克隆策略

In [17]:

m20.data.read().shape

Out[17]:

(806430, 14)

In [18]:

m29.data.read().shape

Out[18]:

(806382, 14)

In [19]:

# 本代码由可视化策略环境自动生成 2020年10月19日 09:25
# 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 m19_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 = 1
    # 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 = 0.5
    context.options['hold_days'] = 1

# 回测引擎：每日数据处理函数，每天执行一次
def m19_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.options['hold_days'] # 是否在建仓期间（前 hold_days 天）
    cash_avg = context.portfolio.portfolio_value / context.options['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.portfolio.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，按机器学习算法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.portfolio.positions.items()}
        instruments = list(reversed(list(ranker_prediction.instrument[ranker_prediction.instrument.apply(
                lambda x: x in equities)])))

        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：按机器学习算法预测的排序，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 m19_prepare_bigquant_run(context):
    pass

# 回测引擎：初始化函数，只执行一次
def m27_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 = 1
    # 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 = 0.5
    context.options['hold_days'] = 1

# 回测引擎：每日数据处理函数，每天执行一次
def m27_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.options['hold_days'] # 是否在建仓期间（前 hold_days 天）
    cash_avg = context.portfolio.portfolio_value / context.options['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.portfolio.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，按机器学习算法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.portfolio.positions.items()}
        instruments = list(reversed(list(ranker_prediction.instrument[ranker_prediction.instrument.apply(
                lambda x: x in equities)])))

        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：按机器学习算法预测的排序，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 m27_prepare_bigquant_run(context):
    pass


m1 = M.instruments.v2(
    start_date='2018-01-01',
    end_date='2019-01-01',
    market='CN_STOCK_A',
    instrument_list='',
    max_count=0
)

m2 = 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, -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)
""",
    start_date='',
    end_date='',
    benchmark='000300.SHA',
    drop_na_label=True,
    cast_label_int=True
)

m3 = M.input_features.v1(
    features="""close_1
open_1
high_1
low_1"""
)

m15 = M.general_feature_extractor.v7(
    instruments=m1.data,
    features=m3.data,
    start_date='',
    end_date='',
    before_start_days=90
)

m16 = M.derived_feature_extractor.v3(
    input_data=m15.data,
    features=m3.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False
)

m7 = M.join.v3(
    data1=m2.data,
    data2=m16.data,
    on='date,instrument',
    how='inner',
    sort=False
)

m5 = M.dropnan.v2(
    input_data=m7.data
)

m4 = M.stock_ranker_train.v6(
    training_ds=m5.data,
    features=m3.data,
    learning_algorithm='排序',
    number_of_leaves=30,
    minimum_docs_per_leaf=1000,
    number_of_trees=20,
    learning_rate=0.1,
    max_bins=1023,
    feature_fraction=1,
    data_row_fraction=1,
    ndcg_discount_base=1,
    m_lazy_run=False
)

m11 = M.input_features.v1(
    features_ds=m3.data,
    features='jin4=close_0 - open_0'
)

m9 = M.instruments.v2(
    start_date=T.live_run_param('trading_date', '2019-01-01'),
    end_date=T.live_run_param('trading_date', '2020-01-01'),
    market='CN_STOCK_A',
    instrument_list='',
    max_count=0
)

m17 = M.general_feature_extractor.v7(
    instruments=m9.data,
    features=m11.data,
    start_date='',
    end_date='',
    before_start_days=90
)

m18 = M.derived_feature_extractor.v3(
    input_data=m17.data,
    features=m11.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False
)

m10 = M.dropnan.v2(
    input_data=m18.data
)

m12 = M.filter.v3(
    input_data=m10.data,
    expr='jin4>0',
    output_left_data=False
)

m8 = M.stock_ranker_predict.v5(
    model=m4.model,
    data=m12.data,
    m_lazy_run=False
)

m19 = M.trade.v4(
    instruments=m9.data,
    options_data=m8.predictions,
    start_date='',
    end_date='',
    initialize=m19_initialize_bigquant_run,
    handle_data=m19_handle_data_bigquant_run,
    prepare=m19_prepare_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='000300.SHA'
)

