Updated tqdm counting issue, added script for setting up environment on head-node

cifar100_network_trainer.py

@@ -83,7 +83,7 @@ with tf.Session() as sess:
                                                    continue_from_epoch))  # restore previous graph to continue operations
 
     best_val_accuracy = 0.
-    with tqdm.tqdm(total=epochs) as epoch_pbar:
+    with tqdm.tqdm(total=epochs - start_epoch) as epoch_pbar:
         for e in range(start_epoch, epochs):
             total_c_loss = 0.
             total_accuracy = 0.

cifar10_network_trainer.py

@@ -83,7 +83,7 @@ with tf.Session() as sess:
                                                    continue_from_epoch))  # restore previous graph to continue operations
 
     best_val_accuracy = 0.
-    with tqdm.tqdm(total=epochs) as epoch_pbar:
+    with tqdm.tqdm(total=epochs-start_epoch) as epoch_pbar:
         for e in range(start_epoch, epochs):
             total_c_loss = 0.
             total_accuracy = 0.

emnist_network_trainer.py

@@ -83,7 +83,7 @@ with tf.Session() as sess:
                                                    continue_from_epoch))  # restore previous graph to continue operations
 
     best_val_accuracy = 0.
-    with tqdm.tqdm(total=epochs) as epoch_pbar:
+    with tqdm.tqdm(total=epochs - start_epoch) as epoch_pbar:
         for e in range(start_epoch, epochs):
             total_c_loss = 0.
             total_accuracy = 0.

gpu_cluster_environment_script.sh

@@ -0,0 +1,20 @@
+#!/bin/sh
+export CUDA_HOME=/opt/cuda-8.0.44
+
+export CUDNN_HOME=/opt/cuDNN-6.0_8.0
+
+export STUDENT_ID=sxxxxxx
+
+export LD_LIBRARY_PATH=${CUDNN_HOME}/lib64:${CUDA_HOME}/lib64:$LD_LIBRARY_PATH
+
+export LIBRARY_PATH=${CUDNN_HOME}/lib64:$LIBRARY_PATH
+
+export CPATH=${CUDNN_HOME}/include:$CPATH
+
+export PATH=${CUDA_HOME}/bin:${PATH}
+
+export PYTHON_PATH=$PATH
+
+# Activate the relevant virtual environment:
+
+source /home/${STUDENT_ID}/miniconda3/bin/activate mlp

msd10_network_trainer.py

@@ -84,7 +84,7 @@ with tf.Session() as sess:
                                                    continue_from_epoch))  # restore previous graph to continue operations
 
     best_val_accuracy = 0.
-    with tqdm.tqdm(total=epochs) as epoch_pbar:
+    with tqdm.tqdm(total=epochs - start_epoch) as epoch_pbar:
         for e in range(start_epoch, epochs):
             total_c_loss = 0.
             total_accuracy = 0.

msd25_network_trainer.py

@@ -83,7 +83,7 @@ with tf.Session() as sess:
                                                    continue_from_epoch))  # restore previous graph to continue operations
 
     best_val_accuracy = 0.
-    with tqdm.tqdm(total=epochs) as epoch_pbar:
+    with tqdm.tqdm(total=epochs - start_epoch) as epoch_pbar:
         for e in range(start_epoch, epochs):
             total_c_loss = 0.
             total_accuracy = 0.