add receipt code

@@ -72,4 +72,8 @@ image_process.py : cropping을 통해 정사각형 형태를 만들고 32*32 크기로 resize,
 
 
 15. params.pkl 추가
-제작한 데이터셋을 트레이닝하고 나온 가중치를 저장한 params.pkl 파일 추가
\ No newline at end of file
+제작한 데이터셋을 트레이닝하고 나온 가중치를 저장한 params.pkl 파일 추가
+
+
+16. simple_convnet_rasberryPi 코드 추가
+인식한 결과에 따라 계산서가 나오도록 코드를 추가
\ No newline at end of file

@@ -18,7 +18,7 @@ struct conv_param {
 };
 
 class SimpleConvNet {
-private:	
+private:
 	std::vector< Layer* > layers;
 
 	std::vector<Mat> W[7];		// weights
@@ -26,9 +26,10 @@ private:
 public:
 	SimpleConvNet() {}
 	~SimpleConvNet() {}
-	SimpleConvNet(input_dim id, conv_param cp, int hidden_size=512, int output_size=10, bool pretrained=true) {
-		if (pretrained) 
-			load_trained("params.txt"); 
+	SimpleConvNet(input_dim id, conv_param cp, int hidden_size = 512, int output_size = 10, bool pretrained = true) {
+		
+		if (pretrained)
+			load_trained("params_noLNORM.txt");
 
 		layers.push_back(new Convolution(W[0], 1, 1));
 		layers.push_back(new LightNormalization());
@@ -47,14 +48,14 @@ public:
 		layers.push_back(new Convolution(W[3], 1, 0));
 		layers.push_back(new LightNormalization());
 		layers.push_back(new Relu());
-		
+
 		layers.push_back(new DW_Convolution(W[4], 1, 1));
 		layers.push_back(new LightNormalization());
 		layers.push_back(new Relu());
 		layers.push_back(new Pooling(2, 2, 2));
 
 		layers.push_back(new Affine(W[5]));
-		layers.push_back(new LightNormalization());
+		//layers.push_back(new LightNormalization());
 		layers.push_back(new Relu());
 
 		layers.push_back(new Affine(W[6]));
@@ -62,12 +63,15 @@ public:
 
 	std::vector< Mat > predict(std::vector<Mat>& x) {
 		for (int i = 0; i < layers.size(); i++) {
+			//printf("%d Layer : (%d, %d, %d, %d)\n",i, x.size(), x[0].dim, x[0].row, x[0].col);
 			x = layers[i]->forward(x);
 		}
+		//printf("Layer : (%d %d %d %d)\n", x.size(), x[0].dim, x[0].row, x[0].col);
 		return x;
 	}
 
-	double accuracy(std::vector< std::vector< unsigned char > > x, std::vector< int > ans, int batch_size=100) {
+	double accuracy(std::vector< std::vector< unsigned char > > x, std::vector< int > ans, int batch_size = 100) {
+		// ...
 		return 1.0;
 	}
 
@@ -87,21 +91,21 @@ public:
 		return pred;
 	}
 
-	void load_trained(const char* filename="params.txt") {
+	void load_trained(const char* filename = "params.txt") {
 		FILE *f = fopen(filename, "r");
 		if (f == NULL) {
 			printf("File not found\n");
 			exit(1);
-		}			
+		}
 		char line[10] = { 0 };
 		int keynum;
-		while (fscanf(f, "%s", line)==1) {
+		while (fscanf(f, "%s", line) == 1) {
 			char s[4][10] = { 0 };
 			keynum = line[1] - '0' - 1;
 
 			// get shape
-			fscanf(f, "%s", s[0]); 
-			fscanf(f, "%s", s[1]); 
+			fscanf(f, "%s", s[0]); // "(num
+			fscanf(f, "%s", s[1]); // num
 			if (s[1][strlen(s[1]) - 1] != '\"') {
 				fscanf(f, "%s", s[2]);
 				fscanf(f, "%s", s[3]);
@@ -109,7 +113,7 @@ public:
 
