克隆策略

In [2]:

# 本代码由可视化策略环境自动生成 2021年8月12日 16:32
# 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。


# 回测引擎：初始化函数，只执行一次
def m6_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 = 3
    # 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 = 1
    context.options['hold_days'] = 0
    #在数据准备函数中一次性计算每日的大盘风控条件相比于在handle中每日计算风控条件可以提高回测速度
    # 多取50天的数据便于计算均值(保证回测的第一天均值不为Nan值)，其中context.start_date和context.end_date是回测指定的起始时间和终止时间
    start_date= (pd.to_datetime(context.start_date) - datetime.timedelta(days=50)).strftime('%Y-%m-%d') 
    df=DataSource('bar1d_index_CN_STOCK_A').read(start_date=start_date,end_date=context.end_date,fields=['close'])

    #这里以上证指数000001.HIX为例
    benchmark_data=df[df.instrument=='000001.HIX']
    #计算上证指数5日涨幅
    benchmark_data['ret5']=benchmark_data['close']/benchmark_data['close'].shift(3)-1
    #计算大盘风控条件，如果5日涨幅小于-4%则设置风险状态risk为1，否则为0
    benchmark_data['risk'] = np.where(benchmark_data['ret5']<-0.04,1,0)
    #修改日期格式为字符串(便于在handle中使用字符串日期索引来查看每日的风险状态)
    benchmark_data['date']=benchmark_data['date'].apply(lambda x:x.strftime('%Y-%m-%d'))
    #设置日期为索引
    benchmark_data.set_index('date',inplace=True)
    #把风控序列输出给全局变量context.benchmark_risk
    context.benchmark_risk=benchmark_data['risk']
# 回测引擎：每日数据处理函数，每天执行一次
def m6_handle_data_bigquant_run(context, data):
# 相隔几天（hold_days）进行一下换仓
    if context.trading_day_index % 1 != 0:
        return 
    
    # 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')]
    # 目前持仓
    positions = {e.symbol: p.amount * p.last_sale_price
                 for e, p in context.portfolio.positions.items()}
    # 权重
    buy_cash_weights = context.stock_weights
    # 今日买入股票列表
    stock_to_buy = list(ranker_prediction.instrument[:len(buy_cash_weights)])
    # 持仓上限
    max_cash_per_instrument = context.portfolio.portfolio_value * context.max_cash_per_instrument

    # 通过positions对象，使用列表生成式的方法获取目前持仓的股票列表
    stock_hold_now = [equity.symbol for equity in context.portfolio.positions ]
    # 继续持有的股票：调仓时，如果买入的股票已经存在于目前的持仓里，那么应继续持有
    no_need_to_sell = [i for i in stock_hold_now if i in stock_to_buy]
    # 需要卖出的股票
    stock_to_sell = [i for i in stock_hold_now if i not in no_need_to_sell]
  
    # 卖出
    for stock in stock_to_sell:
        # 如果该股票停牌，则没法成交。因此需要用can_trade方法检查下该股票的状态
        # 如果返回真值，则可以正常下单，否则会出错
        # 因为stock是字符串格式，我们用symbol方法将其转化成平台可以接受的形式：Equity格式
        if data.can_trade(context.symbol(stock)):
            # order_target_percent是平台的一个下单接口，表明下单使得该股票的权重为0，
            #   即卖出全部股票，可参考回测文档
            context.order_target_percent(context.symbol(stock), 0)
    
    # 如果当天没有买入的股票，就返回
    if len(stock_to_buy) == 0:
        return
    # 3. 生成买入订单：按机器学习算法预测的排序，买入前面的stock_count只股票
    # 买入
    for i, instrument in enumerate(stock_to_buy):
        cash = context.portfolio.portfolio_value * 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:
            price = data.current(context.symbol(instrument), 'price')  # 最新价格
            stock_num = np.floor(cash/price/100)*100  # 向下取整
            context.order(context.symbol(instrument), stock_num) # 整百下单
# 回测引擎：准备数据，只执行一次
def m6_prepare_bigquant_run(context):
    pass


g = T.Graph({

    'm1': 'M.instruments.v2',
    'm1.start_date': '2016-01-01',
    'm1.end_date': '2018-01-01',
    'm1.market': 'CN_STOCK_A',
    'm1.instrument_list': '',
    'm1.max_count': 0,

