[dask] TypeError: Unknown type of parameter:init_score, got:Series

[darshanbint]

Description

Reproducible example

Environment info

LightGBM version or commit hash:

Command(s) you used to install LightGBM

Additional Comments

[StrikerRUS]

Try to pass init_score as an argument not as part of params.
Please provide reproducible example, otherwise we cannot help you.

[darshanbint]

Error log:

Finding random open ports for workers
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-19-4463b2b3ec8a> in <module>
----> 1 model.fit(df2[feature_list], df2[target_list])

~/miniconda3/envs/darshanbint/lib/python3.9/site-packages/lightgbm/dask.py in fit(self, X, y, sample_weight, init_score, **kwargs)
    728     ) -> "DaskLGBMClassifier":
    729         """Docstring is inherited from the lightgbm.LGBMClassifier.fit."""
--> 730         return self._lgb_dask_fit(
    731             model_factory=LGBMClassifier,
    732             X=X,

~/miniconda3/envs/darshanbint/lib/python3.9/site-packages/lightgbm/dask.py in _lgb_dask_fit(self, model_factory, X, y, sample_weight, init_score, group, **kwargs)
    615         params.pop("client", None)
    616 
--> 617         model = _train(
    618             client=_get_dask_client(self.client),
    619             data=X,

~/miniconda3/envs/darshanbint/lib/python3.9/site-packages/lightgbm/dask.py in _train(client, data, label, params, model_factory, sample_weight, init_score, group, **kwargs)
    448     ]
    449 
--> 450     results = client.gather(futures_classifiers)
    451     results = [v for v in results if v]
    452     model = results[0]

~/miniconda3/envs/darshanbint/lib/python3.9/site-packages/distributed/client.py in gather(self, futures, errors, direct, asynchronous)
   1980             else:
   1981                 local_worker = None
-> 1982             return self.sync(
   1983                 self._gather,
   1984                 futures,

~/miniconda3/envs/darshanbint/lib/python3.9/site-packages/distributed/client.py in sync(self, func, asynchronous, callback_timeout, *args, **kwargs)
    851             return future
    852         else:
--> 853             return sync(
    854                 self.loop, func, *args, callback_timeout=callback_timeout, **kwargs
    855             )

~/miniconda3/envs/darshanbint/lib/python3.9/site-packages/distributed/utils.py in sync(loop, func, callback_timeout, *args, **kwargs)
    352     if error[0]:
    353         typ, exc, tb = error[0]
--> 354         raise exc.with_traceback(tb)
    355     else:
    356         return result[0]

~/miniconda3/envs/darshanbint/lib/python3.9/site-packages/distributed/utils.py in f()
    335             if callback_timeout is not None:
    336                 future = asyncio.wait_for(future, callback_timeout)
--> 337             result[0] = yield future
    338         except Exception as exc:
    339             error[0] = sys.exc_info()

~/miniconda3/envs/darshanbint/lib/python3.9/site-packages/tornado/gen.py in run(self)
    760 
    761                     try:
--> 762                         value = future.result()
    763                     except Exception:
    764                         exc_info = sys.exc_info()

~/miniconda3/envs/darshanbint/lib/python3.9/site-packages/distributed/client.py in _gather(self, futures, errors, direct, local_worker)
   1845                             exc = CancelledError(key)
   1846                         else:
-> 1847                             raise exception.with_traceback(traceback)
   1848                         raise exc
   1849                     if errors == "skip":

~/miniconda3/envs/darshanbint/lib/python3.9/site-packages/lightgbm/dask.py in _train_part()
    116             model.fit(data, label, sample_weight=weight, init_score=init_score, group=group, **kwargs)
    117         else:
--> 118             model.fit(data, label, sample_weight=weight, init_score=init_score, **kwargs)
    119 
    120     finally:

~/miniconda3/envs/darshanbint/lib/python3.9/site-packages/lightgbm/sklearn.py in fit()
    888                     valid_sets[i] = (valid_x, self._le.transform(valid_y))
    889 
--> 890         super().fit(X, _y, sample_weight=sample_weight, init_score=init_score, eval_set=valid_sets,
    891                     eval_names=eval_names, eval_sample_weight=eval_sample_weight,
    892                     eval_class_weight=eval_class_weight, eval_init_score=eval_init_score,