m6 = M.instruments.v2(
    start_date='2018-01-01',
    end_date='2019-01-01',
    market='CN_STOCK_A',
    instrument_list='',
    max_count=0
)

m13 = M.advanced_auto_labeler.v2(
    instruments=m6.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, -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)
""",
    start_date='',
    end_date='',
    benchmark='000300.SHA',
    drop_na_label=True,
    cast_label_int=True
)

m14 = M.input_features.v1(
    features="""close_1
open_1
high_1
low_1"""
)

m31 = M.input_features.v1(
    features_ds=m14.data,
    features='jin4=close_0 - open_0'
)

m23 = M.general_feature_extractor.v7(
    instruments=m6.data,
    features=m31.data,
    start_date='',
    end_date='',
    before_start_days=90
)

m24 = M.derived_feature_extractor.v3(
    input_data=m23.data,
    features=m31.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False
)

m20 = M.join.v3(
    data1=m13.data,
    data2=m24.data,
    on='date,instrument',
    how='inner',
    sort=False
)

m29 = M.dropnan.v2(
    input_data=m20.data
)

m28 = M.stock_ranker_train.v6(
    training_ds=m29.data,
    features=m14.data,
    learning_algorithm='排序',
    number_of_leaves=30,
    minimum_docs_per_leaf=1000,
    number_of_trees=20,
    learning_rate=0.1,
    max_bins=1023,
    feature_fraction=1,
    data_row_fraction=1,
    ndcg_discount_base=1,
    m_lazy_run=False
)

m22 = M.instruments.v2(
    start_date='2019-01-01',
    end_date='2020-01-01',
    market='CN_STOCK_A',
    instrument_list='',
    max_count=0
)

m25 = M.general_feature_extractor.v7(
    instruments=m22.data,
    features=m31.data,
    start_date='',
    end_date='',
    before_start_days=90
)

m26 = M.derived_feature_extractor.v3(
    input_data=m25.data,
    features=m31.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False
)

m30 = M.dropnan.v2(
    input_data=m26.data
)

m32 = M.filter.v3(
    input_data=m30.data,
    expr='jin4>0',
    output_left_data=False
)

m21 = M.stock_ranker_predict.v5(
    model=m28.model,
    data=m32.data,
    m_lazy_run=False
)

m27 = M.trade.v4(
    instruments=m22.data,
    options_data=m21.predictions,
    start_date='',
    end_date='',
    initialize=m27_initialize_bigquant_run,
    handle_data=m27_handle_data_bigquant_run,
    prepare=m27_prepare_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='000300.SHA'
)

[2020-10-19 09:17:15.347265] INFO: moduleinvoker: instruments.v2 开始运行..

[2020-10-19 09:17:15.353903] INFO: moduleinvoker: 命中缓存

[2020-10-19 09:17:15.355028] INFO: moduleinvoker: instruments.v2 运行完成[0.007759s].

[2020-10-19 09:17:15.356626] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2020-10-19 09:17:15.361503] INFO: moduleinvoker: 命中缓存

[2020-10-19 09:17:15.362419] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.005787s].

[2020-10-19 09:17:15.363715] INFO: moduleinvoker: input_features.v1 开始运行..

[2020-10-19 09:17:15.368323] INFO: moduleinvoker: 命中缓存

[2020-10-19 09:17:15.369310] INFO: moduleinvoker: input_features.v1 运行完成[0.005587s].

[2020-10-19 09:17:15.444565] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2020-10-19 09:17:15.456982] INFO: moduleinvoker: 命中缓存

[2020-10-19 09:17:15.458410] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.01385s].

[2020-10-19 09:17:15.460238] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2020-10-19 09:17:15.466856] INFO: moduleinvoker: 命中缓存

[2020-10-19 09:17:15.468052] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.007808s].

[2020-10-19 09:17:15.469929] INFO: moduleinvoker: join.v3 开始运行..

[2020-10-19 09:17:15.474784] INFO: moduleinvoker: 命中缓存

[2020-10-19 09:17:15.475770] INFO: moduleinvoker: join.v3 运行完成[0.005836s].

[2020-10-19 09:17:15.477434] INFO: moduleinvoker: dropnan.v2 开始运行..