    'm2': 'M.advanced_auto_labeler.v2',
    'm2.instruments': T.Graph.OutputPort('m1.data'),
    'm2.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, -2) / 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)

""",
    'm2.start_date': '',
    'm2.end_date': '',
    'm2.benchmark': '000300.SHA',
    'm2.drop_na_label': True,
    'm2.cast_label_int': True,

    'm3': 'M.input_features.v1',
    'm3.features': """# #号开始的表示注释
# 多个特征，每行一个，可以包含基础特征和衍生特征
(close_0-low_0)/(high_0-close_0)
std(amount_0,10)
rank_pb_lf_0
rank_beta_industry1_10_0
rank_market_cap_0
(high_0-low_0)/close_1
std(mf_net_amount_l_0,20)
-1*rank(covariance(rank(high_0),rank(volume_0),5))
sum(((close_0-low_0)-(high_0-close_0))/(high_0-low_0)*volume_0,6)
-1*correlation(close_0-shift(close_0,1), amount_0/avg_amount_20,20)
-1*fs_total_liability_0/fs_total_operating_costs_0
sum(max(0,high_0-shift((high_0+low_0+close_0)/3,1)),26)/sum(max(0,delay((high_0+low_0+close_0)/3,1) - low_0), 26)*100
fs_net_profit_margin_0
close_0-shift(close_0, 5)
""",

    'm15': 'M.general_feature_extractor.v7',
    'm15.instruments': T.Graph.OutputPort('m1.data'),
    'm15.features': T.Graph.OutputPort('m3.data'),
    'm15.start_date': '',
    'm15.end_date': '',
    'm15.before_start_days': 90,

    'm16': 'M.derived_feature_extractor.v3',
    'm16.input_data': T.Graph.OutputPort('m15.data'),
    'm16.features': T.Graph.OutputPort('m3.data'),
    'm16.date_col': 'date',
    'm16.instrument_col': 'instrument',
    'm16.drop_na': False,
    'm16.remove_extra_columns': False,

    'm7': 'M.join.v3',
    'm7.data1': T.Graph.OutputPort('m2.data'),
    'm7.data2': T.Graph.OutputPort('m16.data'),
    'm7.on': 'date,instrument',
    'm7.how': 'inner',
    'm7.sort': False,

    'm11': 'M.chinaa_stock_filter.v1',
    'm11.input_data': T.Graph.OutputPort('m7.data'),
    'm11.index_constituent_cond': ['全部'],
    'm11.board_cond': ['上证主板', '深证主板'],
    'm11.industry_cond': ['全部', '交通运输', '公用事业', '化工', '商业贸易', '建筑装饰', '房地产', '机械设备', '汽车/交运设备', '纺织服装'],
    'm11.st_cond': ['正常'],
    'm11.delist_cond': ['全部'],
    'm11.output_left_data': False,

    'm19': 'M.filter.v3',
    'm19.input_data': T.Graph.OutputPort('m11.data'),
    'm19.expr': 'close_0<50',
    'm19.output_left_data': False,

    'm5': 'M.dropnan.v2',
    'm5.input_data': T.Graph.OutputPort('m7.data'),

    'm4': 'M.stock_ranker_train.v6',
    'm4.training_ds': T.Graph.OutputPort('m5.data'),
    'm4.features': T.Graph.OutputPort('m3.data'),
    'm4.learning_algorithm': '排序',
    'm4.number_of_leaves': 30,
    'm4.minimum_docs_per_leaf': 1000,
    'm4.number_of_trees': 20,
    'm4.learning_rate': 0.1,
    'm4.max_bins': 1023,
    'm4.feature_fraction': 1,
    'm4.data_row_fraction': 1,
    'm4.ndcg_discount_base': 1,
    'm4.m_lazy_run': False,

    'm9': 'M.instruments.v2',
    'm9.start_date': T.live_run_param('trading_date', '2019-01-02'),
    'm9.end_date': T.live_run_param('trading_date', '2020-12-27'),
    'm9.market': 'CN_STOCK_A',
    'm9.instrument_list': '000001.HIX',
    'm9.max_count': 0,

    'm17': 'M.general_feature_extractor.v7',
    'm17.instruments': T.Graph.OutputPort('m9.data'),
    'm17.features': T.Graph.OutputPort('m3.data'),
    'm17.start_date': '',
    'm17.end_date': '',
    'm17.before_start_days': 90,