~/miniconda3/envs/darshanbint/lib/python3.9/site-packages/lightgbm/sklearn.py in fit()
    681             init_model = init_model.booster_
    682 
--> 683         self._Booster = train(params, train_set,
    684                               self.n_estimators, valid_sets=valid_sets, valid_names=eval_names,
    685                               early_stopping_rounds=early_stopping_rounds,

~/miniconda3/envs/darshanbint/lib/python3.9/site-packages/lightgbm/engine.py in train()
    226     # construct booster
    227     try:
--> 228         booster = Booster(params=params, train_set=train_set)
    229         if is_valid_contain_train:
    230             booster.set_train_data_name(train_data_name)

~/miniconda3/envs/darshanbint/lib/python3.9/site-packages/lightgbm/basic.py in __init__()
   2227                 )
   2228             # construct booster object
-> 2229             train_set.construct()
   2230             # copy the parameters from train_set
   2231             params.update(train_set.get_params())

~/miniconda3/envs/darshanbint/lib/python3.9/site-packages/lightgbm/basic.py in construct()
   1466             else:
   1467                 # create train
-> 1468                 self._lazy_init(self.data, label=self.label,
   1469                                 weight=self.weight, group=self.group,
   1470                                 init_score=self.init_score, predictor=self._predictor,

~/miniconda3/envs/darshanbint/lib/python3.9/site-packages/lightgbm/basic.py in _lazy_init()
   1247                 params['categorical_column'] = sorted(categorical_indices)
   1248 
-> 1249         params_str = param_dict_to_str(params)
   1250         self.params = params
   1251         # process for reference dataset

~/miniconda3/envs/darshanbint/lib/python3.9/site-packages/lightgbm/basic.py in param_dict_to_str()
    234             pairs.append(str(key) + '=' + str(val))
    235         elif val is not None:
--> 236             raise TypeError('Unknown type of parameter:%s, got:%s'
    237                             % (key, type(val).__name__))
    238     return ' '.join(pairs)

TypeError: Unknown type of parameter:init_score, got:Series

code

init_score = some_float_value * np.log(df[some_column] / (int(1) - df[some_column])) 
# type(df): <class 'dask.dataframe.core.DataFrame'>
# type(init_score): <class 'dask.dataframe.core.Series'> 

# failed on:
model =  lgb.DaskLGBMClassifier(**TRAINING_PARAMETERS, client=client, init_score=init_score)

model.fit(df[feature_list], df[target_list])

[jameslamb]

Thanks very much for including a bit more information. I've reformatted #4409 (comment) a bit for better readability. Please see https://docs.github.com/en/github/writing-on-github/getting-started-with-writing-and-formatting-on-github/basic-writing-and-formatting-syntax if you are new to GitHub and not familiar with GitHub-flavored markdown used in posts here.

---

I can see from your code that you are passing init_score to the constructor of lgb.DaskLGBMClassifier. This is not supported. Please try passing init_score into DaskLGBMClassifier.fit() instead.

Please consult the documentation, which shows where to pass these values:

• https://lightgbm.readthedocs.io/en/latest/pythonapi/lightgbm.DaskLGBMClassifier.html#lightgbm.DaskLGBMClassifier
• https://lightgbm.readthedocs.io/en/latest/pythonapi/lightgbm.DaskLGBMClassifier.html#lightgbm.DaskLGBMClassifier.fit

---

If that works for you, please comment here so the issue can be closed.

If the error persists, please provide a fully reproducible example. That should be code that can be run without modification which reproduces the error. If you are not experienced in this, please see the definition at https://stackoverflow.com/help/minimal-reproducible-example for some guidance.

Your reproducible example should also include answers to these questions which were asked in the issue template when you opened this issue:

• LightGBM version or commit hash:
• Command(s) you used to install LightGBM

[darshanbint]

Worked fine, thanks!

[github-actions]

This issue has been automatically locked since there has not been any recent activity since it was closed. To start a new related discussion, open a new issue at https://github.com/microsoft/LightGBM/issues including a reference to this.