search

leokarlin / LaSO

LaSO: Label-Set Operations networks for multi-label few-shot learning - official implementation
BSD 3-Clause "New" or "Revised" License
86 stars 14 forks source link

doesn‘t start training after running the code #10

Closed MyChocer closed 4 years ago

MyChocer commented 4 years ago

Hi, when I run the training script, it does not start training and the terminal freezes as the content below. Did I omit some important step? '' python train_setops_stripped.py --resume_path=/paperBaseModel --init_inception=True --sets_basic_block_name=SetopResBasicBlock --sets_block_name=SetopResBlock_v1 --sets_network_name=SetOpsResModule --ops_latent_dim=8092 --ops_layer_num=2 --base_network_name=Inception3 --crop_size=299 --epochs=50 --train_base=False --coco_path= --results_path= ''

Terminal output [Main] WARNING | Unrecognized alias: '--results_path=../results_vg', it will probably have no effect. INFO:root:Created results path: /tmp/results/train_setops_stripped/0068_4825993/noJobID/200221_011551 Writing default config to: /tmp/results/train_setops_stripped/0068_4825993/noJobID/200221_011551/config.py GET /api/2.0/mlflow/experiments/list?view_type=ALL HTTP/1.1 Host: localhost:5000 User-Agent: mlflow-python-client/1.2.0 Accept-Encoding: gzip, deflate Accept: */* Connection: keep-alive

mohong96 commented 4 years ago

Have u solved the question? I have the same question now,i will be grateful if u can share the solution,thx.

MyChocer commented 4 years ago

Have u solved the question? I have the same question now,i will be grateful if u can share the solution,thx.

I run the command ''' mlflow server --host localhost ''' before running the training script.

In fact, the page shows that it is the prerequisite to run the training code. But I ignored this before.

mohong96 commented 4 years ago

Have u solved the question? I have the same question now,i will be grateful if u can share the solution,thx.

I run the command ''' mlflow server --host localhost ''' before running the training script.

In fact, the page shows that it is the prerequisite to run the training code. But I ignored this before.

thx,i ignored it too.