File size: 2,507 Bytes
6ac44e1 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 |
import os
import pandas as pd
import matplotlib.pyplot as plt
def plot_training_loss(exp_dir, epoch):
# Build paths
log_dir = os.path.join(exp_dir, 'log')
train_pkl = os.path.join(log_dir, f'temp_train_loss_{epoch}.pkl')
if not os.path.isfile(train_pkl):
raise FileNotFoundError(f"Training pickle not found: {train_pkl}")
# Load data
df = pd.read_pickle(train_pkl)
epoch_losses = [float(x) for x in df['train_epoch_losses']]
# epoch_losses_temp = [float(x) for x in df['train_epoch_losses_temp']]
# epoch_losses_pos = [float(x) for x in df['train_epoch_losses_pos']]
# epoch_losses_von = [float(x) for x in df['train_epoch_losses_von']]
lrs = df['learning_rate']
epochs = list(range(1, len(epoch_losses) + 1))
print("min training loss", df['train_min_epoch_loss'])
# Plot
fig, ax1 = plt.subplots(figsize=(10,5))
ax1.plot(epochs, epoch_losses, marker='o', linestyle='-', label='Total Train Loss', color = 'blue')
# ax1.plot(epochs, epoch_losses_temp, marker='o', linestyle='-', label='Temp Train Loss', color = 'yellow')
# ax1.plot(epochs, epoch_losses_pos, marker='o', linestyle='-', label='Pos Train Loss', color = 'green')
# ax1.plot(epochs, epoch_losses_von, marker='o', linestyle='-', label='Von Train Loss', color = 'pink')
ax1.set_xlabel('Epoch')
ax1.set_ylabel('Loss', color='blue')
ax1.tick_params(axis='y', labelcolor='blue')
ax2 = ax1.twinx()
ax2.plot(epochs, lrs, linestyle='--', label='Learning Rate', color='red')
ax2.set_ylabel('Learning Rate', color='red')
ax2.tick_params(axis='y', labelcolor='red')
# Combine legends
lines, labels = ax1.get_legend_handles_labels()
lines2, labels2 = ax2.get_legend_handles_labels()
ax1.legend(lines + lines2, labels + labels2, loc='upper right')
ax1.set_title(f'Training Loss & Learning Rate (epoch {epoch})')
ax1.grid(True)
# Save
save_dir = os.path.join(exp_dir, 'plots_lr')
os.makedirs(save_dir, exist_ok=True)
out_path = os.path.join(save_dir, f'training_loss_epoch_{epoch}.png')
plt.tight_layout()
plt.savefig(out_path)
plt.close(fig)
print(f"Saved training plot to {out_path}")
def main():
exp_dir = '/home/rachit/GlassForming/t_dest_output/regDGCNN_seg/train/EXPERIMENT_2/Mon-Jul-14-10-01-31-2025/'
epoch = 40
plot_training_loss(exp_dir, epoch)
if __name__ == '__main__':
main() |