Write the parameters of each epoch in training to the log log
Example 1:
with open(os.path.join(args.save_dir, args.exp_name, args.dataset_name, args.log_dir), "a") as train_log:
train_log.write("\r[Epoch%d]-[Loss:%f]"%(epoch+1,loss))
Display results:
Example 2
with open(os.path.join(args.save_dir, args.exp_name, args.dataset_name, args.log_dir), "a") as train_log:
train_log.write("\r[Epoch%d]-[Loss:%f]-[image_AUC:%f]-[pixel_AUC:%f]"% (epoch+1,loss, img_roc_auc, per_pixel_rocauc))
**
Display results:
**
Can learn from:
