The finetune function of model LightGBM has an ERROR

[zhaochaofeng]

Script: /qlib/contrib/model/gbdt.py

Code:
dtrain, _ = self._prepare_data(dataset, reweighter) # pylint: disable=W0632
if dtrain.empty:
raise ValueError("Empty data from dataset, please check your dataset config.")

Problem:
The _prepare_data function returns a list, but a list does not have the attribute empty.

Running Error:
if dtrain.empty:
AttributeError: 'tuple' object has no attribute 'empty'

[SunsetWolf]

Hi, @zhaochaofeng

Could you please share the steps or environment details so we can try to reproduce it?

[zhaochaofeng]

import qlib
from qlib.constant import REG_CN
from qlib.utils import init_instance_by_config
from qlib.workflow import R
from qlib.tests.config import CSI300_BENCH, CSI300_GBDT_TASK

if name == "main":
# use default data
provider_uri = "~/.qlib/qlib_data/cn_data" # target_dir
qlib.init(provider_uri=provider_uri, region=REG_CN)

model = init_instance_by_config(CSI300_GBDT_TASK["model"])
dataset = init_instance_by_config(CSI300_GBDT_TASK["dataset"])

port_analysis_config = {
    "executor": {
        "class": "SimulatorExecutor",
        "module_path": "qlib.backtest.executor",
        "kwargs": {
            "time_per_step": "day",
            "generate_portfolio_metrics": True,
        },
    },
    "strategy": {
        "class": "TopkDropoutStrategy",
        "module_path": "qlib.contrib.strategy.signal_strategy",
        "kwargs": {
            "signal": (model, dataset),
            "topk": 50,
            "n_drop": 5,
        },
    },
    "backtest": {
        "start_time": "2017-01-01",
        "end_time": "2020-08-01",
        "account": 100000000,
        "benchmark": CSI300_BENCH,
        "exchange_kwargs": {
            "freq": "day",
            "limit_threshold": 0.095,
            "deal_price": "close",
            "open_cost": 0.0005,
            "close_cost": 0.0015,
            "min_cost": 5,
        },
    },
}

# start exp to train init model
with R.start(experiment_name="init models"):
    model.fit(dataset)
    R.save_objects(init_model=model)
    rid = R.get_recorder().id

# Finetune model based on previous trained model
with R.start(experiment_name="finetune model"):
    recorder = R.get_recorder(recorder_id=rid, experiment_name="init models")
    model = recorder.load_object("init_model")
    model.finetune(dataset, num_boost_round=10)

[SunsetWolf]

Hi, @zhaochaofeng

This issue has been fixed in PR 2049, thanks for reporting it.

[zhaochaofeng]

Copy !