    'm18': 'M.derived_feature_extractor.v3',
    'm18.input_data': T.Graph.OutputPort('m17.data'),
    'm18.features': T.Graph.OutputPort('m3.data'),
    'm18.date_col': 'date',
    'm18.instrument_col': 'instrument',
    'm18.drop_na': False,
    'm18.remove_extra_columns': False,

    'm20': 'M.chinaa_stock_filter.v1',
    'm20.input_data': T.Graph.OutputPort('m18.data'),
    'm20.index_constituent_cond': ['全部'],
    'm20.board_cond': ['上证主板', '深证主板'],
    'm20.industry_cond': ['全部', '交通运输', '公用事业', '化工', '商业贸易', '建筑装饰', '房地产', '机械设备', '汽车/交运设备', '纺织服装'],
    'm20.st_cond': ['正常'],
    'm20.delist_cond': ['全部'],
    'm20.output_left_data': False,

    'm21': 'M.filter.v3',
    'm21.input_data': T.Graph.OutputPort('m20.data'),
    'm21.expr': 'close_0<50',
    'm21.output_left_data': False,

    'm10': 'M.dropnan.v2',
    'm10.input_data': T.Graph.OutputPort('m18.data'),

    'm8': 'M.stock_ranker_predict.v5',
    'm8.model': T.Graph.OutputPort('m4.model'),
    'm8.data': T.Graph.OutputPort('m10.data'),
    'm8.m_lazy_run': False,

    'm6': 'M.trade.v4',
    'm6.instruments': T.Graph.OutputPort('m9.data'),
    'm6.options_data': T.Graph.OutputPort('m8.predictions'),
    'm6.start_date': '',
    'm6.end_date': '',
    'm6.initialize': m6_initialize_bigquant_run,
    'm6.handle_data': m6_handle_data_bigquant_run,
    'm6.prepare': m6_prepare_bigquant_run,
    'm6.volume_limit': 0.025,
    'm6.order_price_field_buy': 'open',
    'm6.order_price_field_sell': 'close',
    'm6.capital_base': 1000000,
    'm6.auto_cancel_non_tradable_orders': True,
    'm6.data_frequency': 'daily',
    'm6.price_type': '真实价格',
    'm6.product_type': '股票',
    'm6.plot_charts': True,
    'm6.backtest_only': False,
    'm6.benchmark': '',

    'm13': 'M.strategy_ret_risk_analysis.v1',
    'm13.backtest_ds': T.Graph.OutputPort('m6.raw_perf'),
    'm13.benchmark_symbol': '000300.HIX',
    'm13.feature_window': 60,

    'm23': 'M.input_features.v1',
    'm23.features': """# #号开始的表示注释
# 多个特征，每行一个，可以包含基础特征和衍生特征
rank(((stddev(abs((close_0 - open_0)), 5) + (close_0 - open_0)) + correlation(close_0, open_0,10)))
correlation(close_0-shift(close_0,1), amount_0/avg_amount_20,20)
rank_pb_lf_0
rank_fs_roe_ttm_0
fs_total_liability_0/fs_total_operating_costs_0
rank_market_cap_0
rank_beta_industry1_10_0
fs_net_profit_margin_0
std(amount_0,10)
sum(((close_0-low_0)-(high_0-close_0))/(high_0-low_0)*volume_0,6)
mf_net_pct_main_0
""",

    'm14': 'M.input_features.v1',
    'm14.features': """# #号开始的表示注释
# 多个特征，每行一个，可以包含基础特征和衍生特征
correlation(amount_0, close_0, 10)
(close_0-low_0)-(high_0-close_0)
std(amount_0,10)
-1*correlation(close_0-shift(close_0,1), amount_0/avg_amount_20,20)
pb_lf_0
-1*fs_total_liability_0/fs_total_operating_costs_0
rank_beta_industry1_10_0
fs_eqy_belongto_parcomsh_0
rank_market_cap_0
(high_0-low_0)/close_1
mean(mf_net_amount_l_0,20)
std(mf_net_amount_l_0,20)
-1*rank(covariance(rank(high_0),rank(volume_0),5))""",

