# Copyright 2019 IBM Corporation
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

import sklearn.svm
import lale.docstrings
import lale.operators

class SVCImpl():
    def __init__(self, **hyperparams):
        self._hyperparams = hyperparams
        self._wrapped_model = sklearn.svm.SVC(**self._hyperparams)

    def fit(self, X, y=None, sample_weight=None):
        self._wrapped_model.fit(X, y, sample_weight)
        return self

    def predict(self, X):
        return self._wrapped_model.predict(X)

    def predict_proba(self, X):
        return self._wrapped_model.predict_proba(X)

    def decision_function(self, X):
        return self._wrapped_model.decision_function(X)

_hyperparams_schema = {
    'allOf': [
    {   'type': 'object',
        'additionalProperties': False,
        'required': ['kernel', 'degree', 'gamma', 'shrinking', 'tol', 'cache_size', 'max_iter', 'decision_function_shape'],
        'relevantToOptimizer': ['kernel', 'degree', 'gamma', 'shrinking', 'probability', 'tol'],
        'properties': {
            'C': {
                'description': 'Penalty parameter C of the error term.',
                'type': 'number',
                'distribution': 'loguniform',
                'minimum': 0.0,
                'exclusiveMinimum': True,
                'default': 1.0,
                'minimumForOptimizer': 0.03125,
                'maximumForOptimizer': 32768},
            'kernel': {
                'anyOf': [
                {   'enum':['precomputed'], 'forOptimizer': False}, 
                {   'enum': ['linear', 'poly', 'rbf', 'sigmoid']},
                {   'laleType': 'callable', 'forOptimizer': False}],
                'default': 'rbf',
                'description':
                    'Specifies the kernel type to be used in the algorithm.'},
            'degree': {
                'type': 'integer',
                'minimum': 0,
                'minimumForOptimizer': 2,
                'maximumForOptimizer': 5,
                'default': 3,
                'description': "Degree of the polynomial kernel function ('poly')."},
            'gamma': {
                'anyOf': [
                {   'type': 'number',
                    'minimum': 0.0,
                    'exclusiveMinimum': True,
                    'minimumForOptimizer': 3.0517578125e-05,
                    'maximumForOptimizer': 8,
                    'distribution': 'loguniform'},
                {   'enum': ['auto', 'auto_deprecated', 'scale']}],
                'default': 'auto_deprecated', #going to change to 'scale' from sklearn 0.22.
                'description': "Kernel coefficient for 'rbf', 'poly', and 'sigmoid'."},
            'coef0': {
                'type': 'number',
                'default': 0.0,
                'description': 'Independent term in kernel function.'},
            'shrinking': {
                'type': 'boolean',
                'default': True,
                'description': 'Whether to use the shrinking heuristic.'},
            'probability': {
                'type': 'boolean',
                'default': False,
                'description': 'Whether to enable probability estimates.'},
            'tol': {
                'type': 'number',
                'distribution': 'loguniform',
                'minimum': 0.0,
                'exclusiveMinimum': True,
                'maximumForOptimizer': 0.01,
                'default': 0.0001,
                'description': 'Tolerance for stopping criteria.'},
            'cache_size': {
                'type': 'integer',
                'default': 200,
                'description': 'Specify the size of the kernel cache (in MB).'},
            'class_weight': {
                'anyOf': [
                    { 'description': 'By default, all classes have weight 1.',
                      'enum': [None]},
                    { 'description': 'Adjust weights by inverse frequency.',
                      'enum': ['balanced']},
                    { 'description': 'Dictionary mapping class labels to weights.',
                      'type': 'object',
                      'additionalProperties': {'type': 'number'},
                      'forOptimizer': False}],
                'default': None},
            'verbose': {
                'type': 'boolean',
                'default': False,
                'description': 'Enable verbose output.'},
            'max_iter': {
                'type': 'integer',
                'default': -1,
                'description': 'Hard limit on iterations within solver, or -1 for no limit.'},
            'decision_function_shape': {
                'enum': ['ovo', 'ovr'],
                'default': 'ovr',
                'description': "Whether to return a one-vs-rest ('ovr') decision function of shape (n_samples, n_classes) as all other classifiers, or the original one-vs-one (‘ovo’) decision function of libsvm which has shape (n_samples, n_classes * (n_classes - 1) / 2)."},
            'random_state': {
                'description':
                'Seed of pseudo-random number generator.',
                'anyOf': [
                {   'laleType': 'numpy.random.RandomState'},
                {   'description': 'RandomState used by np.random',
                    'enum': [None]},
                {   'description': 'Explicit seed.',
                    'type': 'integer'}],
                'default': None}}},
    {   'description': 'coef0 only significant in kernel ‘poly’ and ‘sigmoid’.',
        'anyOf': [
        {   'type': 'object',
            'properties': {'kernel': {'enum': ['poly', 'sigmoid']}}},
        {   'type': 'object',
            'properties': {'coef0': {'enum': [0.0]}}}]}]}

