
564581 2012-12-31 603993.SHA 36.000198 3

2606084 rows × 4 columns

'one-hot编码结果:'
date instrument pe_ttm_0 pe_ttm_bins_0 pe_ttm_bins_1 pe_ttm_bins_2 pe_ttm_bins_3
0 2010-01-04 000001.SZA 78.793991 0 0 0 1
0 2011-01-04 000001.SZA 9.111742 1 0 0 0
0 2012-01-04 000001.SZA 8.408521 1 0 0 0
0 2013-01-04 000001.SZA 6.385547 1 0 0 0
0 2014-01-02 000001.SZA 6.746183 1 0 0 0
... ... ... ... ... ... ... ...
564577 2012-12-25 603993.SHA 35.913765 0 0 0 1
564578 2012-12-26 603993.SHA 37.037422 0 0 0 1
564579 2012-12-27 603993.SHA 35.568024 0 0 0 1
564580 2012-12-28 603993.SHA 35.654461 0 0 0 1
564581 2012-12-31 603993.SHA 36.000198 0 0 0 1

2606084 rows × 7 columns

In [16]:
df = m15.data_1.read()
df[['date','instrument','pe_ttm_0','pe_ttm_bins_0','pe_ttm_bins_1','pe_ttm_bins_2','pe_ttm_bins_3']]
Out[16]:
date instrument pe_ttm_0 pe_ttm_bins_0 pe_ttm_bins_1 pe_ttm_bins_2 pe_ttm_bins_3
0 2010-01-04 000001.SZA 78.793991 0 0 0 1
0 2011-01-04 000001.SZA 9.111742 1 0 0 0
0 2012-01-04 000001.SZA 8.408521 1 0 0 0
0 2013-01-04 000001.SZA 6.385547 1 0 0 0
0 2014-01-02 000001.SZA 6.746183 1 0 0 0
... ... ... ... ... ... ... ...
564577 2012-12-25 603993.SHA 35.913765 0 0 0 1
564578 2012-12-26 603993.SHA 37.037422 0 0 0 1
564579 2012-12-27 603993.SHA 35.568024 0 0 0 1
564580 2012-12-28 603993.SHA 35.654461 0 0 0 1
564581 2012-12-31 603993.SHA 36.000198 0 0 0 1

2606084 rows × 7 columns

In [17]:
df = m15.data_2.read()
df
Out[17]:
['return_5',
 'return_10',
 'return_20',
 'avg_amount_0/avg_amount_5',
 'avg_amount_5/avg_amount_20',
 'rank_avg_amount_0/rank_avg_amount_5',
 'rank_avg_amount_5/rank_avg_amount_10',
 'rank_return_0',
 'rank_return_5',
 'rank_return_10',
 'rank_return_0/rank_return_5',
 'rank_return_5/rank_return_10',
 'pe_ttm_bins_0',
 'pe_ttm_bins_1',
 'pe_ttm_bins_2',
 'pe_ttm_bins_3',
 'pe_ttm_bins_4']

