limited versions!

.idea/vcs.xml

@@ -2,6 +2,18 @@
 <project version="4">
   <component name="VcsDirectoryMappings">
     <mapping directory="" vcs="Git" />
+    <mapping directory="$PROJECT_DIR$/cmake-build-debug/_deps/freetype-src" vcs="Git" />
+    <mapping directory="$PROJECT_DIR$/cmake-build-debug/_deps/freetype-src/subprojects/dlg" vcs="Git" />
+    <mapping directory="$PROJECT_DIR$/cmake-build-debug/_deps/imgui-src" vcs="Git" />
+    <mapping directory="$PROJECT_DIR$/cmake-build-release/_deps/freetype-src" vcs="Git" />
+    <mapping directory="$PROJECT_DIR$/cmake-build-release/_deps/freetype-src/subprojects/dlg" vcs="Git" />
+    <mapping directory="$PROJECT_DIR$/cmake-build-release/_deps/imgui-src" vcs="Git" />
+    <mapping directory="$PROJECT_DIR$/cmake-build-relwithdebinfo-addrsan/_deps/freetype-src" vcs="Git" />
+    <mapping directory="$PROJECT_DIR$/cmake-build-relwithdebinfo-addrsan/_deps/freetype-src/subprojects/dlg" vcs="Git" />
+    <mapping directory="$PROJECT_DIR$/cmake-build-relwithdebinfo-addrsan/_deps/imgui-src" vcs="Git" />
+    <mapping directory="$PROJECT_DIR$/cmake-build-relwithdebinfo/_deps/freetype-src" vcs="Git" />
+    <mapping directory="$PROJECT_DIR$/cmake-build-relwithdebinfo/_deps/freetype-src/subprojects/dlg" vcs="Git" />
+    <mapping directory="$PROJECT_DIR$/cmake-build-relwithdebinfo/_deps/imgui-src" vcs="Git" />
     <mapping directory="$PROJECT_DIR$/lib/blt-with-graphics" vcs="Git" />
     <mapping directory="$PROJECT_DIR$/lib/blt-with-graphics/libraries/BLT" vcs="Git" />
     <mapping directory="$PROJECT_DIR$/lib/blt-with-graphics/libraries/BLT/libraries/parallel-hashmap" vcs="Git" />

CMakeLists.txt

@@ -1,5 +1,5 @@
 cmake_minimum_required(VERSION 3.25)
-project(COSC-4P80-Final-Project VERSION 0.0.23)
+project(COSC-4P80-Final-Project VERSION 0.0.24)
 
 option(ENABLE_ADDRSAN "Enable the address sanitizer" OFF)
 option(ENABLE_UBSAN "Enable the ub sanitizer" OFF)

graph.py

@@ -2,7 +2,7 @@ import matplotlib.pyplot as plt
 import pandas as pd
 import sys
 
-def plot_stacked_graph(title, output, csv_file1, csv_file2, position, position2):
+def plot_stacked_graph(title, output, csv_file1, csv_file2, name1, name2):
     # Read CSV files
     data1 = pd.read_csv(csv_file1, header=0)
     data2 = pd.read_csv(csv_file2, header=0)
@@ -18,27 +18,27 @@ def plot_stacked_graph(title, output, csv_file1, csv_file2, position, position2)
     # Create the plot
     fig, ax = plt.subplots()
 
-    ax.plot(x1, y1, label=f"{csv_file1}")
-    ax.plot(x2, y2, label=f"{csv_file2}")
+    ax.plot(x1, y1, label=f"{name1}")
+    ax.plot(x2, y2, label=f"{name2}")
     
-    if position < 2 ** 32:
-        ax.axvline(x=position, color='red', linestyle='--')
-        ax.text(position, ax.get_ylim()[1] * 0.95, f"Feed-forward average # of epochs", color='red', fontsize=10, ha='right', va='top', backgroundcolor='white')
+    # if position < 2 ** 32:
+    #     ax.axvline(x=position, color='red', linestyle='--')
+    #     ax.text(position, ax.get_ylim()[1] * 0.95, f"Feed-forward average # of epochs", color='red', fontsize=10, ha='right', va='top', backgroundcolor='white')
         
