--- license: mit library_name: sklearn tags: - sklearn - skops - tabular-classification model_format: pickle model_file: LightGBM_without_hospital_number_01.pkl widget: structuredData: abdomen: - distend_small - distend_small - distend_large abdominal_distention: - none - none - moderate abdomo_appearance: - serosanguious - cloudy - serosanguious abdomo_protein: - 4.1 - 4.3 - 2.0 age: - adult - adult - adult capillary_refill_time: - less_3_sec - less_3_sec - more_3_sec cp_data: - 'yes' - 'yes' - 'no' lesion_1: - 7209 - 2112 - 5400 lesion_2: - 0 - 0 - 0 lesion_3: - 0 - 0 - 0 mucous_membrane: - bright_pink - bright_pink - dark_cyanotic nasogastric_reflux: - none - none - more_1_liter nasogastric_reflux_ph: - 7.0 - 3.5 - 2.0 nasogastric_tube: - slight - none - significant packed_cell_volume: - 37.0 - 44.0 - 65.0 pain: - depressed - mild_pain - extreme_pain peripheral_pulse: - normal - normal - reduced peristalsis: - hypermotile - hypomotile - absent pulse: - 84.0 - 66.0 - 72.0 rectal_exam_feces: - absent - decreased - absent rectal_temp: - 39.0 - 38.5 - 37.3 respiratory_rate: - 24.0 - 21.0 - 30.0 surgery: - 'yes' - 'yes' - 'yes' surgical_lesion: - 'yes' - 'yes' - 'yes' temp_of_extremities: - cool - normal - cool total_protein: - 6.5 - 7.6 - 13.0 --- # Model description This is a `LightGBM` model trained on horse health outcome data from Kaggle. ## Intended uses & limitations This model is not ready to be used in production. ## Training Procedure [More Information Needed] ### Hyperparameters
Click to expand | Hyperparameter | Value | |--------------------------------------------------|-----------------------------------------------------------------------------------------------------------------------------| | memory | | | steps | [('preprocessor', ColumnTransformer(remainder='passthrough',
transformers=[('num',
Pipeline(steps=[('imputer',
SimpleImputer(strategy='median')),
('scaler', StandardScaler())]),
['rectal_temp', 'pulse', 'respiratory_rate',
'nasogastric_reflux_ph', 'packed_cell_volume',
'total_protein', 'abdomo_protein', 'lesion_1',
'lesion_2', 'lesion_3']),
('cat',
Pipeline(steps=[('imputer',
SimpleI...='missing',
strategy='constant')),
('onehot',
OneHotEncoder(handle_unknown='ignore'))]),
['surgery', 'age', 'temp_of_extremities',
'peripheral_pulse', 'mucous_membrane',
'capillary_refill_time', 'pain',
'peristalsis', 'abdominal_distention',
'nasogastric_tube', 'nasogastric_reflux',
'rectal_exam_feces', 'abdomen',
'abdomo_appearance', 'surgical_lesion',
'cp_data'])])), ('classifier', LGBMClassifier(max_depth=3))] | | verbose | False | | preprocessor | ColumnTransformer(remainder='passthrough',
transformers=[('num',
Pipeline(steps=[('imputer',
SimpleImputer(strategy='median')),
('scaler', StandardScaler())]),
['rectal_temp', 'pulse', 'respiratory_rate',
'nasogastric_reflux_ph', 'packed_cell_volume',
'total_protein', 'abdomo_protein', 'lesion_1',
'lesion_2', 'lesion_3']),
('cat',
Pipeline(steps=[('imputer',
SimpleI...='missing',
strategy='constant')),
('onehot',
OneHotEncoder(handle_unknown='ignore'))]),
['surgery', 'age', 'temp_of_extremities',
'peripheral_pulse', 'mucous_membrane',
'capillary_refill_time', 'pain',
'peristalsis', 'abdominal_distention',
'nasogastric_tube', 'nasogastric_reflux',
'rectal_exam_feces', 'abdomen',
'abdomo_appearance', 'surgical_lesion',
'cp_data'])]) | | classifier | LGBMClassifier(max_depth=3) | | preprocessor__n_jobs | | | preprocessor__remainder | passthrough | | preprocessor__sparse_threshold | 0.3 | | preprocessor__transformer_weights | | | preprocessor__transformers | [('num', Pipeline(steps=[('imputer', SimpleImputer(strategy='median')),
('scaler', StandardScaler())]), ['rectal_temp', 'pulse', 'respiratory_rate', 'nasogastric_reflux_ph', 'packed_cell_volume', 'total_protein', 'abdomo_protein', 'lesion_1', 'lesion_2', 'lesion_3']), ('cat', Pipeline(steps=[('imputer',
SimpleImputer(fill_value='missing', strategy='constant')),
('onehot', OneHotEncoder(handle_unknown='ignore'))]), ['surgery', 'age', 'temp_of_extremities', 'peripheral_pulse', 'mucous_membrane', 'capillary_refill_time', 'pain', 'peristalsis', 'abdominal_distention', 'nasogastric_tube', 'nasogastric_reflux', 'rectal_exam_feces', 'abdomen', 'abdomo_appearance', 'surgical_lesion', 'cp_data'])] | | preprocessor__verbose | False | | preprocessor__verbose_feature_names_out | True | | preprocessor__num | Pipeline(steps=[('imputer', SimpleImputer(strategy='median')),
