复制链接

克隆策略

In [ ]:

# 本代码由可视化策略环境自动生成 2023年5月6日 22:47
# 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 m6_run_bigquant_run(input_1, input_2, input_3):
    df = input_1.read()
    
    # 数据格式统一：float
    col_names = list(df.columns)
    col_names.remove('instrument')
    col_names.remove('date')
    df.loc[:,col_names] = df.loc[:,col_names].astype(float)
    
    # 处理数据：inf_
    import numpy as np
    df.loc[:,col_names] = df.loc[:,col_names].where(~np.isinf(df.loc[:,col_names].values),other=np.nan)
    data_1 = DataSource.write_df(df)
    return Outputs(data_1=data_1, 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 m6_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 m8_run_bigquant_run(input_1, input_2, input_3):
    df = input_1.read()
    
    # 数据格式统一：float
    col_names = list(df.columns)
    col_names.remove('instrument')
    col_names.remove('date')
    df.loc[:,col_names] = df.loc[:,col_names].astype(float)
    
    # 处理数据：inf_
    import numpy as np
    df.loc[:,col_names] = df.loc[:,col_names].where(~np.isinf(df.loc[:,col_names].values),other=np.nan)
    data_1 = DataSource.write_df(df)
    return Outputs(data_1=data_1, 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 m8_post_run_bigquant_run(outputs):
    return outputs

# 回测引擎：初始化函数，只执行一次
def m12_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.options['hold_days'] = 1

# 回测引擎：每日数据处理函数，每天执行一次
def m12_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')]
    
    ranker_prediction = ranker_prediction.sort_values('prediction',ascending=False)
    
    # 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.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.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 m12_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 m12_before_trading_start_bigquant_run(context, data):
    pass


m1 = M.instruments.v2(
    start_date='2010-01-01',
    end_date='2021-05-01',
    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, -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
)

m4 = M.standardlize.v9(
    input_1=m2.data,
    standard_func='ZScoreNorm',
    columns_input='label'
)

m3 = M.input_features.v1(
    features="""ta_willr_14_0
rank_return_10/rank_return_30
avg_turn_15/turn_0
ta_rsi_14_0
market_cap_0
turn_5/avg_turn_10
swing_volatility_10_0/swing_volatility_60_0
rank_amount_10/rank_amount_30
rank_amount_5/rank_amount_20
rank_return_10
rank_return_20
rank_return_5/rank_return_20
rank_avg_turn_5
rank_volatility_5_0
ma_2_slope_0 = np.gradient(ta_ma(close_0,2))
"""
)

m15 = M.general_feature_extractor.v7(
    instruments=m1.data,
    features=m3.data,
    start_date='',
    end_date='',
    before_start_days=250
)

m21 = M.chinaa_stock_filter.v1(
    input_data=m15.data,
    index_constituent_cond=['全部'],
    board_cond=['上证主板', '深证主板', '创业板'],
    industry_cond=['全部'],
    st_cond=['正常'],
    delist_cond=['非退市'],
    output_left_data=False
)

m16 = M.derived_feature_extractor.v3(
    input_data=m21.data,
    features=m3.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False
)

m7 = M.join.v3(
    data1=m4.data,
    data2=m16.data,
    on='date,instrument',
    how='inner',
    sort=True
)

m6 = M.cached.v3(
    input_1=m7.data,
    run=m6_run_bigquant_run,
    post_run=m6_post_run_bigquant_run,
    input_ports='',
    params='{}',
    output_ports=''
)

m13 = M.dropnan.v1(
    input_data=m6.data_1
)

m11 = M.features_short_user.v2(
    input_1=m3.data
)

m9 = M.instruments.v2(
    start_date=T.live_run_param('trading_date', '2021-05-02'),
    end_date=T.live_run_param('trading_date', '2023-05-05'),
    market='CN_STOCK_A',
    instrument_list='',
    max_count=0
)