_input_fit_schema = {
    'type': 'object',
    'required': ['X', 'y'],
    'properties': {
        'X': {
            'type': 'array',
            'description': 'The outer array is over samples aka rows.',
            'items': {
                'type': 'array',
                'description': 'The inner array is over features aka columns.',
                'items': {
                    'type': 'number'}}},
        'y': {
            'description': 'The predicted classes.',
            'anyOf': [
            {   'type': 'array', 'items': {'type': 'number'}},
            {   'type': 'array', 'items': {'type': 'string'}},
            {   'type': 'array', 'items': {'type': 'boolean'}}]},
        'sample_weight': {
            'anyOf': [
            {   'type': 'array',
                'items': {'type': 'number'}},
            {   'enum': [None],
                'description': 'Samples are equally weighted.'}],
            'description': 'Sample weights.'}}}

_input_predict_schema = {
    'type': 'object',
    'properties': {
        'X': {
            'type': 'array',
            'description': 'The outer array is over samples aka rows.',
            'items': {
                'type': 'array',
                'description': 'The inner array is over features aka columns.',
                'items': {
                    'type': 'number'}}}}}

_output_predict_schema = {
    'description': 'The predicted classes.',
    'anyOf': [
    {   'type': 'array', 'items': {'type': 'number'}},
    {   'type': 'array', 'items': {'type': 'string'}},
    {   'type': 'array', 'items': {'type': 'boolean'}}]}

_input_predict_proba_schema = {
    'type': 'object',
    'properties': {
        'X': {
            'type': 'array',
            'description': 'The outer array is over samples aka rows.',
            'items': {
                'type': 'array',
                'description': 'The inner array is over features aka columns.',
                'items': {
                    'type': 'number'}}}}}

_output_predict_proba_schema = {
    'type': 'array',
    'description': 'The outer array is over samples aka rows.',
    'items': {
        'type': 'array',
        'description': 'The inner array has items corresponding to each class.',
        'items': {
            'type': 'number'}}}

_input_decision_function_schema = {
    'type': 'object',
    'required': ['X'],
    'properties': {
        'X': {
            'type': 'array',
            'description': 'The outer array is over samples aka rows.',
            'items': {
                'type': 'array',
                'description': 'The inner array is over features aka columns.',
                'items': {
                    'type': 'number'}}}}}

_output_decision_function_schema = {
    'description': 'Confidence scores for samples for each class in the model.',
    'anyOf': [
    {   'description': 'In the multi-way case, score per (sample, class) combination.',
        'type': 'array',
        'items': {'type': 'array', 'items': {'type': 'number'}}},
    {   'description': 'In the binary case, score for `self._classes[1]`.',
        'type': 'array',
        'items': {'type': 'number'}}]}

_combined_schemas = {
    '$schema': 'http://json-schema.org/draft-04/schema#',
    'description': """`Support Vector Classification`_ from scikit-learn.

.. _`Support Vector Classification`: https://scikit-learn.org/0.20/modules/generated/sklearn.svm.SVC.html#sklearn-svm-svc
""",
    'documentation_url': 'https://lale.readthedocs.io/en/latest/modules/lale.lib.sklearn.svc.html',
    'import_from': 'sklearn.svm',
    'type': 'object',
    'tags': {
        'pre': [],
        'op': ['estimator', 'classifier'],
        'post': []},
    'properties': {
        'hyperparams': _hyperparams_schema,
        'input_fit': _input_fit_schema,
        'input_predict': _input_predict_schema,
        'output_predict': _output_predict_schema,
        'input_predict_proba': _input_predict_proba_schema,
        'output_predict_proba': _output_predict_proba_schema,
        'input_decision_function': _input_decision_function_schema,
        'output_decision_function': _output_decision_function_schema}}

SVC : lale.operators.IndividualOp
SVC = lale.operators.make_operator(SVCImpl, _combined_schemas)

if sklearn.__version__ >= '0.22':
    # old: https://scikit-learn.org/0.20/modules/generated/sklearn.svm.SVC.html
    # new: https://scikit-learn.org/0.23/modules/generated/sklearn.svm.SVC.html
    from lale.schemas import AnyOf, Bool, Enum, Float
    SVC = SVC.customize_schema(
        gamma=AnyOf(
            types=[
                Enum(['scale', 'auto']),
                Float(
                    min=0.0,
                    exclusiveMin=True,
                    minForOptimizer=3.0517578125e-05,
                    maxForOptimizer=8,
                    distribution='loguniform')],
            desc="Kernel coefficient for 'rbf', 'poly' and 'sigmoid'.",
            default='scale'),
        break_ties=Bool(
            desc="If true, decision_function_shape='ovr', and number of classes > 2, predict will break ties according to the confidence values of decision_function; otherwise the first class among the tied classes is returned.",
            default=False))

lale.docstrings.set_docstrings(SVCImpl, SVC._schemas)