updated alg init logic in main

MSc-Thesis-Samy · May 9, 2024 · 40ad3e3 · 40ad3e3
1 parent 9068d81
commit 40ad3e3
Show file tree

Hide file tree

Showing 2 changed files with 40 additions and 61 deletions.
diff --git a/src/bin/main.rs b/src/bin/main.rs
@@ -17,78 +17,57 @@ use neuroevolution::constants::*;
 use neuroevolution::neat::*;
 use neuroevolution::neural_network::NeuralNetworkConfig;
 
-fn main() {
-    let cli = Cli::parse();
-    let mut alg: Algorithm;
-    let dim = 2;
-
-    let problem = Benchmark::new(cli.problem);
-
-    match cli.algorithm {
-        AlgorithmType::Oneplusonena => {
-            // match cli.continuous {
-            //     true => {
-            //         let network = Network::new(cli.neurons, dim);
-            //         alg = Algorithm::ContinuousOneplusoneNA(network);
-            //     }
-            //     false => {
-            //         let network = DiscreteNetwork::new(cli.resolution, cli.neurons, dim);
-            //         alg = Algorithm::DiscreteOneplusoneNA(network);
-            //     }
-            // }
-
-            let network = DiscreteNetwork::new(cli.resolution, cli.neurons, dim);
-            alg = Algorithm::DiscreteOneplusoneNA(network);
-        },
-        AlgorithmType::Bna => {
-            // match cli.continuous {
-            //     true => {
-            //         let vneuron = VNeuron::new(dim);
-            //         alg = Algorithm::ContinuousBNA(vneuron);
-            //     },
-            //     false => {
-            //         let vneuron = DiscreteVNeuron::new(cli.resolution, dim);
-            //         alg = Algorithm::DiscreteBNA(vneuron);
-            //     }
-            // }
-
-            let vneuron = DiscreteVNeuron::new(cli.resolution, dim);
-            alg = Algorithm::DiscreteBNA(vneuron);
-        }
+fn get_algorithm(algorithm_type: AlgorithmType, resolution: usize, neurons: usize, dim: usize, toml_config: &Option<String>) -> Algorithm {
+    match algorithm_type {
         AlgorithmType::Neat => {
-            let config_file_path = match cli.file {
-                Some(file) => file,
-                None => panic!("No configuration file provided"),
-            };
-
-            let mut neat_config_file = File::open(config_file_path).unwrap();
-            let mut toml_config = String::new();
-            neat_config_file.read_to_string(&mut toml_config).unwrap();
-            let config: Config = toml::from_str(&toml_config).unwrap();
+            let toml_config: &str = toml_config.as_deref().unwrap();
+            let config: Config = toml::from_str(toml_config).unwrap();
 
             let neat = Neat::new(config);
-            alg = Algorithm::Neat(neat);
+            Algorithm::Neat(neat)
         }
         AlgorithmType::NeuralNetwork => {
-            let config_file_path = match cli.file {
-                Some(file) => file,
-                None => panic!("No configuration file provided"),
-            };
-
-            let mut network_config_file = File::open(config_file_path).unwrap();
-            let mut toml_config = String::new();
-            network_config_file.read_to_string(&mut toml_config).unwrap();
-
+            let toml_config: &str = toml_config.as_deref().unwrap();
             let network_config: NeuralNetworkConfig = toml::from_str(&toml_config).unwrap();
 
             let network = network_config.to_neural_network();
-            alg = Algorithm::NeuralNetwork(network);
+            Algorithm::NeuralNetwork(network)
+        }
+        AlgorithmType::Oneplusonena => {
+            let network = DiscreteNetwork::new(resolution, neurons, dim);
+            Algorithm::DiscreteOneplusoneNA(network)
+        }
+        AlgorithmType::Bna => {
+            let vneuron = DiscreteVNeuron::new(resolution, dim);
+            Algorithm::DiscreteBNA(vneuron)
         }
     }
+}
+
+fn main() {
+    let cli = Cli::parse();
+
+    let dim = 2;
+    let problem = Benchmark::new(cli.problem);
+    let resolution = cli.resolution;
+    let neurons = cli.neurons;
+
+    let mut alg: Algorithm;
+
+    let toml_config = if let Some(file) = cli.file {
+        let mut config_file = File::open(file).unwrap();
+        let mut toml_config = String::new();
+        config_file.read_to_string(&mut toml_config).unwrap();
+        Some(toml_config)
+    } else {
+        None
+    };
+
+    alg = get_algorithm(cli.algorithm, resolution, neurons, dim, &toml_config);
 
     if let Some(n_runs) = cli.test_runs {
         let results = (0..n_runs).into_par_iter().map(|_| {
-            let mut algorithm = alg.clone(); // TODO initialize with different initial values
+            let mut algorithm = get_algorithm(cli.algorithm, resolution, neurons, dim, &toml_config);
             let problem = Benchmark::new(cli.problem);
 
             let start = Instant::now();
@@ -100,7 +79,7 @@ fn main() {
 
         if let Some(output_path) = cli.output {
             let mut output_file = File::create(output_path).unwrap();
-            writeln!(output_file, "Fitness,Iterations,Elapsed time").unwrap();
+            writeln!(output_file, "fitness,iterations,cpu").unwrap();
             for (fitness, n_iters, elapsed) in results {
                 writeln!(output_file, "{:.2},{},{:.3}", fitness, n_iters, elapsed).unwrap();
             }

diff --git a/testing_job.sh b/testing_job.sh
@@ -34,5 +34,5 @@ cd ~/code-master
 
 for resolution in $(seq 100 100 1000)
 do
-      ./target/release/main oneplusonena half -i 500 -n 1 -r $resolution -t $n_runs -o output/oneplusone_na_half_$resolution.csv
+      ./target/release/main oneplusonena half -i 500 -n 1 -r $resolution -t $n_runs -o ~/output/oneplusone_na_half_$resolution.csv
 done