Comments (4)
elisa-aleman
commented on September 28, 2024
Also related to #553
from mmrazor.
chenjie04
commented on September 28, 2024
If you look at the source code, this error does not require torch to be higher than 1.13, but less than or equal to 1.13, and degrading torch will fix the problem.
from mmrazor.
elisa-aleman
commented on September 28, 2024
the source code
Then the requirements files need to be updated and follow PEP.
degrading torch will fix the problem.
Regardless, torch 1.13 is extremely outdated, please update the source code.
from mmrazor.
elisa-aleman
commented on September 28, 2024
Note, the suggested fix above will not work with fusions because of the changes in the BackendPatternConfig from torch 1 to torch 2. Any model with potential fusions will have mishaps in torch 2 unless updating these BackendPatternConfig to match the new version.
from mmrazor.
Related Issues (20)
- [Bug] TypeError: 'NoneType' object is not iterable
- Try to reproduce CWD in VOC data set
- [Bug] (suggested fix) `nn.Parameter` are not added to root after being traced in `mmrazor.models.task_utils.tracer.fx.custom_tracer.build_graphmodule()` HOT 2
- [Bug] (suggested fix) `mmrazor.models.algorithms.quantization.mm_architecture.MMArchitectureQuant.sync_qparams()` fails if there are modules present in other modes but not in forward `mode='tensor'` HOT 4
- I want to obtain the current epoch value and associate it with the custom distillation loss
- cannot use recorder to obtain panoptic_head info from mask2former
- [Bug] `mmrazor.engine.runner.quantization_loops.QATValLoop` calls `after_val_epoch` hook twice with different keys, causing `mmengine.hooks.checkpoint_hook._save_best_checkpoint()` to fail with `KeyError` for the `save_best` config
- [Bug] Custom Distillation MMSeg CWD loss nan problem
- When I use methodoutputs to access the results of assigner, I only obtain one sample
- Regarding tables and accuracy
- [Bug] (suggested fix) `mmrazor.models.algorithms.mm_architecture.MMArchitectureQuant.get_deploy_model()` fails if `predict` mode lacks nodes from the `model.quantizer.tracer.skipped_methods` configuration, but the architecture `quantizer.prepare(fp32_model)` has these nodes. HOT 4
- Is this a dead project ? HOT 1
- 我在用mmrazor通过yolov5-x蒸馏yolov5-s时候遇到了问题 HOT 1
- No Sign of activation quantization with QAT HOT 1
- MAP is stucked at 0 for Mobilenet V2 SSD QAT without pretrained model [Bug]
- [Docs] A100算力加持！书生大模型实战营第3期全面升级，趣味闯关模式等你开启
- Missing keys after RTMDET knowledge distillation HOT 1
- Problematic difference between FakeQuantize for activations and weights HOT 1
- TypeError: 'ABCMeta' Object is not subscriptable when configuring mmseg model for PTQ and QAT HOT 4
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 mmrazor.