[2020-10-19 09:17:15.482421] INFO: moduleinvoker: 命中缓存

[2020-10-19 09:17:15.483573] INFO: moduleinvoker: dropnan.v2 运行完成[0.006122s].

[2020-10-19 09:17:15.488021] INFO: moduleinvoker: stock_ranker_train.v6 开始运行..

[2020-10-19 09:17:15.831386] INFO: StockRanker: 特征预处理 ..

[2020-10-19 09:17:16.818652] INFO: StockRanker: prepare data: training ..

[2020-10-19 09:17:17.461327] INFO: StockRanker: sort ..

[2020-10-19 09:17:24.199184] INFO: StockRanker训练: d2e1f81a 准备训练: 806424 行数

[2020-10-19 09:17:24.200920] INFO: StockRanker训练: AI模型训练，将在806424*4=322.57万数据上对模型训练进行20轮迭代训练。预计将需要2～4分钟。请耐心等待。

[2020-10-19 09:17:24.284366] INFO: StockRanker训练: 正在训练 ..

[2020-10-19 09:17:24.363696] INFO: StockRanker训练: 任务状态: Pending

[2020-10-19 09:17:34.405106] INFO: StockRanker训练: 任务状态: Running

[2020-10-19 09:17:44.445599] INFO: StockRanker训练: 00:00:08.2071614, finished iteration 1

[2020-10-19 09:17:44.447049] INFO: StockRanker训练: 00:00:14.6791504, finished iteration 2

[2020-10-19 09:17:54.481400] INFO: StockRanker训练: 00:00:21.1765269, finished iteration 3

[2020-10-19 09:18:04.520577] INFO: StockRanker训练: 00:00:27.7166896, finished iteration 4

[2020-10-19 09:18:04.522554] INFO: StockRanker训练: 00:00:34.3312873, finished iteration 5

[2020-10-19 09:18:14.556539] INFO: StockRanker训练: 00:00:40.9807140, finished iteration 6

[2020-10-19 09:18:24.593111] INFO: StockRanker训练: 00:00:47.7762302, finished iteration 7

[2020-10-19 09:18:24.595073] INFO: StockRanker训练: 00:00:54.6482893, finished iteration 8

[2020-10-19 09:18:34.639885] INFO: StockRanker训练: 00:01:01.5577361, finished iteration 9

[2020-10-19 09:18:44.674582] INFO: StockRanker训练: 00:01:08.5528007, finished iteration 10

[2020-10-19 09:18:44.676210] INFO: StockRanker训练: 00:01:15.7152593, finished iteration 11

[2020-10-19 09:18:54.710709] INFO: StockRanker训练: 00:01:22.9380870, finished iteration 12

[2020-10-19 09:19:04.749056] INFO: StockRanker训练: 00:01:30.4589081, finished iteration 13

[2020-10-19 09:19:14.782842] INFO: StockRanker训练: 00:01:37.9441313, finished iteration 14

[2020-10-19 09:19:14.784361] INFO: StockRanker训练: 00:01:45.5032204, finished iteration 15

[2020-10-19 09:19:24.819260] INFO: StockRanker训练: 00:01:52.9336557, finished iteration 16

[2020-10-19 09:19:34.855801] INFO: StockRanker训练: 00:02:00.4956279, finished iteration 17

[2020-10-19 09:19:44.893902] INFO: StockRanker训练: 00:02:08.0603684, finished iteration 18

[2020-10-19 09:19:44.895638] INFO: StockRanker训练: 00:02:15.6367345, finished iteration 19

[2020-10-19 09:19:54.936831] INFO: StockRanker训练: 00:02:23.1370488, finished iteration 20

[2020-10-19 09:19:54.938531] INFO: StockRanker训练: 任务状态: Succeeded

[2020-10-19 09:19:55.082482] INFO: moduleinvoker: stock_ranker_train.v6 运行完成[159.594419s].

[2020-10-19 09:19:55.085015] INFO: moduleinvoker: input_features.v1 开始运行..

[2020-10-19 09:19:55.137360] INFO: moduleinvoker: input_features.v1 运行完成[0.052323s].

[2020-10-19 09:19:55.143474] INFO: moduleinvoker: instruments.v2 开始运行..

