Comments (4)
I am working on the fix will raise a PR soon.
from pymarlin.
From more debugging it seems like a bug in gather_tensors_on_cpu
def gather_tensors_on_cpu(self, x: torch.tensor):
n_samples = len(x)
self._set_gather_frequency(n_samples)
gathered = []
n_chunks = n_samples // self.gather_frequency + 1
print(n_chunks, n_samples, self.gather_frequency) # Debug code introduced
Output
1510 3018 2
This bug is because of self._set_gather_frequency(n_samples).
In case of multiple outputs, if dimension 0 of first output was 2, then gather_frequency will be set as 2 for rest of the outputs. Class variable assignment needs to be avoided here.
from pymarlin.
Also if n_chunks is different in 2 processes, all-gather gets stuck as the process with higher number of chunks keeps waiting. For this, either make num_chunks = 1 or gather the num_chunks tensor first and take the maximum.
from pymarlin.
Thanks for debugging this @krishansubudhi - any chance you'd be able to put these changes in or should we try to resource for next sprint?
from pymarlin.
Related Issues (20)
- refactor to units of steps instead of epochs
- CustomArgParser doesn't support the yaml_file_arg_key functionality
- user defined save_best_checkpoint() hook HOT 1
- Custom Argparser doesn't support multi-level commandline override HOT 2
- DeepSpeed Trainer Backend
- Update installation documentation HOT 1
- Add integration test for DDP HOT 1
- Dockerfile and CI pipeline for Docker image
- Easy self.log() and other QoL improvements such as sane defaults for module_interface
- Running validation more often than once an epoch.
- 这个真没必要
- tb_log_hist_steps is misplaced in the Tensorboard writer
- log_graph is pretty restrictive
- Link the blogposts?
- DDP Hangs with uneven batches
- Add support for uneven number of batches on different GPUs
- Gracefully handle hang issue when training with uneven batches via throwing an exception
- Restrict scope of modification of root logger to logging_utils HOT 1
- Enable distributed training over multiple node in compliant detonation chamber 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 pymarlin.