m17 = M.general_feature_extractor.v7(
    instruments=m9.data,
    features=m3.data,
    start_date='',
    end_date='',
    before_start_days=250
)

m22 = M.chinaa_stock_filter.v1(
    input_data=m17.data,
    index_constituent_cond=['全部'],
    board_cond=['上证主板', '深证主板', '创业板'],
    industry_cond=['全部'],
    st_cond=['正常'],
    delist_cond=['非退市'],
    output_left_data=False
)

m18 = M.derived_feature_extractor.v3(
    input_data=m22.data,
    features=m3.data,
    date_col='date',
    instrument_col='instrument',
    drop_na=False,
    remove_extra_columns=False
)

m8 = M.cached.v3(
    input_1=m18.data,
    run=m8_run_bigquant_run,
    post_run=m8_post_run_bigquant_run,
    input_ports='',
    params='{}',
    output_ports=''
)

m14 = M.dropnan.v1(
    input_data=m8.data_1
)

m10 = M.xgboost.v1(
    training_ds=m13.data,
    features=m11.data_1,
    predict_ds=m14.data,
    num_boost_round=5,
    objective='排序(ndcg)',
    booster='gbtree',
    max_depth=5,
    key_cols='date,instrument',
    group_col='date',
    nthread=4,
    n_gpus=-1,
    other_train_parameters={
    'tree_method':'hist',
    'learning_rat':'0.01',
    'subsample':'0.7',
}
)

m12 = M.trade.v4(
    instruments=m9.data,
    options_data=m10.predictions,
    start_date='',
    end_date='',
    initialize=m12_initialize_bigquant_run,
    handle_data=m12_handle_data_bigquant_run,
    prepare=m12_prepare_bigquant_run,
    before_trading_start=m12_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='000300.HIX'
)

[2023-05-06 22:46:37.069319] INFO: moduleinvoker: instruments.v2 开始运行..

[2023-05-06 22:46:37.080641] INFO: moduleinvoker: 命中缓存

[2023-05-06 22:46:37.082606] INFO: moduleinvoker: instruments.v2 运行完成[0.01332s].

[2023-05-06 22:46:37.101214] INFO: moduleinvoker: advanced_auto_labeler.v2 开始运行..

[2023-05-06 22:46:37.111391] INFO: moduleinvoker: 命中缓存

[2023-05-06 22:46:37.113362] INFO: moduleinvoker: advanced_auto_labeler.v2 运行完成[0.01215s].

[2023-05-06 22:46:37.119310] INFO: moduleinvoker: standardlize.v9 开始运行..

[2023-05-06 22:46:37.124985] INFO: moduleinvoker: 命中缓存

[2023-05-06 22:46:37.126813] INFO: moduleinvoker: standardlize.v9 运行完成[0.007509s].

[2023-05-06 22:46:37.131060] INFO: moduleinvoker: input_features.v1 开始运行..

[2023-05-06 22:46:37.139387] INFO: moduleinvoker: 命中缓存

[2023-05-06 22:46:37.141330] INFO: moduleinvoker: input_features.v1 运行完成[0.010282s].

[2023-05-06 22:46:37.154917] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2023-05-06 22:46:37.162296] INFO: moduleinvoker: 命中缓存

[2023-05-06 22:46:37.164755] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.009865s].

[2023-05-06 22:46:37.175347] INFO: moduleinvoker: chinaa_stock_filter.v1 开始运行..

[2023-05-06 22:46:37.183833] INFO: moduleinvoker: 命中缓存

[2023-05-06 22:46:37.186377] INFO: moduleinvoker: chinaa_stock_filter.v1 运行完成[0.011029s].

[2023-05-06 22:46:37.195692] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2023-05-06 22:46:37.205371] INFO: moduleinvoker: 命中缓存

[2023-05-06 22:46:37.207345] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.011669s].

[2023-05-06 22:46:37.223297] INFO: moduleinvoker: join.v3 开始运行..

