Comments (6)
@SkafteNicki mind have look? :]
from torchmetrics.
Definitely an lack of support from pytorch side, but seems to be on track to be solved:
Issue: pytorch/pytorch#49889
PR with fix: pytorch/pytorch#49895
I can try to see if the code can be rewritten with operations that support float16
Probably should also add test for float16
in general for metrics.
from torchmetrics.
Just a follow up: the following metrics does not work with float16 CPU because of lack of support from a list of pytorch operations (in parentheses)
- SSIM (torch.arange)
- PSNR (torch.pow)
- MeanSquaredLogError (torch.logp1 & torch.pow)
- MeanSquaredError (torch.pow)
- ExplainedVariance (torch.pow)
- PrecisionRecallCurve (torch.flip)
- AveragePrecision (torch.flip)
@Borda how to move forward with this? I think some of these can be converted in operations that support float16 CPU (pow can be made to multiplication) but I am not sure about all of them
from torchmetrics.
from torchmetrics.
Hi! thanks for your contribution!, great first issue!
from torchmetrics.
looks like this was fixed in pytorch/pytorch@5d93e2b
from torchmetrics.
Related Issues (20)
- Support for DLM (AIM) metric HOT 1
- Error for index tensor verification in GeneralizedDice if target is zeros HOT 1
- Incorrect caching (`m._compute`) of metrics inside a `MetricCollection` if `compute_groups` are used and `.compute` is called twice HOT 1
- Classification Report, maybe with the option to select even more metrics HOT 1
- `MeanAveragePrecision.tm_to_coco` example fails with `IndexError: list index out of range` HOT 1
- The compute_groups may lead to unexpected behavior HOT 2
- GPU ram memory increase until overflow when using PSNR and SSIM HOT 2
- Need higher level RAG metrics HOT 2
- Multi output functionality for MeanAbsoluteError HOT 1
- _cumsum moves the data to default GPU when deterministic flag is on HOT 1
- Retrieval metrics are misleading - here's how it should be done instead HOT 6
- Multiple threshold for Multilabel metrics HOT 1
- Cannot import the "clustering" module HOT 5
- Potentially Misleading Error Message for multiclass_precision HOT 2
- Accuracy incurs a GPU/CPU sync point HOT 3
- Support equivalent of torcheval's `merge_state` HOT 1
- Multilabel metrics based on threshold should accept a threshold per label
- Recall and Accuracy produce different values when used with logits HOT 1
- Cloning `MultitaskWrapper` with postfix parameters results in incorrect dictionary HOT 5
- f-string should be used in exception HOT 2
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 torchmetrics.