('scaler', StandardScaler())]) | | preprocessor__cat | Pipeline(steps=[('imputer',
SimpleImputer(fill_value='missing', strategy='constant')),
('onehot', OneHotEncoder(handle_unknown='ignore'))]) | | preprocessor__num__memory | | | preprocessor__num__steps | [('imputer', SimpleImputer(strategy='median')), ('scaler', StandardScaler())] | | preprocessor__num__verbose | False | | preprocessor__num__imputer | SimpleImputer(strategy='median') | | preprocessor__num__scaler | StandardScaler() | | preprocessor__num__imputer__add_indicator | False | | preprocessor__num__imputer__copy | True | | preprocessor__num__imputer__fill_value | | | preprocessor__num__imputer__keep_empty_features | False | | preprocessor__num__imputer__missing_values | nan | | preprocessor__num__imputer__strategy | median | | preprocessor__num__scaler__copy | True | | preprocessor__num__scaler__with_mean | True | | preprocessor__num__scaler__with_std | True | | preprocessor__cat__memory | | | preprocessor__cat__steps | [('imputer', SimpleImputer(fill_value='missing', strategy='constant')), ('onehot', OneHotEncoder(handle_unknown='ignore'))] | | preprocessor__cat__verbose | False | | preprocessor__cat__imputer | SimpleImputer(fill_value='missing', strategy='constant') | | preprocessor__cat__onehot | OneHotEncoder(handle_unknown='ignore') | | preprocessor__cat__imputer__add_indicator | False | | preprocessor__cat__imputer__copy | True | | preprocessor__cat__imputer__fill_value | missing | | preprocessor__cat__imputer__keep_empty_features | False | | preprocessor__cat__imputer__missing_values | nan | | preprocessor__cat__imputer__strategy | constant | | preprocessor__cat__onehot__categories | auto | | preprocessor__cat__onehot__drop | | | preprocessor__cat__onehot__dtype | | | preprocessor__cat__onehot__feature_name_combiner | concat | | preprocessor__cat__onehot__handle_unknown | ignore | | preprocessor__cat__onehot__max_categories | | | preprocessor__cat__onehot__min_frequency | | | preprocessor__cat__onehot__sparse | deprecated | | preprocessor__cat__onehot__sparse_output | True | | classifier__boosting_type | gbdt | | classifier__class_weight | | | classifier__colsample_bytree | 1.0 | | classifier__importance_type | split | | classifier__learning_rate | 0.1 | | classifier__max_depth | 3 | | classifier__min_child_samples | 20 | | classifier__min_child_weight | 0.001 | | classifier__min_split_gain | 0.0 | | classifier__n_estimators | 100 | | classifier__n_jobs | | | classifier__num_leaves | 31 | | classifier__objective | | | classifier__random_state | | | classifier__reg_alpha | 0.0 | | classifier__reg_lambda | 0.0 | | classifier__subsample | 1.0 | | classifier__subsample_for_bin | 200000 | | classifier__subsample_freq | 0 |
### Model Plot 
Pipeline(steps=[('preprocessor',ColumnTransformer(remainder='passthrough',transformers=[('num',Pipeline(steps=[('imputer',SimpleImputer(strategy='median')),('scaler',StandardScaler())]),['rectal_temp', 'pulse','respiratory_rate','nasogastric_reflux_ph','packed_cell_volume','total_protein','abdomo_protein', 'lesion_1','lesion_2', 'lesion_3']),('cat',Pi...OneHotEncoder(handle_unknown='ignore'))]),['surgery', 'age','temp_of_extremities','peripheral_pulse','mucous_membrane','capillary_refill_time','pain', 'peristalsis','abdominal_distention','nasogastric_tube','nasogastric_reflux','rectal_exam_feces','abdomen','abdomo_appearance','surgical_lesion','cp_data'])])),('classifier', LGBMClassifier(max_depth=3))])
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
## Evaluation Results | Metric | Value | |----------|----------| | accuracy | 0.740891 | | f1 score | 0.740891 | ### Confusion Matrix ![Confusion Matrix](confusion_matrix.png) ## Permutation Importance ![Permutation Importance](feature_importance.png) # How to Get Started with the Model [More Information Needed] # Model Card Authors kmposkid # Model Card Contact You can contact the model card authors through following channels: [More Information Needed] # Citation Below you can find information related to citation. **BibTeX:** ``` [More Information Needed] ```