[FAQ] Module 6: ValueError: not enough values to unpack (expected 3, got 1)

[dmitrykosintsev]

Course

machine-learning-zoomcamp

Question

When running parse_xg_output(output), I get the following error:

ValueError                                Traceback (most recent call last)
Cell In[67], line 1
----> 1 df_score = parse_xg_output(output)

Cell In[66], line 5, in parse_xg_output(output)
      2 results = []
      4 for line in output.stdout.strip().split('\n'):
----> 5     it_line, train_line, val_line = line.split('\t')
      7     it = int(it_line.strip('[]'))
      8     train = float(train_line.split(':')[1])

ValueError: not enough values to unpack (expected 3, got 1)

Answer

Check the following outputs:

print(repr(output.stdout))

and

print(repr(output.stderr))

If they are empty, this means XGBoost does not print training logs to stdout.
One solution is to use the training results returned by the Python API:

evals_result = {}

model = xgb.train(
    params=xgb_params,
    dtrain=dtrain,
    num_boost_round=200,
    evals=[(dtrain, 'train'), (dval, 'val')],
    evals_result=evals_result,
    verbose_eval=True
)

# Convert results to a DataFrame directly
df_results = pd.DataFrame({
    'num_iter': range(len(evals_result['train']['auc'])),
    'train_auc': evals_result['train']['auc'],
    'val_auc': evals_result['val']['auc']
})

df_results.head()

This will replace relevant code in parse_xg_output(). Also, the function now does not take any parameters.

Checklist

• I have searched existing FAQs and this question is not already answered
• The answer provides accurate, helpful information
• I have included any relevant code examples or links

[github-actions]

✅ FAQ NEW proposal created in PR #36

Please review and approve the changes.