/*!
 * \file FeatureEvaluator.h
 * \date 2019/11/12
 *
 * \author Lin, Chi
 * Contact: lin.chi@hzleaper.com
 *
 *
 * \note
*/

#ifndef __FeatureEvaluator_h_
#define __FeatureEvaluator_h_

#include "CyclopsCommon.h"
#include "CyclopsModules.h"
#include "CyclopsEnums.h"
#include "CyclopsFeature.h"
#include "StdUtils.h"
#include "CVUtils.h"
#include "SampleDBManager.h"

/*! \brief Feature evaluator, for feature extraction and selection
* Should be used together with LLClassifier
*/
class FeatureEvaluator : public ICyclopsModuleInstance
{
public:
	FeatureEvaluator() : mDirtyBit(false) {}
	virtual ~FeatureEvaluator() {}

	//! Smart pointer to hold an instance of FeatureEvaluator
	typedef std::shared_ptr<FeatureEvaluator> Ptr;
	DECL_GET_INSTANCE(FeatureEvaluator::Ptr)

	/*! \fn serializeToMemory
	 * Serialize information of the feature evaluator into a in-memory string, see also deserializeFromMemory()
	 * @param str used to take the output serialization result
	 * @return true for succeed, false for fail
	 * \fn serializeToFile
	 * Serialize information of the feature evaluator into a text file, see also deserializeFromFile()
	 * @param filename file name (full path) where we will write the data
	 * @return true for succeed, false for fail
	 * \fn deserializeFromMemory
	 * Deserialize the feature evaluator from in-memory string, see also serializeToMemory()
	 * @param str in-memory string
	 * @return true for succeed, false for fail
	 * \fn deserializeFromFile
	 * Deserialize the feature evaluator from a text file, see also serializeToFile()
	 * @param filename file name (full path) where we will read the data
	 * @return true for succeed, false for fail
	*/
	DECL_SERIALIZE_FUNCS

	/*! Add a new feature of given type */
	virtual CyclopsFeature::Ptr add(FeatureType ftype);
	/*! Total count of features defined in this evaluator */
	int size() const { return mFeatures.size(); }
	/*! Get feature by index */
	CyclopsFeature::Ptr get(int index) {
		if (index < 0 || index >= size()) return nullptr;
		return mFeatures[index];
	}
	/*! Remove feature by index */
	virtual bool remove(int index);
	/*! Remove all features */
	virtual bool removeAll();
	/*! Query feature index by its name */
	virtual int indexByName(const std::string& name);

	/*! Total count of all enabled features */
	virtual int enabledSize() const;
	/*! Enable or disable all features */
	virtual void enableAll(bool val);

	/*! Check whether we need to fix some features of bad parameter configuration
	* @param sampleSize uniformed sample size
	* @param doFix true if we should also do the real fix
	* @return true if fix is needed
	*/
	virtual bool needFix(const Size& sampleSize, bool doFix = false);

	/*! Check whether we need to re-train some features if parameter changed
	* @param sampleSize uniformed sample size
	* @return true if re-train is needed
	*/
	virtual bool needTrain(const Size& sampleSize);

	/*! Train features with provided samples */
	virtual bool train(std::vector<SampleInstIterator>& samples, const Size& sampleSize);

	/** @overload Only train feature with given index */
	virtual bool train(int index, std::vector<SampleInstIterator>& samples, const Size& sampleSize);

	/*! Evaluate current parameter configuration with provided samples
	* @param index feature index
	* @param samples samples used for training
	* @param sampleSize uniformed sample size
	* @param score output evaluation score, higher is better
	* @param time output computation time of the feature
	* @return true if everything is fine
	*/
	virtual bool evaluate(int index, std::vector<SampleInstIterator>& samples, const Size& sampleSize,
		double& score, double& time);

	/*! Evaluate whether it is hard to distinguish the provided sample collections */
	virtual bool confuse(std::vector<SampleInstIterator>& samples,
		const Size& sampleSize, Mat& confuseMat);

	/*! Total length of feature vector contains all */
	virtual int featureVecSize(const Size& sampleSize) const;

	/*! Get range of features in feature vector */
	virtual vector<Range> featureVecRanges(const Size& sampleSize) const;

	/*! Compute feature vector for given image */
	virtual void compute(const SampleInstPtr& sample, Mat& result);

	/*! True if any feature's any parameter is changed */
	virtual bool isDirty() const;

	/*! Clear trained cache of all features */
	virtual void cleanAll();

private:
	virtual bool serialize(FileStorage& fs);
	virtual bool deserialize(const FileNode& fs);

	void silhouetteAnalysis(int classCount, int totalSampleCount,
		const std::vector<Mat>& featureMats, const Mat& labelMat, const Mat& weightMat,
		vector<double>& shMeans);

private:
	std::vector<CyclopsFeature::Ptr> mFeatures;
	bool mDirtyBit;
};

#endif // FeatureEvaluator_h_