[2020-10-19 09:19:55.149669] INFO: moduleinvoker: 命中缓存

[2020-10-19 09:19:55.151095] INFO: moduleinvoker: instruments.v2 运行完成[0.007623s].

[2020-10-19 09:19:55.157948] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2020-10-19 09:19:56.059785] INFO: 基础特征抽取: 年份 2018, 特征行数=210561

[2020-10-19 09:19:57.847359] INFO: 基础特征抽取: 年份 2019, 特征行数=884867

[2020-10-19 09:19:58.362318] INFO: 基础特征抽取: 年份 2020, 特征行数=0

[2020-10-19 09:19:58.751108] INFO: 基础特征抽取: 总行数: 1095428

[2020-10-19 09:19:58.755051] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[3.597095s].

[2020-10-19 09:19:58.757089] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2020-10-19 09:19:59.092081] INFO: derived_feature_extractor: 提取完成 jin4=close_0 - open_0, 0.007s

[2020-10-19 09:19:59.214049] INFO: derived_feature_extractor: /y_2018, 210561

[2020-10-19 09:19:59.532306] INFO: derived_feature_extractor: /y_2019, 884867

[2020-10-19 09:20:00.487976] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[1.73084s].

[2020-10-19 09:20:00.490853] INFO: moduleinvoker: dropnan.v2 开始运行..

[2020-10-19 09:20:00.757692] INFO: dropnan: /y_2018, 210542/210561

[2020-10-19 09:20:01.808710] INFO: dropnan: /y_2019, 884658/884867

[2020-10-19 09:20:02.085567] INFO: dropnan: 行数: 1095200/1095428

[2020-10-19 09:20:02.096451] INFO: moduleinvoker: dropnan.v2 运行完成[1.605584s].

[2020-10-19 09:20:02.099513] INFO: moduleinvoker: filter.v3 开始运行..

[2020-10-19 09:20:02.128034] INFO: filter: 使用表达式 jin4>0 过滤

[2020-10-19 09:20:02.257719] INFO: filter: 过滤 /y_2018, 103634/0/210542

[2020-10-19 09:20:02.670792] INFO: filter: 过滤 /y_2019, 437221/0/884658

[2020-10-19 09:20:02.986317] INFO: moduleinvoker: filter.v3 运行完成[0.886795s].

[2020-10-19 09:20:02.989707] INFO: moduleinvoker: stock_ranker_predict.v5 开始运行..

[2020-10-19 09:20:03.909794] INFO: StockRanker预测: /y_2018 ..

[2020-10-19 09:20:04.983672] INFO: StockRanker预测: /y_2019 ..

[2020-10-19 09:20:08.383251] INFO: moduleinvoker: stock_ranker_predict.v5 运行完成[5.393511s].

[2020-10-19 09:20:09.414982] INFO: moduleinvoker: backtest.v8 开始运行..

[2020-10-19 09:20:09.419553] INFO: backtest: biglearning backtest:V8.4.2

[2020-10-19 09:20:09.420666] INFO: backtest: product_type:stock by specified

[2020-10-19 09:20:09.530174] INFO: moduleinvoker: cached.v2 开始运行..

[2020-10-19 09:20:19.815623] INFO: backtest: 读取股票行情完成:1893451

[2020-10-19 09:20:23.482631] INFO: moduleinvoker: cached.v2 运行完成[13.952447s].

[2020-10-19 09:20:24.695298] INFO: algo: TradingAlgorithm V1.6.9

[2020-10-19 09:20:25.735181] INFO: algo: trading transform...

[2020-10-19 09:20:26.735408] INFO: algo: handle_splits get splits [dt:2019-05-17 00:00:00+00:00] [asset:Equity(2013 [603577.SHA]), ratio:0.99615079164505]

[2020-10-19 09:20:26.736705] INFO: Position: position stock handle split[sid:2013, orig_amount:1000, new_amount:1003.0, orig_cost:13.04000005083917, new_cost:12.9898, ratio:0.99615079164505, last_sale_price:12.9399995803833]

[2020-10-19 09:20:26.737734] INFO: Position: after split: PositionStock(asset:Equity(2013 [603577.SHA]), amount:1003.0, cost_basis:12.9898, last_sale_price:12.99000072479248)

[2020-10-19 09:20:26.738609] INFO: Position: returning cash: 11.1812

[2020-10-19 09:20:28.119230] INFO: Performance: Simulated 244 trading days out of 244.