[2023-05-06 22:46:37.233045] INFO: moduleinvoker: 命中缓存

[2023-05-06 22:46:37.235816] INFO: moduleinvoker: join.v3 运行完成[0.012525s].

[2023-05-06 22:46:37.266462] INFO: moduleinvoker: cached.v3 开始运行..

[2023-05-06 22:46:37.272916] INFO: moduleinvoker: 命中缓存

[2023-05-06 22:46:37.275504] INFO: moduleinvoker: cached.v3 运行完成[0.009058s].

[2023-05-06 22:46:37.285013] INFO: moduleinvoker: dropnan.v1 开始运行..

[2023-05-06 22:46:37.292245] INFO: moduleinvoker: 命中缓存

[2023-05-06 22:46:37.294692] INFO: moduleinvoker: dropnan.v1 运行完成[0.00969s].

[2023-05-06 22:46:37.306875] INFO: moduleinvoker: features_short_user.v2 开始运行..

[2023-05-06 22:46:37.313035] INFO: moduleinvoker: 命中缓存

[2023-05-06 22:46:37.315077] INFO: moduleinvoker: features_short_user.v2 运行完成[0.008239s].

[2023-05-06 22:46:37.320867] INFO: moduleinvoker: instruments.v2 开始运行..

[2023-05-06 22:46:37.328349] INFO: moduleinvoker: 命中缓存

[2023-05-06 22:46:37.331040] INFO: moduleinvoker: instruments.v2 运行完成[0.010168s].

[2023-05-06 22:46:37.344705] INFO: moduleinvoker: general_feature_extractor.v7 开始运行..

[2023-05-06 22:46:37.351619] INFO: moduleinvoker: 命中缓存

[2023-05-06 22:46:37.354211] INFO: moduleinvoker: general_feature_extractor.v7 运行完成[0.009482s].

[2023-05-06 22:46:37.363402] INFO: moduleinvoker: chinaa_stock_filter.v1 开始运行..

[2023-05-06 22:46:37.370508] INFO: moduleinvoker: 命中缓存

[2023-05-06 22:46:37.372376] INFO: moduleinvoker: chinaa_stock_filter.v1 运行完成[0.00898s].

[2023-05-06 22:46:37.379784] INFO: moduleinvoker: derived_feature_extractor.v3 开始运行..

[2023-05-06 22:46:37.389008] INFO: moduleinvoker: 命中缓存

[2023-05-06 22:46:37.391915] INFO: moduleinvoker: derived_feature_extractor.v3 运行完成[0.012135s].

[2023-05-06 22:46:37.404694] INFO: moduleinvoker: cached.v3 开始运行..

[2023-05-06 22:46:37.411357] INFO: moduleinvoker: 命中缓存

[2023-05-06 22:46:37.413811] INFO: moduleinvoker: cached.v3 运行完成[0.009122s].

[2023-05-06 22:46:37.424135] INFO: moduleinvoker: dropnan.v1 开始运行..

[2023-05-06 22:46:37.431594] INFO: moduleinvoker: 命中缓存

[2023-05-06 22:46:37.433488] INFO: moduleinvoker: dropnan.v1 运行完成[0.009362s].

[2023-05-06 22:46:37.442554] INFO: moduleinvoker: xgboost.v1 开始运行..

[2023-05-06 22:47:44.677684] INFO: moduleinvoker: xgboost.v1 运行完成[67.235122s].

[2023-05-06 22:47:44.732427] INFO: moduleinvoker: backtest.v8 开始运行..

[2023-05-06 22:47:44.746392] INFO: backtest: biglearning backtest:V8.6.3

[2023-05-06 22:47:44.748524] INFO: backtest: product_type:stock by specified

[2023-05-06 22:47:44.817980] INFO: moduleinvoker: cached.v2 开始运行..

[2023-05-06 22:47:44.825263] INFO: moduleinvoker: 命中缓存

[2023-05-06 22:47:44.827600] INFO: moduleinvoker: cached.v2 运行完成[0.009661s].