Created by: xinaiwunai

在win10下操作，命令行如下。完成训练后，并没有任何结果在--output_dir ./model_files中。请问如何才能将训练后的结果保存？

PS E:\paddlehub\models\PaddleNLP\similarity_net> python run_classifier.py    --task_name 'wyh_simmilarity_net_fine_tune_test'    --use_cuda false     --do_train True      --do_valid True      --do_test True       --do_infer True     --batch_size 128     --train_data_dir ./data/qqsim     --valid_data_dir ./data/qqsim     --test_data_dir ./data/qqsim       --infer_data_dir ./data/infer_data     --output_dir ./model_files      --config_path ./config/bow_pairwise.json     --vocab_path ./data/term2id.dict       --epoch 10        --save_steps 1000      --validation_steps 100   --task_mode 'pairwise'  --compute_accuracy False      --lamda 0.958
-----------  Configuration Arguments -----------
batch_size: 128
compute_accuracy: False
config_path: ./config/bow_pairwise.json
do_infer: True
do_test: True
do_train: True
do_valid: True
enable_ce: False
epoch: 10
infer_data_dir: ./data/infer_data
infer_result_path: infer_result
init_checkpoint: None
lamda: 0.958
output_dir: ./model_files
save_steps: 1000
skip_steps: 10
task_mode: pairwise
task_name: wyh_simmilarity_net_fine_tune_test
test_data_dir: ./data/qqsim
test_result_path: test_result
train_data_dir: ./data/qqsim
use_cuda: False
valid_data_dir: ./data/qqsim
validation_steps: 100
verbose_result: True
vocab_path: ./data/term2id.dict
------------------------------------------------

device count: 1
start train process ...
I0918 08:48:18.885494  9960 parallel_executor.cc:334] The number of CPUPlace, which is used in ParallelExecutor, is 1. And the Program will be copied 1 copies
I0918 08:48:18.889495  9960 build_strategy.cc:340] SeqOnlyAllReduceOps:0, num_trainers:1
epoch: 0, loss: 0.004120, used time: 7 sec
epoch: 1, loss: 0.000024, used time: 7 sec
epoch: 2, loss: 0.000006, used time: 7 sec
epoch: 3, loss: 0.000000, used time: 7 sec
epoch: 4, loss: 0.000000, used time: 7 sec
epoch: 5, loss: 0.000000, used time: 7 sec
global_steps: 100, valid_auc: 0.605413
epoch: 6, loss: 0.000000, used time: 7 sec
epoch: 7, loss: 0.000000, used time: 7 sec
epoch: 8, loss: 0.000000, used time: 7 sec
epoch: 9, loss: 0.000000, used time: 7 sec
AUC of test is 0.605461