Closed gaurish1212 closed 2 months ago
lbj96347
commented
5 months ago I guess Torchserve is failing because it doesn't have enough RAM. You may also try to increase RAM for docker.
How about checking your RAM after u enter python image_to_animation.py drawings/garlic.png garlic_out ?
gaurish1212
commented
5 months ago Hy How can i increase ram for docker. In my system have 32GB ram
corporateGhost
commented
5 months ago Same Question
lbj96347
commented
5 months ago @gaurish1212 by default, docker can use unlimited RAM in your system.
After running docker run -d --name docker_torchserve -p 8080:8080 -p 8081:8081 docker_torchserve
Try to get more detail from container : docker exec -it docker_torchserve bash
All logs while running docker_torchserve cat /home/torchserve/logs/ts_log.log
Then we may figure out which part goes wrong. Or post it here.
Also I found a simular issue: https://github.com/facebookresearch/AnimatedDrawings/issues/136
lbj96347
commented
5 months ago corporateGhost
commented
5 months ago @gaurish1212 try it: #240 (comment)
I tried this. I changed it from RUN pip install torch to RUN pip install torch==2.0.0 but still it gives me the same error
corporateGhost
commented
5 months ago @gaurish1212 by default, docker can use unlimited RAM in your system.
After running
docker run -d --name docker_torchserve -p 8080:8080 -p 8081:8081 docker_torchserveTry to get more detail from container :
docker exec -it docker_torchserve bashAll logs while running docker_torchserve
cat /home/torchserve/logs/ts_log.logThen we may figure out which part goes wrong. Or post it here.
Also I found a simular issue: #136
I tried this and I am getting this (base) H:\AnimatedDrawings\torchserve>docker exec -it docker_torchserve bash (base) root@6eab2021b0eb:/# cat /home/torchserve/logs/ts_log.log cat: /home/torchserve/logs/ts_log.log: No such file or directory
lbj96347
commented
5 months ago @corporateGhost if u can't see ts_log.log , u can try to run bash inside docker container.
step 1:
docker exec -it docker_torchserve bash
step 2:
torchserve --stop
step3:
/opt/conda/bin/torchserve --start --ts-config /home/torchserve/config.properties
Now you can see logs popping up.
gaurish1212
commented
3 months ago This is our logs
2024-03-20T12:17:46,825 [INFO ] W-9015-drawn_humanoid_detector_1.0-stdout MODEL_LOG - File "/opt/conda/lib/python3.11/site-packages/mmcv/ops/active_rotated_filter.py", line 10, in
gaurish1212
commented
3 months ago We didn't get any output
This is my docker image
Please guide me
gaurish1212
commented
3 months ago @corporateGhost if u can't see
ts_log.log, u can try to run bash inside docker container.step 1:
docker exec -it docker_torchserve bash
step 2:
torchserve --stop
step3:
/opt/conda/bin/torchserve --start --ts-config /home/torchserve/config.properties
Now you can see logs popping up.
Plaese Reply
gaurish1212
commented
3 months ago
This is my docker logs image what's wrong and how can i fix ?
hjessmith
commented
3 months ago I just merged a PR that deals with the Docker not correctly building. Please try pulling the latest version of the Animated Drawings code and following the installation steps again. Changing mmcv-full version as you did is triggering an assertion error in the torchserve handler code which is causing it to fail.
gaurish1212
commented
2 months ago solved thanks
Hy
When we build image after change version of mmsc-full==1.7.0 instaed mmcv-full==1.7.2. Then image build successfully and we get status healthy. Then after we hit this command "python image_to_animation.py drawings/garlic.png garlic_out" then we get error. Traceback (most recent call last): File "image_to_animation.py", line 41, in
image_to_animation(img_fn, char_anno_dir, motion_cfg_fn, retarget_cfg_fn)
File "image_to_animation.py", line 19, in image_to_animation
image_to_annotations(img_fn, char_anno_dir)
File "C:\Users\DELL4\AnimatedDrawings\examples\image_to_annotations.py", line 53, in image_to_annotations
raise Exception(f"Failed to get bounding box, please check if the 'docker_torchserve' is running and healthy, resp: {resp}")
Exception: Failed to get bounding box, please check if the 'docker_torchserve' is running and healthy, resp: <Response [500]>