    {"description":"实验创建于2020/4/10","graph":{"edges":[{"to_node_id":"-526:instruments","from_node_id":"-513:data"},{"to_node_id":"-565:instruments","from_node_id":"-513:data"},{"to_node_id":"-526:features","from_node_id":"-521:data"},{"to_node_id":"-541:features","from_node_id":"-521:data"},{"to_node_id":"-548:features","from_node_id":"-521:data"},{"to_node_id":"-557:features","from_node_id":"-521:data"},{"to_node_id":"-792:input_2","from_node_id":"-521:data"},{"to_node_id":"-548:input_data","from_node_id":"-526:data"},{"to_node_id":"-541:instruments","from_node_id":"-532:data"},{"to_node_id":"-614:instruments","from_node_id":"-532:data"},{"to_node_id":"-557:input_data","from_node_id":"-541:data"},{"to_node_id":"-576:data2","from_node_id":"-548:data"},{"to_node_id":"-593:input_data","from_node_id":"-557:data"},{"to_node_id":"-576:data1","from_node_id":"-565:data"},{"to_node_id":"-590:input_data","from_node_id":"-576:data"},{"to_node_id":"-792:input_1","from_node_id":"-590:data"},{"to_node_id":"-2878:input_1","from_node_id":"-593:data"},{"to_node_id":"-635:input_ds","from_node_id":"-600:predictions"},{"to_node_id":"-614:options_data","from_node_id":"-635:sorted_data"},{"to_node_id":"-600:training_ds","from_node_id":"-792:data_1"},{"to_node_id":"-600:features","from_node_id":"-792:data_2"},{"to_node_id":"-600:predict_ds","from_node_id":"-2878:data_1"}],"nodes":[{"node_id":"-513","module_id":"BigQuantSpace.instruments.instruments-v2","parameters":[{"name":"start_date","value":"2010-01-01","type":"Literal","bound_global_parameter":null},{"name":"end_date","value":"2014-12-31","type":"Literal","bound_global_parameter":null},{"name":"market","value":"CN_STOCK_A","type":"Literal","bound_global_parameter":null},{"name":"instrument_list","value":"","type":"Literal","bound_global_parameter":null},{"name":"max_count","value":0,"type":"Literal","bound_global_parameter":null}],"input_ports":[{"name":"rolling_conf","node_id":"-513"}],"output_ports":[{"name":"data","node_id":"-513"}],"cacheable":true,"seq_num":1,"comment":"","comment_collapsed":true},{"node_id":"-521","module_id":"BigQuantSpace.input_features.input_features-v1","parameters":[{"name":"features","value":"\n# #号开始的表示注释,注释需单独一行\n# 多个特征,每行一个,可以包含基础特征和衍生特征,特征须为本平台特征\nreturn_5\nreturn_10\nreturn_20\navg_amount_0/avg_amount_5\navg_amount_5/avg_amount_20\nrank_avg_amount_0/rank_avg_amount_5\nrank_avg_amount_5/rank_avg_amount_10\nrank_return_0\nrank_return_5\nrank_return_10\nrank_return_0/rank_return_5\nrank_return_5/rank_return_10\npe_ttm_0\n","type":"Literal","bound_global_parameter":null}],"input_ports":[{"name":"features_ds","node_id":"-521"}],"output_ports":[{"name":"data","node_id":"-521"}],"cacheable":true,"seq_num":2,"comment":"","comment_collapsed":true},{"node_id":"-526","module_id":"BigQuantSpace.general_feature_extractor.general_feature_extractor-v7","parameters":[{"name":"start_date","value":"","type":"Literal","bound_global_parameter":null},{"name":"end_date","value":"","type":"Literal","bound_global_parameter":null},{"name":"before_start_days","value":90,"type":"Literal","bound_global_parameter":null}],"input_ports":[{"name":"instruments","node_id":"-526"},{"name":"features","node_id":"-526"}],"output_ports":[{"name":"data","node_id":"-526"}],"cacheable":true,"seq_num":3,"comment":"","comment_collapsed":true},{"node_id":"-532","module_id":"BigQuantSpace.instruments.instruments-v2","parameters":[{"name":"start_date","value":"2015-01-01","type":"Literal","bound_global_parameter":null},{"name":"end_date","value":"2017-01-01","type":"Literal","bound_global_parameter":null},{"name":"market","value":"CN_STOCK_A","type":"Literal","bound_global_parameter":null},{"name":"instrument_list","value":"","type":"Literal","bound_global_parameter":null},{"name":"max_count","value":0,"type":"Literal","bound_global_parameter":null}],"input_ports":[{"name":"rolling_conf","node_id":"-532"}],"output_ports":[{"name":"data","node_id":"-532"}],"cacheable":true,"seq_num":4,"comment":"","comment_collapsed":true},{"node_id":"-541","module_id":"BigQuantSpace.general_feature_extractor.general_feature_extractor-v7","parameters":[{"name":"start_date","value":"","type":"Literal","bound_global_parameter":null},{"name":"end_date","value":"","type":"Literal","bound_global_parameter":null},{"name":"before_start_days","value":90,"type":"Literal","bound_global_parameter":null}],"input_ports":[{"name":"instruments","node_id":"-541"},{"name":"features","node_id":"-541"}],"output_ports":[{"name":"data","node_id":"-541"}],"cacheable":true,"seq_num":5,"comment":"","comment_collapsed":true},{"node_id":"-548","module_id":"BigQuantSpace.derived_feature_extractor.derived_feature_extractor-v3","parameters":[{"name":"date_col","value":"date","type":"Literal","bound_global_parameter":null},{"name