Comments (4)
In CornerNet.json,you should change both batch_size and chunk_sizes.For example, the batch_size is 30, the chunk_sizes is [15, 15]
from cornernet-lite.
In CornerNet.json,you should change both batch_size and chunk_sizes.For example, the batch_size is 30, the chunk_sizes is [15, 15]
thx for your answer.^_^
my CornerNet.json is like below:
But it's still OOM T_T
"system": {
"dataset": "COCO",
"batch_size": 1,
"sampling_function": "cornernet",
"train_split": "trainval",
"val_split": "minival",
"learning_rate": 0.00025,
"decay_rate": 10,
"val_iter": 100,
"opt_algo": "adam",
"prefetch_size": 5,
"max_iter": 500000,
"stepsize": 450000,
"snapshot": 5000,
"chunk_sizes": [15, 15],
"data_dir": "./data"
}
[root@localhost CornerNet-Lite]# python3 demo.py
total parameters: 116969339
loading from /data/MachineLearning/CornerNet-Lite/core/../cache/nnet/CornerNet_Saccade/CornerNet_Saccade_500000.pkl
/usr/local/python3/lib/python3.7/site-packages/torch/nn/functional.py:2423: UserWarning: Default upsampling behavior when mode=bilinear is changed to align_corners=False since 0.4.0. Please specify align_corners=True if the old behavior is desired. See the documentation of nn.Upsample for details.
"See the documentation of nn.Upsample for details.".format(mode))
Traceback (most recent call last):
File "demo.py", line 10, in
bboxes = detector(image)
File "/data/MachineLearning/CornerNet-Lite/core/base.py", line 25, in call
bboxes = self._inference(image, *args, **kwargs)
File "/data/MachineLearning/CornerNet-Lite/core/base.py", line 21, in _inference
return self._func(self._db, self._nnet, image.copy(), *args, **kwargs)
File "/data/MachineLearning/CornerNet-Lite/core/test/cornernet_saccade.py", line 346, in cornernet_saccade_inference
dets, atts = decode_func(db, nnet, images, no_att=no_att)
File "/data/MachineLearning/CornerNet-Lite/core/test/cornernet_saccade.py", line 77, in batch_decode
test=True, num_dets=num_dets, no_border=True, no_att=no_att
File "/data/MachineLearning/CornerNet-Lite/core/nnet/py_factory.py", line 112, in test
return self.model(*xs, **kwargs)
File "/usr/local/python3/lib/python3.7/site-packages/torch/nn/modules/module.py", line 489, in call
result = self.forward(*input, **kwargs)
File "/data/MachineLearning/CornerNet-Lite/core/nnet/py_factory.py", line 30, in forward
return self.module(*xs, **kwargs)
File "/usr/local/python3/lib/python3.7/site-packages/torch/nn/modules/module.py", line 489, in call
result = self.forward(*input, **kwargs)
File "/data/MachineLearning/CornerNet-Lite/core/models/py_utils/modules.py", line 303, in forward
return self._test(*xs, **kwargs)
File "/data/MachineLearning/CornerNet-Lite/core/models/py_utils/modules.py", line 280, in _test
cnvs, ups = self.hg(image)
File "/usr/local/python3/lib/python3.7/site-packages/torch/nn/modules/module.py", line 489, in call
result = self.forward(*input, **kwargs)
File "/data/MachineLearning/CornerNet-Lite/core/models/py_utils/modules.py", line 234, in forward
inter = self.intersind
File "/usr/local/python3/lib/python3.7/site-packages/torch/nn/modules/module.py", line 489, in call
result = self.forward(*input, **kwargs)
File "/data/MachineLearning/CornerNet-Lite/core/models/py_utils/utils.py", line 194, in forward
return self.relu(bn2 + skip)
RuntimeError: CUDA out of memory. Tried to allocate 44.00 MiB (GPU 0; 1.95 GiB total capacity; 1.21 GiB already allocated; 20.62 MiB free; 75.49 MiB cached)
[root@localhost CornerNet-Lite]#
from cornernet-lite.
How many GPU do you have? If you have just one, I think you can try set chunk_sizes=[1].
from cornernet-lite.
How many GPU do you have? If you have just one, I think you can try set chunk_sizes=[1].
OOM T_T
But if I use debug mode to execute setp by setp, it can be done.
Maybe I should change my video card >_<
[root@localhost ~]# nvidia-smi
Tue May 7 20:31:20 2019
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 418.56 Driver Version: 418.56 CUDA Version: 10.1 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 GeForce GTX 750 Ti Off | 00000000:01:00.0 On | N/A |
| 38% 28C P8 1W / 38W | 222MiB / 1999MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 12655 G /usr/bin/X 109MiB |
| 0 13192 G /usr/bin/gnome-shell 110MiB |
+-----------------------------------------------------------------------------+
[root@localhost ~]#
from cornernet-lite.
Related Issues (20)
- Using a data set with only one category, Loss dropped to 0.003, but when testing, ap = -1 HOT 1
- AttributeError: 'builtin_function_or_method' object has no attribute 'view' HOT 1
- ModuleNotFoundError: No module named 'core' HOT 3
- Has anyone use another backbone networks to test the performance?
- About "add downsampling lyaer before the hourglass module and remove one in hourglass modue" HOT 1
- train error
- When I testing my own data, occur No module named 'test.xxx'
- The network architecture of CornerNet-Saccade
- Can not create the envs on the first step, list many config package when created from file conda_packagelist.txt Please help me!! HOT 1
- Duplicated boxes during soft_nms HOT 1
- some training issue HOT 1
- some questions about the structure of cornerNet-saccade HOT 1
- when run the demo.py, the program is stuck
- 0%| | 0/90000 [00:00<?, ?it/s]段错误(吐核)
- ImportError undefined symbol: _ZNSt19basic_ostringstreamIcSt11char_traitsIcESaIcEEC1Ev HOT 1
- When I train the model on my own dataset, I met IndexError in cornernet_saccade.py HOT 1
- error while tarining on my new dataset which has same COCO format HOT 1
- A small running error
- About the software requirements HOT 1
- [W Resize.cpp:19] Warning: An output with one or more elements was resized since it had shape [16263], which does not match the required output shape [14926].This behavior is deprecated, and in a future PyTorch release outputs will not be resized unless they have zero elements. You can explicitly reuse an out tensor t by resizing it, inplace, to zero elements with t.resize_(0). (function resize_output) HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from cornernet-lite.