    'm25': 'M.input_features.v1',
    'm25.features': """# #号开始的表示注释
# 多个特征，每行一个，可以包含基础特征和衍生特征
(close_0-low_0)/(high_0-close_0)
std(amount_0,10)
rank_pb_lf_0
rank_beta_industry1_10_0
rank_market_cap_0
(high_0-low_0)/close_1
mean(mf_net_amount_l_0,3)/mean(mf_net_amount_0,30)
std(mf_net_amount_l_0,20)
-1*rank(covariance(rank(high_0),rank(volume_0),5))
rank_beta_industry1_10_0
sum(((close_0-low_0)-(high_0-close_0))/(high_0-low_0)*volume_0,6)
-1*correlation(close_0-shift(close_0,1), amount_0/avg_amount_20,20)
-1*fs_total_liability_0/fs_total_operating_costs_0
rank_market_cap_0""",
})

# g.run({})


def m22_run_bigquant_run(
    bq_graph,
    inputs,
    trading_days_market='CN', # 使用那个市场的交易日历, TODO
    train_instruments_mid='m1', # 训练数据 证券代码列表 模块id
    test_instruments_mid='m9', # 测试数据 证券代码列表 模块id
    predict_mid='m8', # 预测 模块id
    trade_mid='m6', # 回测 模块id
    start_date='2018-01-01', # 数据开始日期
    end_date=T.live_run_param('trading_date', '2019-12-31'), # 数据结束日期
    train_update_days=30, # 更新周期，按交易日计算，每多少天更新一次
    train_update_days_for_live=None, #模拟实盘模式下的更新周期，按交易日计算，每多少天更新一次。如果需要在模拟实盘阶段使用不同的模型更新周期，可以设置这个参数
    train_data_min_days=360, # 最小数据天数，按交易日计算，所以第一个滚动的结束日期是 从开始日期到开始日期+最小数据天数
    train_data_max_days=370, # 最大数据天数，按交易日计算，0，表示没有限制，否则每一个滚动的开始日期=max(此滚动的结束日期-最大数据天数, 开始日期
    rolling_count_for_live=0, #实盘模式下滚动次数，模拟实盘模式下，取最后多少次滚动。一般在模拟实盘模式下，只用到最后一次滚动训练的模型，这里可以设置为1；如果你的滚动训练数据时间段很短，以至于期间可能没有训练数据，这里可以设置大一点。0表示没有限制
):
    def merge_datasources(input_1):
        df_list = [ds[0].read_df().set_index('date').ix[ds[1]:].reset_index() for ds in input_1]
        df = pd.concat(df_list)
        instrument_data = {
            'start_date': df['date'].min().strftime('%Y-%m-%d'),
            'end_date': df['date'].max().strftime('%Y-%m-%d'),
            'instruments': list(set(df['instrument'])),
        }
        return Outputs(data=DataSource.write_df(df), instrument_data=DataSource.write_pickle(instrument_data))

    def gen_rolling_dates(trading_days_market, start_date, end_date, train_update_days, train_update_days_for_live, train_data_min_days, train_data_max_days, rolling_count_for_live):
        # 是否实盘模式
        tdays = list(D.trading_days(market=trading_days_market, start_date=start_date, end_date=end_date)['date'])
        is_live_run = T.live_run_param('trading_date', None) is not None

        if is_live_run and train_update_days_for_live:
            train_update_days = train_update_days_for_live

        rollings = []
        train_end_date = train_data_min_days
        while train_end_date < len(tdays):
            if train_data_max_days is not None and train_data_max_days > 0:
                train_start_date = max(train_end_date - train_data_max_days, 0)
            else:
                train_start_date = 0
            rollings.append({
                'train_start_date': tdays[train_start_date].strftime('%Y-%m-%d'),
                'train_end_date': tdays[train_end_date - 1].strftime('%Y-%m-%d'),
                'test_start_date': tdays[train_end_date].strftime('%Y-%m-%d'),
                'test_end_date': tdays[min(train_end_date + train_update_days, len(tdays)) - 1].strftime('%Y-%m-%d'),
            })
            train_end_date += train_update_days

        if not rollings:
            raise Exception('没有滚动需要执行，请检查配置')

        if is_live_run and rolling_count_for_live:
            rollings = rollings[-rolling_count_for_live:]

        return rollings

    g = bq_graph

    rolling_dates = gen_rolling_dates(
        trading_days_market, start_date, end_date, train_update_days, train_update_days_for_live, train_data_min_days, train_data_max_days, rolling_count_for_live)