":"instrument_col","value":"instrument","type":"Literal","bound_global_parameter":null},{"name":"drop_na","value":"False","type":"Literal","bound_global_parameter":null},{"name":"remove_extra_columns","value":"False","type":"Literal","bound_global_parameter":null},{"name":"user_functions","value":"{}","type":"Literal","bound_global_parameter":null}],"input_ports":[{"name":"input_data","node_id":"-548"},{"name":"features","node_id":"-548"}],"output_ports":[{"name":"data","node_id":"-548"}],"cacheable":true,"seq_num":6,"comment":"","comment_collapsed":true},{"node_id":"-557","module_id":"BigQuantSpace.derived_feature_extractor.derived_feature_extractor-v3","parameters":[{"name":"date_col","value":"date","type":"Literal","bound_global_parameter":null},{"name":"instrument_col","value":"instrument","type":"Literal","bound_global_parameter":null},{"name":"drop_na","value":"False","type":"Literal","bound_global_parameter":null},{"name":"remove_extra_columns","value":"False","type":"Literal","bound_global_parameter":null},{"name":"user_functions","value":"{}","type":"Literal","bound_global_parameter":null}],"input_ports":[{"name":"input_data","node_id":"-557"},{"name":"features","node_id":"-557"}],"output_ports":[{"name":"data","node_id":"-557"}],"cacheable":true,"seq_num":7,"comment":"","comment_collapsed":true},{"node_id":"-565","module_id":"BigQuantSpace.advanced_auto_labeler.advanced_auto_labeler-v2","parameters":[{"name":"label_expr","value":"# #号开始的表示注释\n# 0. 每行一个,顺序执行,从第二个开始,可以使用label字段\n# 1. 可用数据字段见 https://bigquant.com/docs/develop/datasource/deprecated/history_data.html\n# 添加benchmark_前缀,可使用对应的benchmark数据\n# 2. 可用操作符和函数见 `表达式引擎 <https://bigquant.com/docs/develop/bigexpr/usage.html>`_\n\n# 计算收益:5日收盘价(作为卖出价格)除以明日开盘价(作为买入价格)\nshift(close, -5) / shift(open, -1)\n\n# 极值处理:用1%和99%分位的值做clip\nclip(label, all_quantile(label, 0.01), all_quantile(label, 0.99))\n\n# 将分数映射到分类,这里使用20个分类\n#all_wbins(label, 20)\n\n# 过滤掉一字涨停的情况 (设置label为NaN,在后续处理和训练中会忽略NaN的label)\nwhere(shift(high, -1) == shift(low, -1), NaN, label)\n","type":"Literal","bound_global_parameter":null},{"name":"start_date","value":"","type":"Literal","bound_global_parameter":null},{"name":"end_date","value":"","type":"Literal","bound_global_parameter":null},{"name":"benchmark","value":"000300.SHA","type":"Literal","bound_global_parameter":null},{"name":"drop_na_label","value":"True","type":"Literal","bound_global_parameter":null},{"name":"cast_label_int","value":"False","type":"Literal","bound_global_parameter":null},{"name":"user_functions","value":"{}","type":"Literal","bound_global_parameter":null}],"input_ports":[{"name":"instruments","node_id":"-565"}],"output_ports":[{"name":"data","node_id":"-565"}],"cacheable":false,"seq_num":8,"comment":"","comment_collapsed":true},{"node_id":"-576","module_id":"BigQuantSpace.join.join-v3","parameters":[{"name":"on","value":"date,instrument","type":"Literal","bound_global_parameter":null},{"name":"how","value":"inner","type":"Literal","bound_global_parameter":null},{"name":"sort","value":"False","type":"Literal","bound_global_parameter":null}],"input_ports":[{"name":"data1","node_id":"-576"},{"name":"data2","node_id":"-576"}],"output_ports":[{"name":"data","node_id":"-576"}],"cacheable":true,"seq_num":9,"comment":"","comment_collapsed":true},{"node_id":"-590","module_id":"BigQuantSpace.dropnan.dropnan-v1","parameters":[],"input_ports":[{"name":"input_data","node_id":"-590"}],"output_ports":[{"name":"data","node_id":"-590"}],"cacheable":true,"seq_num":11,"comment":"","comment_collapsed":true},{"node_id":"-593","module_id":"BigQuantSpace.dropnan.dropnan-v1","parameters":[],"input_ports":[{"name":"input_data","node_id":"-593"}],"output_ports":[{"name":"data","node_id":"-593"}],"cacheable":true,"seq_num":10,"comment":"","comment_collapsed":true},{"node_id":"-600","module_id":"BigQuantSpace.random_forest_regressor.random_forest_regressor-v1","parameters":[{"name":"iterations","value":10,"type":"Literal","bound_global_parameter":null},{"name":"feature_fraction","value":1,"type":"Literal","bound_global_parameter":null},{"name":"max_depth","value":30,"type":"Literal","bound_global_parameter":null},{"name":"min_samples_per_leaf","value":200,"type":"Literal","bound_global_parameter":null},{"name":"key_cols","value":"date,instrument","type":"Literal","bound_global_parameter":null},{"name":"workers","value":1,"type":"Literal","bound_global_parameter":null},{"name":"random_state","value":0,"type":"Literal","bound_global_parameter":null},{"name":"other_train_parameters","value":"{\"random_state\":0}","type":"Literal","bound_global_parameter":null}],"input_ports":[{"name":"training_ds","node_id":"-600"},{"name":"features","node_id":"-600"},{"name":"model","node_id":"-600"},{"name":"predict_ds","node_id":"-600"}],"output_ports":[{"name":"output_model","node_id":"-600"},{"name":"predictions","node_id":"-600"}],"cacheable":true,"seq_num":12,"comment":"","comment_collapsed":true},{"node_id":"-614","module_id":"BigQuantSpace.trade.trade-v4","parameters":[{"name":"start_date","value":"","type":"Literal","bound_global_parameter":null},{"name":"end_date","value":"","type":"Literal","bound_global_parameter":null},{"name":"initialize","value":"# 