[2020-10-19 09:20:28.120517] INFO: Performance: first open: 2019-01-02 09:30:00+00:00

[2020-10-19 09:20:28.121743] INFO: Performance: last close: 2019-12-31 15:00:00+00:00

[2020-10-19 09:20:33.314717] INFO: moduleinvoker: backtest.v8 运行完成[23.899728s].

[2020-10-19 09:20:33.316775] INFO: moduleinvoker: trade.v4 运行完成[24.930187s].

[2020-10-19 09:20:33.319145] INFO: moduleinvoker: instruments.v2 开始运行..

[2020-10-19 09:20:33.325238] INFO: moduleinvoker: 命中缓存

[2020-10-19 09:20:33.326572] INFO: moduleinvoker: instruments.v2 运行完成[0.007424s].

[2020-10-19 09:20:33.328421] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2020-10-19 09:20:33.334057] INFO: moduleinvoker: 命中缓存

[2020-10-19 09:20:33.335102] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.006675s].

[2020-10-19 09:20:33.336577] INFO: moduleinvoker: input_features.v1 开始运行..

[2020-10-19 09:20:33.341306] INFO: moduleinvoker: 命中缓存

[2020-10-19 09:20:33.342203] INFO: moduleinvoker: input_features.v1 运行完成[0.005625s].

[2020-10-19 09:20:33.343589] INFO: moduleinvoker: input_features.v1 开始运行..

[2020-10-19 09:20:33.348205] INFO: moduleinvoker: 命中缓存

[2020-10-19 09:20:33.349043] INFO: moduleinvoker: input_features.v1 运行完成[0.00545s].

[2020-10-19 09:20:33.354903] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2020-10-19 09:20:34.012681] INFO: 基础特征抽取: 年份 2017, 特征行数=193398

[2020-10-19 09:20:35.509839] INFO: 基础特征抽取: 年份 2018, 特征行数=816987

[2020-10-19 09:20:35.963369] INFO: 基础特征抽取: 年份 2019, 特征行数=0

[2020-10-19 09:20:36.224081] INFO: 基础特征抽取: 总行数: 1010385

[2020-10-19 09:20:36.230245] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[2.875319s].

[2020-10-19 09:20:36.232156] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2020-10-19 09:20:36.516423] INFO: derived_feature_extractor: 提取完成 jin4=close_0 - open_0, 0.004s

[2020-10-19 09:20:36.587661] INFO: derived_feature_extractor: /y_2017, 193398

[2020-10-19 09:20:36.853276] INFO: derived_feature_extractor: /y_2018, 816987

[2020-10-19 09:20:37.733752] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[1.501566s].

[2020-10-19 09:20:37.736336] INFO: moduleinvoker: join.v3 开始运行..

[2020-10-19 09:20:38.380329] INFO: join: /y_2017, 行数=0/193398, 耗时=0.3696s

[2020-10-19 09:20:39.571179] INFO: join: /y_2018, 行数=806430/816987, 耗时=1.186191s

[2020-10-19 09:20:39.988234] INFO: join: 最终行数: 806430

[2020-10-19 09:20:40.020129] INFO: moduleinvoker: join.v3 运行完成[2.283773s].

[2020-10-19 09:20:40.022910] INFO: moduleinvoker: dropnan.v2 开始运行..

[2020-10-19 09:20:40.063958] INFO: dropnan: /y_2017, 0/0

[2020-10-19 09:20:40.962573] INFO: dropnan: /y_2018, 806424/806430

[2020-10-19 09:20:41.234152] INFO: dropnan: 行数: 806424/806430

[2020-10-19 09:20:41.244705] INFO: moduleinvoker: dropnan.v2 运行完成[1.221782s].

[2020-10-19 09:20:41.246490] INFO: moduleinvoker: stock_ranker_train.v6 开始运行..

[2020-10-19 09:20:41.694488] INFO: StockRanker: 特征预处理 ..

