| @@ -52,38 +52,37 @@ namespace TensorFlowNET.Examples.CnnTextClassification | |||||
| protected virtual bool RunWithImportedGraph(Session sess, Graph graph) | protected virtual bool RunWithImportedGraph(Session sess, Graph graph) | ||||
| { | { | ||||
| var stopwatch = Stopwatch.StartNew(); | |||||
| Console.WriteLine("Building dataset..."); | Console.WriteLine("Building dataset..."); | ||||
| var (x, y, alphabet_size) = DataHelpers.build_char_dataset("train", model_name, CHAR_MAX_LEN, DataLimit=null); | var (x, y, alphabet_size) = DataHelpers.build_char_dataset("train", model_name, CHAR_MAX_LEN, DataLimit=null); | ||||
| Console.WriteLine("\tDONE"); | |||||
| Console.WriteLine("\tDONE "); | |||||
| var (train_x, valid_x, train_y, valid_y) = train_test_split(x, y, test_size: 0.15f); | var (train_x, valid_x, train_y, valid_y) = train_test_split(x, y, test_size: 0.15f); | ||||
| Console.WriteLine("Import graph..."); | Console.WriteLine("Import graph..."); | ||||
| var meta_file = model_name + ".meta"; | var meta_file = model_name + ".meta"; | ||||
| tf.train.import_meta_graph(Path.Join("graph", meta_file)); | tf.train.import_meta_graph(Path.Join("graph", meta_file)); | ||||
| Console.WriteLine("\tDONE"); | |||||
| // definitely necessary, otherwize will get the exception of "use uninitialized variable" | |||||
| Console.WriteLine("\tDONE " + stopwatch.Elapsed); | |||||
| sess.run(tf.global_variables_initializer()); | sess.run(tf.global_variables_initializer()); | ||||
| var train_batches = batch_iter(train_x, train_y, BATCH_SIZE, NUM_EPOCHS); | var train_batches = batch_iter(train_x, train_y, BATCH_SIZE, NUM_EPOCHS); | ||||
| var num_batches_per_epoch = (len(train_x) - 1); // BATCH_SIZE + 1 | |||||
| var num_batches_per_epoch = (len(train_x) - 1) / BATCH_SIZE + 1; | |||||
| double max_accuracy = 0; | double max_accuracy = 0; | ||||
| Tensor is_training = graph.get_operation_by_name("is_training"); | Tensor is_training = graph.get_operation_by_name("is_training"); | ||||
| Tensor model_x = graph.get_operation_by_name("x"); | Tensor model_x = graph.get_operation_by_name("x"); | ||||
| Tensor model_y = graph.get_operation_by_name("y"); | Tensor model_y = graph.get_operation_by_name("y"); | ||||
| Tensor loss = graph.get_operation_by_name("loss/loss"); | |||||
| Tensor loss = graph.get_operation_by_name("loss/value"); | |||||
| //var optimizer_nodes = graph._nodes_by_name.Keys.Where(key => key.Contains("optimizer")).ToArray(); | //var optimizer_nodes = graph._nodes_by_name.Keys.Where(key => key.Contains("optimizer")).ToArray(); | ||||
| Tensor optimizer = graph.get_operation_by_name("loss/optimizer"); | Tensor optimizer = graph.get_operation_by_name("loss/optimizer"); | ||||
| Tensor global_step = graph.get_operation_by_name("global_step"); | Tensor global_step = graph.get_operation_by_name("global_step"); | ||||
| Tensor accuracy = graph.get_operation_by_name("accuracy/accuracy"); | |||||
| var stopwatch = Stopwatch.StartNew(); | |||||
| Tensor accuracy = graph.get_operation_by_name("accuracy/value"); | |||||
| stopwatch = Stopwatch.StartNew(); | |||||
| int i = 0; | int i = 0; | ||||
| foreach (var (x_batch, y_batch, total) in train_batches) | foreach (var (x_batch, y_batch, total) in train_batches) | ||||
| { | { | ||||
| i++; | i++; | ||||
| var estimate = TimeSpan.FromSeconds((stopwatch.Elapsed.TotalSeconds / i) * total); | |||||
| Console.WriteLine($"Training on batch {i}/{total}. Estimated training time: {estimate}"); | |||||
| var train_feed_dict = new Hashtable | var train_feed_dict = new Hashtable | ||||
| { | { | ||||
| [model_x] = x_batch, | [model_x] = x_batch, | ||||
| @@ -94,9 +93,14 @@ namespace TensorFlowNET.Examples.CnnTextClassification | |||||
| //_, step, loss = sess.run([model.optimizer, model.global_step, model.loss], feed_dict = train_feed_dict) | //_, step, loss = sess.run([model.optimizer, model.global_step, model.loss], feed_dict = train_feed_dict) | ||||
| var result = sess.run(new ITensorOrOperation[] { optimizer, global_step, loss }, train_feed_dict); | var result = sess.run(new ITensorOrOperation[] { optimizer, global_step, loss }, train_feed_dict); | ||||
| //loss_value = result[2]; | //loss_value = result[2]; | ||||
| var step = result[1]; | |||||
| var step = result[1]; | |||||
| if (step % 10 == 0) | if (step % 10 == 0) | ||||
| { | |||||
| var estimate = TimeSpan.FromSeconds((stopwatch.Elapsed.TotalSeconds / i) * total); | |||||
| Console.WriteLine($"Training on batch {i}/{total}. Estimated training time: {estimate}"); | |||||
| Console.WriteLine($"Step {step} loss: {result[2]}"); | Console.WriteLine($"Step {step} loss: {result[2]}"); | ||||
| } | |||||
| if (step % 100 == 0) | if (step % 100 == 0) | ||||
| { | { | ||||
| continue; | continue; | ||||
| @@ -198,6 +202,8 @@ namespace TensorFlowNET.Examples.CnnTextClassification | |||||
| { | { | ||||
| // download graph meta data | // download graph meta data | ||||
| var meta_file = model_name + ".meta"; | var meta_file = model_name + ".meta"; | ||||
| if (File.GetLastWriteTime(meta_file) < new DateTime(2019,05,11)) // delete old cached file which contains errors | |||||
| File.Delete(meta_file); | |||||
| url = "https://raw.githubusercontent.com/SciSharp/TensorFlow.NET/master/graph/" + meta_file; | url = "https://raw.githubusercontent.com/SciSharp/TensorFlow.NET/master/graph/" + meta_file; | ||||
| Web.Download(url, "graph", meta_file); | Web.Download(url, "graph", meta_file); | ||||
| } | } | ||||