回测引擎:初始化函数,只执行一次\ndef bigquant_run(context):\n # 加载预测数据\n context.ranker_prediction = context.options['data'].read_df()\n\n # 系统已经设置了默认的交易手续费和滑点,要修改手续费可使用如下函数\n context.set_commission(PerOrder(buy_cost=0.0003, sell_cost=0.0013, min_cost=5))\n # 预测数据,通过options传入进来,使用 read_df 函数,加载到内存 (DataFrame)\n # 设置买入的股票数量,这里买入预测股票列表排名靠前的5只\n stock_count = 5\n # 每只的股票的权重,如下的权重分配会使得靠前的股票分配多一点的资金,[0.339160, 0.213986, 0.169580, ..]\n context.stock_weights = T.norm([1 / math.log(i + 2) for i in range(0, stock_count)])\n # 设置每只股票占用的最大资金比例\n context.max_cash_per_instrument = 0.2\n context.hold_days = 5\n","type":"Literal","bound_global_parameter":null},{"name":"handle_data","value":"# 回测引擎:每日数据处理函数,每天执行一次\ndef bigquant_run(context, data):\n # 按日期过滤得到今日的预测数据\n ranker_prediction = context.ranker_prediction[\n context.ranker_prediction.date == data.current_dt.strftime('%Y-%m-%d')]\n\n # 1. 资金分配\n # 平均持仓时间是hold_days,每日都将买入股票,每日预期使用 1/hold_days 的资金\n # 实际操作中,会存在一定的买入误差,所以在前hold_days天,等量使用资金;之后,尽量使用剩余资金(这里设置最多用等量的1.5倍)\n is_staging = context.trading_day_index < context.hold_days # 是否在建仓期间(前 hold_days 天)\n cash_avg = context.portfolio.portfolio_value / context.hold_days\n cash_for_buy = min(context.portfolio.cash, (1 if is_staging else 1.5) * cash_avg)\n cash_for_sell = cash_avg - (context.portfolio.cash - cash_for_buy)\n positions = {e.symbol: p.amount * p.last_sale_price\n for e, p in context.perf_tracker.position_tracker.positions.items()}\n\n # 2. 生成卖出订单:hold_days天之后才开始卖出;对持仓的股票,按StockRanker预测的排序末位淘汰\n if not is_staging and cash_for_sell > 0:\n equities = {e.symbol: e for e, p in context.perf_tracker.position_tracker.positions.items()}\n instruments = list(reversed(list(ranker_prediction.instrument[ranker_prediction.instrument.apply(\n lambda x: x in equities and not context.has_unfinished_sell_order(equities[x]))])))\n # print('rank order for sell %s' % instruments)\n for instrument in instruments:\n context.order_target(context.symbol(instrument), 0)\n cash_for_sell -= positions[instrument]\n if cash_for_sell <= 0:\n break\n\n # 3. 生成买入订单:按StockRanker预测的排序,买入前面的stock_count只股票\n buy_cash_weights = context.stock_weights\n buy_instruments = list(ranker_prediction.instrument[:len(buy_cash_weights)])\n max_cash_per_instrument = context.portfolio.portfolio_value * context.max_cash_per_instrument\n for i, instrument in enumerate(buy_instruments):\n cash = cash_for_buy * buy_cash_weights[i]\n if cash > max_cash_per_instrument - positions.get(instrument, 0):\n # 确保股票持仓量不会超过每次股票最大的占用资金量\n cash = max_cash_per_instrument - positions.get(instrument, 0)\n if cash > 0:\n context.order_value(context.symbol(instrument), cash)\n","type":"Literal","bound_global_parameter":null},{"name":"prepare","value":"# 回测引擎:准备数据,只执行一次\ndef bigquant_run(context):\n pass\n","type":"Literal","bound_global_parameter":null},{"name":"before_trading_start","value":"# 回测引擎:每个单位时间开始前调用一次,即每日开盘前调用一次。\ndef bigquant_run(context, data):\n pass\n","type":"Literal","bound_global_parameter":null},{"name":"volume_limit","value":0.025,"type":"Literal","bound_global_parameter":null},{"name":"order_price_field_buy","value":"open","type":"Literal","bound_global_parameter":null},{"name":"order_price_field_sell","value":"close","type":"Literal","bound_global_parameter":null},{"name":"capital_base","value":1000000,"type":"Literal","bound_global_parameter":null},{"name":"auto_cancel_non_tradable_orders","value":"True","type":"Literal","bound_global_parameter":null},{"name":"data_frequency","value":"daily","type":"Literal","bound_global_parameter":null},{"name":"price_type","value":"真实价格","type":"Literal","bound_global_parameter":null},{"name":"product_type","value":"股票","type":"Literal","bound_global_parameter":null},{"name":"plot_charts","value":"True","type":"Literal","bound_global_parameter":null},{"name":"backtest_only","value":"False","type":"Literal","bound_global_parameter":null},{"name":"benchmark","value":"","type":"Literal","bound_global_parameter":null}],"input_ports":[{"name":"instruments","node_id":"-614"},{"name":"options_data","node_id":"-614"},{"name":"history_ds","node_id":"-614"},{"name":"benchmark_ds","node_id":"-614"},{"name":"trading_calendar","node_id":"-614"}],"output_ports":[{"name":"raw_perf","node_id":"-614"}],"cacheable":false,"seq_num":13,"comment":"","comment_collapsed":true},{"node_id":"-635","module_id":"BigQuantSpace.sort.sort-v4","parameters":[{"name":"sort_by","value":"pred_label","type":"Literal","bound_global_parameter":null},{"name":"group_by","value":"date","type":"Literal","bound_global_parameter":null},{"name":"keep_columns","value":"--","type":"Literal","bound_global_parameter":null},{"name":"ascending","value":"False","type":"Literal","bound_global_parameter":null}],"input_ports":[{"name":"input_ds","node_id":"-635"},{"name":"sort_by_ds","node_id":"-635"}],"output_ports":[{"name":"sorted_data","node_id":"-635"}],"cacheable":true,"seq_num":14,"comment":"","comment_collapsed":true},{"node_id":"-792","module_id":"BigQuantSpace.cached.cached-v3","parameters":[{"name":"run","value":"# Python 