    # 训练和预测
    results = []
    for rolling in rolling_dates:
        parameters = {}
        # 先禁用回测
        parameters[trade_mid + '.__enabled__'] = False
        parameters[train_instruments_mid + '.start_date'] = rolling['train_start_date']
        parameters[train_instruments_mid + '.end_date'] = rolling['train_end_date']
        parameters[test_instruments_mid + '.start_date'] = rolling['test_start_date']
        parameters[test_instruments_mid + '.end_date'] = rolling['test_end_date']
        # print('------ rolling_train:', parameters)
        results.append(g.run(parameters))

    # 合并预测结果并回测
    mx = M.cached.v3(run=merge_datasources, input_1=[[result[predict_mid].predictions, result[test_instruments_mid].data.read_pickle()['start_date']] for result in results])
    parameters = {}
    parameters['*.__enabled__'] = False
    parameters[trade_mid + '.__enabled__'] = True
    parameters[trade_mid + '.instruments'] = mx.instrument_data
    parameters[trade_mid + '.options_data'] = mx.data

    trade = g.run(parameters)

    return {'rollings': results, 'trade': trade}


m22 = M.hyper_rolling_train.v1(
    run=m22_run_bigquant_run,
    run_now=True,
    bq_graph=g
)

[2021-07-21 21:32:34.822478] INFO: moduleinvoker: instruments.v2 开始运行..

[2021-07-21 21:32:34.921670] INFO: moduleinvoker: 命中缓存

[2021-07-21 21:32:34.924634] INFO: moduleinvoker: instruments.v2 运行完成[0.10217s].

[2021-07-21 21:32:34.964528] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2021-07-21 21:32:34.976522] INFO: moduleinvoker: 命中缓存

[2021-07-21 21:32:35.031785] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.067257s].

[2021-07-21 21:32:35.131834] INFO: moduleinvoker: input_features.v1 开始运行..

[2021-07-21 21:32:35.140080] INFO: moduleinvoker: 命中缓存

[2021-07-21 21:32:35.141862] INFO: moduleinvoker: input_features.v1 运行完成[0.010046s].

[2021-07-21 21:32:36.859865] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2021-07-21 21:32:36.868730] INFO: moduleinvoker: 命中缓存

[2021-07-21 21:32:36.871162] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.01132s].

[2021-07-21 21:32:36.926006] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2021-07-21 21:32:36.939490] INFO: moduleinvoker: 命中缓存

[2021-07-21 21:32:36.941990] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.015997s].

[2021-07-21 21:32:37.001883] INFO: moduleinvoker: join.v3 开始运行..

[2021-07-21 21:32:37.075987] INFO: moduleinvoker: 命中缓存

[2021-07-21 21:32:37.077785] INFO: moduleinvoker: join.v3 运行完成[0.075939s].

[2021-07-21 21:32:37.148174] INFO: moduleinvoker: dropnan.v2 开始运行..

[2021-07-21 21:32:37.455821] INFO: dropnan: /y_2015, 0/0

[2021-07-21 21:32:41.789465] INFO: dropnan: /y_2016, 620002/637477

[2021-07-21 21:32:46.845831] INFO: dropnan: /y_2017, 702555/731481

[2021-07-21 21:32:47.387670] INFO: dropnan: 行数: 1322557/1368958

[2021-07-21 21:32:47.429796] INFO: moduleinvoker: dropnan.v2 运行完成[10.281609s].

[2021-07-21 21:32:47.497725] INFO: moduleinvoker: stock_ranker_train.v6 开始运行..

[2021-07-21 21:32:50.157173] INFO: StockRanker: 特征预处理 ..

[2021-07-21 21:32:52.632013] INFO: StockRanker: prepare data: training ..

[2021-07-21 21:32:55.143571] INFO: StockRanker: sort ..

[2021-07-21 21:33:19.470463] INFO: StockRanker训练: 2333aebe 准备训练: 1322557 行数

[2021-07-21 21:33:19.472398] INFO: StockRanker训练: AI模型训练，将在1322557*14=1851.58万数据上对模型训练进行20轮迭代训练。预计将需要6～13分钟。请耐心等待。

[2021-07-21 21:33:19.741927] INFO: StockRanker训练: 正在训练 ..