[2020-10-19 09:20:42.028293] INFO: StockRanker: prepare data: training ..

[2020-10-19 09:20:42.793317] INFO: StockRanker: sort ..

[2020-10-19 09:20:49.338041] INFO: StockRanker训练: 4d8ac6d2 准备训练: 806424 行数

[2020-10-19 09:20:49.339222] INFO: StockRanker训练: AI模型训练，将在806424*4=322.57万数据上对模型训练进行20轮迭代训练。预计将需要2～4分钟。请耐心等待。

[2020-10-19 09:20:49.358118] INFO: StockRanker训练: 正在训练 ..

[2020-10-19 09:20:49.408876] INFO: StockRanker训练: 任务状态: Pending

[2020-10-19 09:20:59.444406] INFO: StockRanker训练: 任务状态: Running

[2020-10-19 09:21:09.478297] INFO: StockRanker训练: 00:00:08.1605583, finished iteration 1

[2020-10-19 09:21:09.479478] INFO: StockRanker训练: 00:00:14.6180183, finished iteration 2

[2020-10-19 09:21:19.526074] INFO: StockRanker训练: 00:00:21.1064929, finished iteration 3

[2020-10-19 09:21:29.563529] INFO: StockRanker训练: 00:00:27.6525338, finished iteration 4

[2020-10-19 09:21:29.565148] INFO: StockRanker训练: 00:00:34.2893192, finished iteration 5

[2020-10-19 09:21:39.598944] INFO: StockRanker训练: 00:00:40.9650214, finished iteration 6

[2020-10-19 09:21:49.640494] INFO: StockRanker训练: 00:00:47.9079590, finished iteration 7

[2020-10-19 09:21:49.641651] INFO: StockRanker训练: 00:00:54.7883720, finished iteration 8

[2020-10-19 09:21:59.676349] INFO: StockRanker训练: 00:01:01.7207653, finished iteration 9

[2020-10-19 09:22:09.711303] INFO: StockRanker训练: 00:01:08.8625450, finished iteration 10

[2020-10-19 09:22:09.713230] INFO: StockRanker训练: 00:01:15.9629242, finished iteration 11

[2020-10-19 09:22:19.749816] INFO: StockRanker训练: 00:01:23.0840056, finished iteration 12

[2020-10-19 09:22:29.786499] INFO: StockRanker训练: 00:01:30.5638073, finished iteration 13

[2020-10-19 09:22:39.820332] INFO: StockRanker训练: 00:01:38.0395249, finished iteration 14

[2020-10-19 09:22:39.821949] INFO: StockRanker训练: 00:01:45.4917661, finished iteration 15

[2020-10-19 09:22:49.862550] INFO: StockRanker训练: 00:01:52.7632459, finished iteration 16

[2020-10-19 09:22:59.897380] INFO: StockRanker训练: 00:02:00.3948108, finished iteration 17

[2020-10-19 09:23:09.931278] INFO: StockRanker训练: 00:02:07.9574828, finished iteration 18

[2020-10-19 09:23:09.932961] INFO: StockRanker训练: 00:02:15.5234539, finished iteration 19

[2020-10-19 09:23:19.971276] INFO: StockRanker训练: 00:02:23.0164767, finished iteration 20

[2020-10-19 09:23:19.972556] INFO: StockRanker训练: 任务状态: Succeeded

[2020-10-19 09:23:20.127749] INFO: moduleinvoker: stock_ranker_train.v6 运行完成[158.88126s].

[2020-10-19 09:23:20.129580] INFO: moduleinvoker: instruments.v2 开始运行..

[2020-10-19 09:23:20.135747] INFO: moduleinvoker: 命中缓存

[2020-10-19 09:23:20.137060] INFO: moduleinvoker: instruments.v2 运行完成[0.007474s].

[2020-10-19 09:23:20.142495] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2020-10-19 09:23:20.146741] INFO: moduleinvoker: 命中缓存

[2020-10-19 09:23:20.147619] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.005121s].

[2020-10-19 09:23:20.149299] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2020-10-19 09:23:20.153609] INFO: moduleinvoker: 命中缓存

[2020-10-19 09:23:20.154434] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.005131s].