代码入口函数,input_1/2/3 对应三个输入端,data_1/2/3 对应三个输出端\ndef bigquant_run(input_1, input_2, input_3):\n # 示例代码如下。在这里编写您的代码\n #分箱\n col_name = 'pe_ttm_bins'\n def cal_bins(df):\n bins=5\n df[col_name] = np.array(pd.qcut(df.pe_ttm_0, bins, labels=range(0, bins)))\n return df\n df = input_1.read().groupby('date').apply(cal_bins)\n #换成one-hot编码\n df = pd.get_dummies(df,columns=[col_name])\n cols = df.columns\n need_cols = [c for c in cols if col_name in c]\n features = input_2.read()\n features = features + need_cols\n features.remove('pe_ttm_0')\n features = DataSource.write_pickle(features)\n return Outputs(data_1=DataSource.write_df(df), data_2=features, data_3=None)\n","type":"Literal","bound_global_parameter":null},{"name":"post_run","value":"# 后处理函数,可选。输入是主函数的输出,可以在这里对数据做处理,或者返回更友好的outputs数据格式。此函数输出不会被缓存。\ndef bigquant_run(outputs):\n return outputs\n","type":"Literal","bound_global_parameter":null},{"name":"input_ports","value":"","type":"Literal","bound_global_parameter":null},{"name":"params","value":"{}","type":"Literal","bound_global_parameter":null},{"name":"output_ports","value":"","type":"Literal","bound_global_parameter":null}],"input_ports":[{"name":"input_1","node_id":"-792"},{"name":"input_2","node_id":"-792"},{"name":"input_3","node_id":"-792"}],"output_ports":[{"name":"data_1","node_id":"-792"},{"name":"data_2","node_id":"-792"},{"name":"data_3","node_id":"-792"}],"cacheable":true,"seq_num":15,"comment":"","comment_collapsed":true},{"node_id":"-2878","module_id":"BigQuantSpace.cached.cached-v3","parameters":[{"name":"run","value":"# Python 代码入口函数,input_1/2/3 对应三个输入端,data_1/2/3 对应三个输出端\ndef bigquant_run(input_1, input_2, input_3):\n # 示例代码如下。在这里编写您的代码\n col_name = 'pe_ttm_bins'\n def cal_bins(df):\n bins=5\n df[col_name] = np.array(pd.qcut(df.pe_ttm_0, bins, labels=range(0, bins)))\n return df\n df = input_1.read().groupby('date').apply(cal_bins)\n df = pd.get_dummies(df,columns=[col_name])\n# cols = df.columns\n# need_cols = [c for c in cols if col_name in c]\n# features = input_2.read()\n# features = features + need_cols\n# features = DataSource.write_pickle(features)\n return Outputs(data_1=DataSource.write_df(df), data_2=None, data_3=None)\n","type":"Literal","bound_global_parameter":null},{"name":"post_run","value":"# 后处理函数,可选。输入是主函数的输出,可以在这里对数据做处理,或者返回更友好的outputs数据格式。此函数输出不会被缓存。\ndef bigquant_run(outputs):\n return outputs\n","type":"Literal","bound_global_parameter":null},{"name":"input_ports","value":"","type":"Literal","bound_global_parameter":null},{"name":"params","value":"{}","type":"Literal","bound_global_parameter":null},{"name":"output_ports","value":"","type":"Literal","bound_global_parameter":null}],"input_ports":[{"name":"input_1","node_id":"-2878"},{"name":"input_2","node_id":"-2878"},{"name":"input_3","node_id":"-2878"}],"output_ports":[{"name":"data_1","node_id":"-2878"},{"name":"data_2","node_id":"-2878"},{"name":"data_3","node_id":"-2878"}],"cacheable":true,"seq_num":16,"comment":"","comment_collapsed":true}],"node_layout":"<node_postions><node_position Node='-513' Position='-13,53,200,200'/><node_position Node='-521' Position='311,16.788578033447266,200,200'/><node_position Node='-526' Position='123,157,200,200'/><node_position Node='-532' Position='641,23,200,200'/><node_position Node='-541' Position='579,153,200,200'/><node_position Node='-548' Position='134,243,200,200'/><node_position Node='-557' Position='581,264,200,200'/><node_position Node='-565' Position='-177,162,200,200'/><node_position Node='-576' Position='-3,314,200,200'/><node_position Node='-590' Position='2,386,200,200'/><node_position Node='-593' Position='590,354,200,200'/><node_position Node='-600' Position='183.42282104492188,577.05712890625,200,200'/><node_position Node='-614' Position='309,739,200,200'/><node_position Node='-635' Position='410,655,200,200'/><node_position Node='-792' Position='171,483,200,200'/><node_position Node='-2878' Position='572,467,200,200'/></node_postions>"},"nodes_readonly":false,"studio_version":"v2"}
    In [2]:
    # 本代码由可视化策略环境自动生成 2022年11月5日 15:11
    # 本代码单元只能在可视化模式下编辑。您也可以拷贝代码,粘贴到新建的代码单元或者策略,然后修改。
    