[2021-07-21 21:33:19.805136] INFO: StockRanker训练: 任务状态: Pending

[2021-07-21 21:33:29.857026] INFO: StockRanker训练: 任务状态: Running

[2021-07-21 21:33:39.888836] INFO: StockRanker训练: 00:00:08.8711422, finished iteration 1

[2021-07-21 21:33:39.892612] INFO: StockRanker训练: 00:00:15.9000916, finished iteration 2

[2021-07-21 21:33:49.941898] INFO: StockRanker训练: 00:00:23.4427136, finished iteration 3

[2021-07-21 21:33:59.977562] INFO: StockRanker训练: 00:00:32.0373576, finished iteration 4

[2021-07-21 21:34:10.024608] INFO: StockRanker训练: 00:00:41.1924034, finished iteration 5

[2021-07-21 21:34:20.064442] INFO: StockRanker训练: 00:00:50.2298460, finished iteration 6

[2021-07-21 21:34:30.106522] INFO: StockRanker训练: 00:00:59.4240147, finished iteration 7

[2021-07-21 21:34:40.153176] INFO: StockRanker训练: 00:01:08.6659225, finished iteration 8

[2021-07-21 21:34:50.192843] INFO: StockRanker训练: 00:01:18.6820141, finished iteration 9

[2021-07-21 21:35:00.237636] INFO: StockRanker训练: 00:01:30.4158521, finished iteration 10

[2021-07-21 21:35:10.289336] INFO: StockRanker训练: 00:01:41.2848800, finished iteration 11

[2021-07-21 21:35:20.332441] INFO: StockRanker训练: 00:01:53.8934966, finished iteration 12

[2021-07-21 21:35:30.376069] INFO: StockRanker训练: 00:02:07.4634009, finished iteration 13

[2021-07-21 21:35:50.469824] INFO: StockRanker训练: 00:02:20.9509568, finished iteration 14

[2021-07-21 21:36:00.514256] INFO: StockRanker训练: 00:02:33.1622538, finished iteration 15

[2021-07-21 21:36:10.571722] INFO: StockRanker训练: 00:02:46.6574687, finished iteration 16

[2021-07-21 21:36:30.660778] INFO: StockRanker训练: 00:03:01.2531978, finished iteration 17

[2021-07-21 21:36:40.706300] INFO: StockRanker训练: 00:03:13.7828312, finished iteration 18

[2021-07-21 21:36:50.762542] INFO: StockRanker训练: 00:03:25.9563024, finished iteration 19

[2021-07-21 21:37:00.812631] INFO: StockRanker训练: 00:03:37.7273646, finished iteration 20

[2021-07-21 21:37:10.857228] INFO: StockRanker训练: 任务状态: Succeeded

[2021-07-21 21:37:11.148537] INFO: moduleinvoker: stock_ranker_train.v6 运行完成[263.651849s].

[2021-07-21 21:37:11.164947] INFO: moduleinvoker: instruments.v2 开始运行..

[2021-07-21 21:37:11.231697] INFO: moduleinvoker: instruments.v2 运行完成[0.066733s].

[2021-07-21 21:37:11.242134] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2021-07-21 21:37:15.073219] INFO: 基础特征抽取: 年份 2018, 特征行数=0

[2021-07-21 21:37:19.767106] INFO: 基础特征抽取: 年份 2019, 特征行数=0

[2021-07-21 21:37:24.427006] INFO: 基础特征抽取: 年份 2020, 特征行数=0

[2021-07-21 21:37:27.654689] ERROR: moduleinvoker: module name: general_feature_extractor, module version: v7, trackeback: Exception: no features extracted.

设置评估测试数据集，查看训练曲线
[视频教程]StockRanker训练曲线

bigcharts-data-start/{"__type":"tabs","__id":"bigchart-906c2945671b45fea41e2ac70231c8a8"}/bigcharts-data-end

---------------------------------------------------------------------------
Exception                                 Traceback (most recent call last)
<ipython-input-2-61124cbf5f79> in <module>
    197 )
    198 
--> 199 m17 = M.general_feature_extractor.v7(
    200     instruments=m9.data,
    201     features=m3.data,

Exception: no features extracted.