-    if position2 < 2 ** 32:
-        ax.axvline(x=position2, color='red', linestyle='--')
-        ax.text(position2, ax.get_ylim()[1] * 0.95, f"Deep learning average # of epochs", color='red', fontsize=10, ha='right', va='top', backgroundcolor='white')
+    # if position2 < 2 ** 32:
+    #     ax.axvline(x=position2, color='red', linestyle='--')
+    #     ax.text(position2, ax.get_ylim()[1] * 0.95, f"Deep learning average # of epochs", color='red', fontsize=10, ha='right', va='top', backgroundcolor='white')
 
     ax.set_xlabel(x1_label)
     ax.set_ylabel(y1_label)
-    # ax.legend()
-    ax.set_title(title)
+    ax.legend()
+    ax.set_title(title, fontsize=12)
 
     plt.savefig(output)
 
 if __name__ == "__main__":
     if len(sys.argv) != 7:
-        print("Usage: python script.py <title> <output_file> <csv_file1> <csv_file2> <position_feed_forward> <position_deep>")
+        print("Usage: python script.py <title> <output_file> <csv_file1> <csv_file2> <csv1_name> <csv2_name>")
         sys.exit(1)
 
     csv_file1 = sys.argv[3]
@@ -48,4 +48,4 @@ if __name__ == "__main__":
     position = sys.argv[5]
     position2 = sys.argv[6]
 
-    plot_stacked_graph(title, output, csv_file1, csv_file2, int(position), int(position2))
+    plot_stacked_graph(title, output, csv_file1, csv_file2, position, position2)

part1/average_epochs.txt

@@ -0,0 +1 @@
+1000,1000

part1/deep_train_results.csv

@@ -0,0 +1,101 @@
+Epoch,Loss
+0,0.467267
+1,0.290678
+2,0.207312
+3,0.0649596
+4,0.0599905
+5,0.0543778
+6,0.0424468
+7,0.0403263
+8,0.0378105
+9,0.0316262
+10,0.030276
+11,0.0285475
+12,0.0243477
+13,0.0233952
+14,0.0222361
+15,0.0193495
+16,0.0187022
+17,0.0179005
+18,0.0157537
+19,0.0152478
+20,0.0146454
+21,0.0129936
+22,0.0126169
+23,0.0121665
+24,0.0109912
+25,0.0106991
+26,0.0103499
+27,0.00939932
+28,0.00917552
+29,0.00890531
+30,0.00822634
+31,0.00804983
+32,0.00783543
+33,0.00727694
+34,0.00714944
+35,0.0069743
+36,0.00651646
+37,0.00638716
+38,0.0062143
+39,0.00583569
+40,0.0057501
+41,0.00563407
+42,0.00535179
+43,0.00527912
+44,0.00518645
+45,0.00492411
+46,0.00486457
+47,0.00479207
+48,0.00457332
+49,0.00452252
+50,0.0044633
+51,0.00427591
+52,0.00423684
+53,0.00419028
+54,0.00405542
+55,0.00402287
+56,0.00398389
+57,0.00388008
+58,0.00385705
+59,0.00383098
+60,0.00375437
+61,0.003736
+62,0.00371335
+63,0.00365119
+64,0.00363692
+65,0.00362008
+66,0.00356688
+67,0.00355586
+68,0.00354138
+69,0.00349546
+70,0.00348745
+71,0.0034773
+72,0.00344903
+73,0.00343734
+74,0.00342712
+75,0.0033967
+76,0.00339011
+77,0.0033799
+78,0.00335454
+79,0.00335041
+80,0.00334154
+81,0.00332021
+82,0.00331562
+83,0.00330733
+84,0.00328681
+85,0.00328304
+86,0.0032763
+87,0.00326311
+88,0.00325899
+89,0.003252
+90,0.00324033
+91,0.0032339
+92,0.00322834
+93,0.00321506
+94,0.0032241
+95,0.00323033
+96,0.0032095
+97,0.00320514
+98,0.00321329
+99,0.00318569

