nohup
The command can make the command execute permanently, and it has nothing to do with the terminal, and exiting the terminal will not affect the running of the program;
&
it means running in the background, but when the user exits, the command automatically exits.
Then, combine the two nohup 命令 &
so that the command is permanently executed in the background
Take the run_train.sh
file as an example
source env_set.sh
nohup python -u train_image_classifier.py \
--dataset_name=$DATASET_NAME \
--dataset_dir=$DATASET_DIR \
--checkpoint_path=$CHECKPOINT_PATH \
--model_name=inception_v4 \
--checkpoint_exclude_scopes=InceptionV4/Logits,InceptionV4/AuxLogits/Aux_logits \
--trainable_scopes=InceptionV4/Logits,InceptionV4/AuxLogits/Aux_logits \
--train_dir=$TRAIN_DIR \
--learning_rate=0.001 \
--learning_rate_decay_factor=0.76\
--num_epochs_per_decay=50 \
--moving_average_decay=0.9999 \
--optimizer=adam \
--ignore_missing_vars=True \
--batch_size=32 > output.log 2>&1 &
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 12
- 13
- 14
- 15
- 16
- 17
- 18
To run the program in TensorFlow, run_train.sh
add nohup 命令 > output.log 2>&1 &
the command before and after the file to execute the command in the background.
0, 1, and 2 represent the following meanings:
0 – stdin (standard input)
1 – stdout (standard output)
2 – stderr (standard error)
nohup
+ The last &
is to let the command execute in the background
>output.log
is to output information to the output.log log
2>&1
It is to convert the standard error information into standard output, so that the error information can be output to the output.log log.
View log (dynamic display)
tail -f output.log
- 1
View the log (show the entire file at once)
cat output.log
- 1
View the current Python process
ps -ef |grep python
- 1
kill process
sudo kill 进程号
- 1
kill 9 进程号 #绝杀
- 1
nohup
The command can make the command execute permanently, and it has nothing to do with the terminal, and exiting the terminal will not affect the running of the program;
&
it means running in the background, but when the user exits, the command automatically exits.
Then, combine the two nohup 命令 &
so that the command is permanently executed in the background
Take the run_train.sh
file as an example
source env_set.sh
nohup python -u train_image_classifier.py \
--dataset_name=$DATASET_NAME \
--dataset_dir=$DATASET_DIR \
--checkpoint_path=$CHECKPOINT_PATH \
--model_name=inception_v4 \
--checkpoint_exclude_scopes=InceptionV4/Logits,InceptionV4/AuxLogits/Aux_logits \
--trainable_scopes=InceptionV4/Logits,InceptionV4/AuxLogits/Aux_logits \
--train_dir=$TRAIN_DIR \
--learning_rate=0.001 \
--learning_rate_decay_factor=0.76\
--num_epochs_per_decay=50 \
--moving_average_decay=0.9999 \
--optimizer=adam \
--ignore_missing_vars=True \
--batch_size=32 > output.log 2>&1 &
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 12
- 13
- 14
- 15
- 16
- 17
- 18
To run the program in TensorFlow, run_train.sh
add nohup 命令 > output.log 2>&1 &
the command before and after the file to execute the command in the background.
0, 1, and 2 represent the following meanings:
0 – stdin (standard input)
1 – stdout (standard output)
2 – stderr (standard error)
nohup
+ The last &
is to let the command execute in the background
>output.log
is to output information to the output.log log
2>&1
It is to convert the standard error information into standard output, so that the error information can be output to the output.log log.
View log (dynamic display)
tail -f output.log
- 1
View the log (show the entire file at once)
cat output.log
- 1
View the current Python process
ps -ef |grep python
- 1
kill process
sudo kill 进程号
- 1
kill 9 进程号 #绝杀
- 1