[2020-10-19 09:23:20.156053] INFO: moduleinvoker: dropnan.v2 开始运行..

[2020-10-19 09:23:20.160238] INFO: moduleinvoker: 命中缓存

[2020-10-19 09:23:20.161101] INFO: moduleinvoker: dropnan.v2 运行完成[0.005043s].

[2020-10-19 09:23:20.162761] INFO: moduleinvoker: filter.v3 开始运行..

[2020-10-19 09:23:20.167567] INFO: moduleinvoker: 命中缓存

[2020-10-19 09:23:20.168425] INFO: moduleinvoker: filter.v3 运行完成[0.005658s].

[2020-10-19 09:23:20.170124] INFO: moduleinvoker: stock_ranker_predict.v5 开始运行..

[2020-10-19 09:23:20.870522] INFO: StockRanker预测: /y_2018 ..

[2020-10-19 09:23:21.995191] INFO: StockRanker预测: /y_2019 ..

[2020-10-19 09:23:24.670645] INFO: moduleinvoker: stock_ranker_predict.v5 运行完成[4.500484s].

[2020-10-19 09:23:24.721187] INFO: moduleinvoker: backtest.v8 开始运行..

[2020-10-19 09:23:24.726195] INFO: backtest: biglearning backtest:V8.4.2

[2020-10-19 09:23:24.727744] INFO: backtest: product_type:stock by specified

[2020-10-19 09:23:24.865165] INFO: moduleinvoker: cached.v2 开始运行..

[2020-10-19 09:23:24.870437] INFO: moduleinvoker: 命中缓存

[2020-10-19 09:23:24.871479] INFO: moduleinvoker: cached.v2 运行完成[0.006323s].

[2020-10-19 09:23:26.052604] INFO: algo: TradingAlgorithm V1.6.9

[2020-10-19 09:23:26.958230] INFO: algo: trading transform...

[2020-10-19 09:23:27.977315] INFO: algo: handle_splits get splits [dt:2019-05-17 00:00:00+00:00] [asset:Equity(2013 [603577.SHA]), ratio:0.99615079164505]

[2020-10-19 09:23:27.978507] INFO: Position: position stock handle split[sid:2013, orig_amount:1000, new_amount:1003.0, orig_cost:13.04000005083917, new_cost:12.9898, ratio:0.99615079164505, last_sale_price:12.9399995803833]

[2020-10-19 09:23:27.979414] INFO: Position: after split: PositionStock(asset:Equity(2013 [603577.SHA]), amount:1003.0, cost_basis:12.9898, last_sale_price:12.99000072479248)

[2020-10-19 09:23:27.980381] INFO: Position: returning cash: 11.1812

[2020-10-19 09:23:29.474298] INFO: Performance: Simulated 244 trading days out of 244.

[2020-10-19 09:23:29.475422] INFO: Performance: first open: 2019-01-02 09:30:00+00:00

[2020-10-19 09:23:29.476267] INFO: Performance: last close: 2019-12-31 15:00:00+00:00

[2020-10-19 09:23:34.266074] INFO: moduleinvoker: backtest.v8 运行完成[9.5449s].

[2020-10-19 09:23:34.267542] INFO: moduleinvoker: trade.v4 运行完成[9.594301s].

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-e91eb29e59964268a37b401950b11ea0"}/bigcharts-data-end

收益率-22.78%
年化收益率-23.43%
基准收益率36.07%
阿尔法-0.5
贝塔0.94
夏普比率-0.53
胜率0.57
盈亏比0.67
收益波动率40.26%
信息比率-0.09
最大回撤43.26%

bigcharts-data-start/{"__type":"tabs","__id":"bigchart-25622311fc9644779ab504b6b04402aa"}/bigcharts-data-end

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-a39867bc6c084df286802e710e29661a"}/bigcharts-data-end

收益率-22.78%
年化收益率-23.43%
基准收益率36.07%
阿尔法-0.5
贝塔0.94
夏普比率-0.53
胜率0.57
盈亏比0.67
收益波动率40.26%
信息比率-0.09
最大回撤43.26%

bigcharts-data-start/{"__type":"tabs","__id":"bigchart-07fa9608e88d41c28ae30a49d00fb9df"}/bigcharts-data-end