Views
No views yet
[I 2024-05-08 06:09:44,865] # of parameters: 783,674,400 | # of trainable parameters: 524,320
{'eval_loss': 0.796862781047821, 'eval_rouge1': 0.7621970683993027, 'eval_rouge2': 0.5604146098335534, 'eval_rougeL': 0.6793146845525677, 'eval_rougeLsum': 0.6793904348801474, 'eval_exact_match': 0.04385601985932975, 'eval_answer_f1': 0.755, 'eval_bleu': 0.4230708837673167, 'eval_bleu_precision': [0.7748865411799717, 0.560967534617407, 0.4202738606172083, 0.33109023283813127], 'eval_runtime': 65.5651, 'eval_samples_per_second': 36.864, 'eval_steps_per_second': 0.58, 'epoch': 1.0}
{'eval_loss': 0.7385475635528564, 'eval_rouge1': 0.7784561001768054, 'eval_rouge2': 0.582245821304826, 'eval_rougeL': 0.7003109404941752, 'eval_rougeLsum': 0.7006334562748413, 'eval_exact_match': 0.05544062887877534, 'eval_answer_f1': 0.772, 'eval_bleu': 0.46308663903699265, 'eval_bleu_precision': [0.7800815892160341, 0.5761540314660045, 0.44073254956737673, 0.35349793401046226], 'eval_runtime': 68.8285, 'eval_samples_per_second': 35.116, 'eval_steps_per_second': 0.552, 'epoch': 2.0}
{'eval_loss': 0.7233059406280518, 'eval_rouge1': 0.7838531547066641, 'eval_rouge2': 0.5913115317921096, 'eval_rougeL': 0.7090890606183786, 'eval_rougeLsum': 0.7090586788376629, 'eval_exact_match': 0.063301613570542, 'eval_answer_f1': 0.777, 'eval_bleu': 0.4810574504704876, 'eval_bleu_precision': [0.7944260779922505, 0.5963970656514068, 0.4629190324336937, 0.37524229254657115], 'eval_runtime': 66.6726, 'eval_samples_per_second': 36.252, 'eval_steps_per_second': 0.57, 'epoch': 3.0}
{'eval_loss': 0.7169296741485596, 'eval_rouge1': 0.7869214304468117, 'eval_rouge2': 0.5977398799036677, 'eval_rougeL': 0.7144684199858549, 'eval_rougeLsum': 0.7146933230523775, 'eval_exact_match': 0.06992139015308234, 'eval_answer_f1': 0.781, 'eval_bleu': 0.4883685660838088, 'eval_bleu_precision': [0.8045431672776939, 0.610235406842392, 0.4788033851604015, 0.39063500919554484], 'eval_runtime': 66.0209, 'eval_samples_per_second': 36.61, 'eval_steps_per_second': 0.576, 'epoch': 4.0}
{'eval_loss': 0.714337170124054, 'eval_rouge1': 0.7868999187069199, 'eval_rouge2': 0.5981371847854916, 'eval_rougeL': 0.715028234747944, 'eval_rougeLsum': 0.7149340809828948, 'eval_exact_match': 0.06950765411667356, 'eval_answer_f1': 0.781, 'eval_bleu': 0.48919438065439813, 'eval_bleu_precision': [0.8054662090528472, 0.6109817034226246, 0.4796426392742586, 0.39144628099173556], 'eval_runtime': 65.7858, 'eval_samples_per_second': 36.74, 'eval_steps_per_second': 0.578, 'epoch': 5.0}
{'train_runtime': 1864.0565, 'train_samples_per_second': 53.48, 'train_steps_per_second': 0.837, 'train_loss': 0.7318433712690304, 'epoch': 5.0}
100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1560/1560 [31:04<00:00, 1.19s/it]
[I 2024-05-08 06:40:50,207] Trial 32 finished with value: 0.7149340809828948 and parameters: {'batch_size': 64, 'dropout_p': 0.5, 'learning_rate': 0.0528246346225972, 'num_train_epochs': 5}. Best is trial 32 with value: 0.7149340809828948.