    
    # Python 代码入口函数,input_1/2/3 对应三个输入端,data_1/2/3 对应三个输出端
    def m15_run_bigquant_run(input_1, input_2, input_3):
        # 示例代码如下。在这里编写您的代码
        #分箱
        col_name = 'pe_ttm_bins'
        def cal_bins(df):
            bins=5
            df[col_name] = np.array(pd.qcut(df.pe_ttm_0, bins, labels=range(0, bins)))
            return df
        df = input_1.read().groupby('date').apply(cal_bins)
        #换成one-hot编码
        df = pd.get_dummies(df,columns=[col_name])
        cols = df.columns
        need_cols = [c for c in cols if col_name in c]
        features = input_2.read()
        features = features + need_cols
        features.remove('pe_ttm_0')
        features = DataSource.write_pickle(features)
        return Outputs(data_1=DataSource.write_df(df), data_2=features, data_3=None)
    
    # 后处理函数,可选。输入是主函数的输出,可以在这里对数据做处理,或者返回更友好的outputs数据格式。此函数输出不会被缓存。
    def m15_post_run_bigquant_run(outputs):
        return outputs
    
    # Python 代码入口函数,input_1/2/3 对应三个输入端,data_1/2/3 对应三个输出端
    def m16_run_bigquant_run(input_1, input_2, input_3):
        # 示例代码如下。在这里编写您的代码
        col_name = 'pe_ttm_bins'
        def cal_bins(df):
            bins=5
            df[col_name] = np.array(pd.qcut(df.pe_ttm_0, bins, labels=range(0, bins)))
            return df
        df = input_1.read().groupby('date').apply(cal_bins)
        df = pd.get_dummies(df,columns=[col_name])
    #     cols = df.columns
    #     need_cols = [c for c in cols if col_name in c]
    #     features = input_2.read()
    #     features = features + need_cols
    #     features = DataSource.write_pickle(features)
        return Outputs(data_1=DataSource.write_df(df), data_2=None, data_3=None)
    
    # 后处理函数,可选。输入是主函数的输出,可以在这里对数据做处理,或者返回更友好的outputs数据格式。此函数输出不会被缓存。
    def m16_post_run_bigquant_run(outputs):
        return outputs
    
    # 回测引擎:初始化函数,只执行一次
    def m13_initialize_bigquant_run(context):
        # 加载预测数据
        context.ranker_prediction = context.options['data'].read_df()
    
        # 系统已经设置了默认的交易手续费和滑点,要修改手续费可使用如下函数
        context.set_commission(PerOrder(buy_cost=0.0003, sell_cost=0.0013, min_cost=5))
        # 预测数据,通过options传入进来,使用 read_df 函数,加载到内存 (DataFrame)
        # 设置买入的股票数量,这里买入预测股票列表排名靠前的5只
        stock_count = 5
        # 每只的股票的权重,如下的权重分配会使得靠前的股票分配多一点的资金,[0.339160, 0.213986, 0.169580, ..]
        context.stock_weights = T.norm([1 / math.log(i + 2) for i in range(0, stock_count)])
        # 设置每只股票占用的最大资金比例
        context.max_cash_per_instrument = 0.2
        context.hold_days = 5
    
    # 回测引擎:每日数据处理函数,每天执行一次
    def m13_handle_data_bigquant_run(context, data):
        # 按日期过滤得到今日的预测数据
        ranker_prediction = context.ranker_prediction[
            context.ranker_prediction.date == data.current_dt.strftime('%Y-%m-%d')]
    
        # 1. 资金分配
        # 平均持仓时间是hold_days,每日都将买入股票,每日预期使用 1/hold_days 的资金
        # 实际操作中,会存在一定的买入误差,所以在前hold_days天,等量使用资金;之后,尽量使用剩余资金(这里设置最多用等量的1.5倍)
        is_staging = context.trading_day_index < context.hold_days # 是否在建仓期间(前 hold_days 天)
        cash_avg = context.portfolio.portfolio_value / context.hold_days
        cash_for_buy = min(context.portfolio.cash, (1 if is_staging else 1.5) * cash_avg)
        cash_for_sell = cash_avg - (context.portfolio.cash - cash_for_buy)
        positions = {e.symbol: p.amount * p.last_sale_price
                     for e, p in context.perf_tracker.position_tracker.positions.items()}
    
        # 2. 生成卖出订单:hold_days天之后才开始卖出;对持仓的股票,按StockRanker预测的排序末位淘汰
        if not is_staging and cash_for_sell > 0:
            equities = {e.symbol: e for e, p in context.perf_tracker.position_tracker.positions.items()}
            instruments = list(reversed(list(ranker_prediction.instrument[ranker_prediction.instrument.apply(
                    lambda x: x in equities and not context.has_unfinished_sell_order(equities[x]))])))
            # print('rank order for sell %s' % instruments)
            for instrument in instruments:
                context.order_target(context.symbol(instrument), 0)
                cash_for_sell -= positions[instrument]
                if cash_for_sell <= 0:
                    break
    
        # 3. 生成买入订单:按StockRanker预测的排序,买入前面的stock_count只股票
        buy_cash_weights = context.stock_weights
        buy_instruments = list(ranker_prediction.instrument[:len(buy_cash_weights)])
        max_cash_per_instrument = context.portfolio.portfolio_value * context.max_cash_per_instrument
        for i, instrument in enumerate(buy_instruments):
            cash = cash_for_buy * buy_cash_weights[i]
            if cash > max_cash_per_instrument - positions.get(instrument, 0):
                # 确保股票持仓量不会超过每次股票最大的占用资金量
                cash = max_cash_per_instrument - positions.get(instrument, 0)
            if cash > 0:
                context.order_value(context.symbol(instrument), cash)
    
    # 回测引擎:准备数据,只执行一次
    def m13_prepare_bigquant_run(context):
        pass
    
    # 回测引擎:每个单位时间开始前调用一次,即每日开盘前调用一次。
    def m13_before_trading_start_bigquant_run(context, data):
        pass
    
    
    m1 = M.instruments.v2(
        start_date='2010-01-01',
        end_date='2014-12-31',
        market='CN_STOCK_A',
        instrument_list='',
        max_count=0
    )
    
    m8 = M.advanced_auto_labeler.v2(
        instruments=m1.data,
        label_expr="""# #号开始的表示注释
    # 0. 每行一个,顺序执行,从第二个开始,可以使用label字段
    # 1. 可用数据字段见 https://bigquant.com/docs/develop/datasource/deprecated/history_data.html
    #   添加benchmark_前缀,可使用对应的benchmark数据
    # 2. 可用操作符和函数见 `表达式引擎 <https://bigquant.com/docs/develop/bigexpr/usage.html>`_
    