 			// nw = number of weights : shape[0]
 			// size = input size of W[key]
-			int size = 1, nw=0;
+			int size = 1, nw = 0;
 			for (int i = 0; i < 4; i++) {
 				int val = 0;
 				for (int j = 0; j < strlen(s[i]); j++) {
@@ -130,9 +134,9 @@ public:
 			for (int i = 0; i < size; i++) {
 				fscanf(f, "%lf", &mm[i%fsize]);
 				if (i%fsize == fsize - 1) {
-					if(shape[keynum].size() == 2)
+					if (shape[keynum].size() == 2)
 						W[keynum].push_back(Mat(1, 1, shape[keynum][1], std::vector<double>(mm, mm + fsize)));
-					else if(shape[keynum].size() == 4)
+					else if (shape[keynum].size() == 4)
 						W[keynum].push_back(Mat(shape[keynum][1], shape[keynum][2],
 							shape[keynum][3], std::vector<double>(mm, mm + fsize)));
 				}

 #include"Layers.hpp"
 #include"SimpleConvNet.hpp"
+#include"fstream"
 using namespace std;
 int main() {
 
@@ -7,7 +8,7 @@ int main() {
 	conv_param cp = { 32,32,64, 3,1,1 };
 	SimpleConvNet SCN(id, cp);
 
-	freopen("input.txt", "r", stdin);	
+	freopen("input.txt", "r", stdin);
 	vector<Mat> X;
 	int nx = 1, dim = 3, row = 32, col = 32;
 	double tmp;
@@ -43,7 +44,93 @@ int main() {
 	auto pred = SCN.argmax(x);
 
 	int num = 0, pd;
+
+
+
+	char receipt[100];
+	int total = 0;
 	
-	printf("predict : %d ", pred[0]);
+	ifstream fin;
+	ofstream fout;
+	fin.open("receipt.txt");
+	char c;
+	int cnt=0;
+
+	while(fin.get(c)){
+		receipt[cnt] = c;
+		cnt++;
+	}
+	fin.close();
+
+	fout.open("receipt.txt");	
+	for(int i=0; i<cnt; i++){
+		fout << receipt[i];
+	}
+
+	switch(pred[0]){
+		case 0:
+			cout << "can\n";
+			fout << 0;
+			break;
+		case 1:
+			cout << "ramen\n";
+			fout << 1;
+			break;
+ 		case 2:
+			cout << "cigarette\n";
+			fout << 2;
+			break;
+		case 3:
+			cout << "instant rice\n";
+			fout << 3;
+			break;
+		case 4:
+			cout << "wet tissue\n";
+			fout << 4;
+			break;
+		default:
+			cout << "try again\n";
+			break;
+	}
+
+	fout.close();
+
+	fin.open("receipt.txt");
+	cnt=0;
+
+	while(fin.get(c)){
+		receipt[cnt] = c;
+		cnt++;
+	}
+	fin.close();
+	cout << "\n\n===================receipt===================\n\n\n";
+	for(int i = 0; i<cnt; i++){
+		switch(receipt[i]){
+		case '0':
+			cout << "can---------------------------------1,200Won\n";
+			total += 1200;
+			break;
+		case '1':
+			cout << "ramen-------------------------------1,000Won\n";
+			total += 1000;			
+			break;
+ 		case '2':
+			cout << "cigarette---------------------------4,500Won\n";
+			total += 4500;
+			break;
+		case '3':
+			cout << "instant rice------------------------1,500Won\n";
+			total += 1500;
+			break;
+		case '4':
+			cout << "wet tissue--------------------------2,000Won\n";
+			total += 2000;
+			break;
+		default:
+			break;
+		}
+	}
+	cout << "\nTotal-------------------------------" << total << "Won\n";
+	cout << "\n\n=============================================\n\n\n";
 	return 0;
-}
\ No newline at end of file
+}

@@ -2,9 +2,14 @@ from numpy import array, savetxt
 from PIL import Image
 import sys
 import picamera
+import time
 
 camera = picamera.PiCamera()
 
+camera.start_preview()
+time.sleep(3)
+camera.stop_preview()
+
 filename = 'image.jpg'
 camera.capture(filename)
 

+0201
+112
\ No newline at end of file