part1/forward_train_results.csv

@@ -0,0 +1,101 @@
+Epoch,Loss
+0,0.699479
+1,0.496314
+2,0.410334
+3,0.358543
+4,0.322386
+5,0.295046
+6,0.27331
+7,0.255423
+8,0.240327
+9,0.227347
+10,0.216018
+11,0.206008
+12,0.197078
+13,0.189043
+14,0.181763
+15,0.175128
+16,0.169047
+17,0.163448
+18,0.158271
+19,0.153466
+20,0.148992
+21,0.144812
+22,0.140896
+23,0.137217
+24,0.133754
+25,0.130487
+26,0.127397
+27,0.124471
+28,0.121694
+29,0.119055
+30,0.116543
+31,0.114149
+32,0.111865
+33,0.109682
+34,0.107593
+35,0.105593
+36,0.103676
+37,0.101836
+38,0.10007
+39,0.0983709
+40,0.0967364
+41,0.0951626
+42,0.0936461
+43,0.0921836
+44,0.0907725
+45,0.0894097
+46,0.088093
+47,0.08682
+48,0.0855883
+49,0.0843962
+50,0.0832418
+51,0.0821232
+52,0.0810387
+53,0.0799869
+54,0.0789662
+55,0.0779754
+56,0.077013
+57,0.0760781
+58,0.0751694
+59,0.0742859
+60,0.0734264
+61,0.0725901
+62,0.0717762
+63,0.0709837
+64,0.0702117
+65,0.0694594
+66,0.0687262
+67,0.0680115
+68,0.0673145
+69,0.0666345
+70,0.0659711
+71,0.0653235
+72,0.0646913
+73,0.064074
+74,0.063471
+75,0.0628818
+76,0.062306
+77,0.0617431
+78,0.0611928
+79,0.0606546
+80,0.0601282
+81,0.0596132
+82,0.0591092
+83,0.0586158
+84,0.0416983
+85,0.0168184
+86,0.0174127
+87,0.0173631
+88,0.0173133
+89,0.0172655
+90,0.017219
+91,0.0171735
+92,0.0171297
+93,0.0170865
+94,0.0170442
+95,0.017003
+96,0.0169629
+97,0.0169233
+98,0.0168849
+99,0.0168473

part1/test_results_table.txt

@@ -0,0 +1,9 @@
+\begin{figure}
+	\begin{tabular}{|c|c|c|c|}
+		\hline
+		Test & Correct & Incorrect & Accuracy (\%) \\
+		\hline
+		Feed-Forward & 9800 & 199 & 98\\
+		Deep Learning & 9898 & 101 & 99\\
+	\end{tabular}
+\end{figure}

src/MNIST.cpp

@@ -684,6 +684,34 @@ namespace fp
         return run_network_tests<net_type_ff>(path, "feed_forward", runs, restore);
     }
 
+    auto run_deep_learning_tests_v2(const std::string& path, const blt::i32 runs, const bool restore)
+    {
+        using namespace dlib;
+        using net_type_dl = loss_multiclass_log<
+            fc<10,
+               relu<fc<16,
+                       relu<fc<16,
+                               max_pool<2, 2, 2, 2, relu<con<16, 5, 5, 1, 1,
+                                                             max_pool<2, 2, 2, 2, relu<con<6, 5, 5, 1, 1,
+                                                                                           input<matrix<blt::u8>>>>>>>>>>>>>>;
+        BLT_TRACE("Running deep learning tests");
+        return run_network_tests<net_type_dl>(path, "deep_learning", runs, restore);
+    }
+
+    auto run_feed_forward_tests_v2(const std::string& path, const blt::i32 runs, const bool restore)
+    {
+        using namespace dlib;
+
+        using net_type_ff = loss_multiclass_log<
+            fc<10,
+               relu<fc<16,
+                       relu<fc<16,
+                               input<matrix<blt::u8>>>>>>>>;
+
+        BLT_TRACE("Running feed forward tests");
+        return run_network_tests<net_type_ff>(path, "feed_forward", runs, restore);
+    }
+
     void run_mnist(const int argc, const char** argv)
     {
         binary_directory = std::filesystem::current_path();
@@ -756,8 +784,10 @@ namespace fp
         const auto restore = args.get<bool>("restore");
         auto path = binary_directory + args.get<std::string>("network");
 
-        auto [deep_stats, deep_tests] = run_deep_learning_tests(path, runs, restore);
-        auto [forward_stats, forward_tests] = run_feed_forward_tests(path, runs, restore);
+        // auto [deep_stats, deep_tests] = run_deep_learning_tests(path, runs, restore);
+        auto [forward_stats, forward_tests] = run_feed_forward_tests_v2(path, runs, restore);
+        auto [deep_stats, deep_tests] = run_deep_learning_tests_v2(path, runs, restore);
+        // auto [forward_stats, forward_tests] = run_feed_forward_tests(path, runs, restore);
 
         auto average_forward_size = forward_stats.average_size();
         auto average_deep_size = deep_stats.average_size();