    # 计算收益:5日收盘价(作为卖出价格)除以明日开盘价(作为买入价格)
    shift(close, -5) / shift(open, -1)
    
    # 极值处理:用1%和99%分位的值做clip
    clip(label, all_quantile(label, 0.01), all_quantile(label, 0.99))
    
    # 将分数映射到分类,这里使用20个分类
    #all_wbins(label, 20)
    
    # 过滤掉一字涨停的情况 (设置label为NaN,在后续处理和训练中会忽略NaN的label)
    where(shift(high, -1) == shift(low, -1), NaN, label)
    """,
        start_date='',
        end_date='',
        benchmark='000300.SHA',
        drop_na_label=True,
        cast_label_int=False,
        user_functions={},
        m_cached=False
    )
    
    m2 = M.input_features.v1(
        features="""
    # #号开始的表示注释,注释需单独一行
    # 多个特征,每行一个,可以包含基础特征和衍生特征,特征须为本平台特征
    return_5
    return_10
    return_20
    avg_amount_0/avg_amount_5
    avg_amount_5/avg_amount_20
    rank_avg_amount_0/rank_avg_amount_5
    rank_avg_amount_5/rank_avg_amount_10
    rank_return_0
    rank_return_5
    rank_return_10
    rank_return_0/rank_return_5
    rank_return_5/rank_return_10
    pe_ttm_0
    """
    )
    
    m3 = M.general_feature_extractor.v7(
        instruments=m1.data,
        features=m2.data,
        start_date='',
        end_date='',
        before_start_days=90
    )
    
    m6 = M.derived_feature_extractor.v3(
        input_data=m3.data,
        features=m2.data,
        date_col='date',
        instrument_col='instrument',
        drop_na=False,
        remove_extra_columns=False,
        user_functions={}
    )
    
    m9 = M.join.v3(
        data1=m8.data,
        data2=m6.data,
        on='date,instrument',
        how='inner',
        sort=False
    )
    
    m11 = M.dropnan.v1(
        input_data=m9.data
    )
    
    m15 = M.cached.v3(
        input_1=m11.data,
        input_2=m2.data,
        run=m15_run_bigquant_run,
        post_run=m15_post_run_bigquant_run,
        input_ports='',
        params='{}',
        output_ports=''
    )
    
    m4 = M.instruments.v2(
        start_date='2015-01-01',
        end_date='2017-01-01',
        market='CN_STOCK_A',
        instrument_list='',
        max_count=0
    )
    
    m5 = M.general_feature_extractor.v7(
        instruments=m4.data,
        features=m2.data,
        start_date='',
        end_date='',
        before_start_days=90
    )
    
    m7 = M.derived_feature_extractor.v3(
        input_data=m5.data,
        features=m2.data,
        date_col='date',
        instrument_col='instrument',
        drop_na=False,
        remove_extra_columns=False,
        user_functions={}
    )
    
    m10 = M.dropnan.v1(
        input_data=m7.data
    )
    
    m16 = M.cached.v3(
        input_1=m10.data,
        run=m16_run_bigquant_run,
        post_run=m16_post_run_bigquant_run,
        input_ports='',
        params='{}',
        output_ports=''
    )
    
    m12 = M.random_forest_regressor.v1(
        training_ds=m15.data_1,
        features=m15.data_2,
        predict_ds=m16.data_1,
        iterations=10,
        feature_fraction=1,
        max_depth=30,
        min_samples_per_leaf=200,
        key_cols='date,instrument',
        workers=1,
        random_state=0,
        other_train_parameters={"random_state":0}
    )
    
    m14 = M.sort.v4(
        input_ds=m12.predictions,
        sort_by='pred_label',
        group_by='date',
        keep_columns='--',
        ascending=False
    )
    
    m13 = M.trade.v4(
        instruments=m4.data,
        options_data=m14.sorted_data,
        start_date='',
        end_date='',
        initialize=m13_initialize_bigquant_run,
        handle_data=m13_handle_data_bigquant_run,
        prepare=m13_prepare_bigquant_run,
        before_trading_start=m13_before_trading_start_bigquant_run,
        volume_limit=0.025,
        order_price_field_buy='open',
        order_price_field_sell='close',
        capital_base=1000000,
        auto_cancel_non_tradable_orders=True,
        data_frequency='daily',
        price_type='真实价格',
        product_type='股票',
        plot_charts=True,
        backtest_only=False,
        benchmark=''
    )
    
    • 收益率199.14%
    • 年化收益率76.09%
    • 基准收益率-6.33%
    • 阿尔法0.9
    • 贝塔0.99
    • 夏普比率1.45
    • 胜率0.6
    • 盈亏比0.85
    • 收益波动率43.53%
    • 信息比率0.14
    • 最大回撤48.37%
    bigcharts-data-start/{"__type":"tabs","__id":"bigchart-d3dae2e7094646df8feb3af092d77ab0"}/bigcharts-data-end