Comments (4)
SunMarc
commented on September 12, 2024
1
Models are not trained to be used in that dtype. However, if you want to use 8bit model, I advise you to load in fp16 then apply a quantization method like bnb to get the 8-bit or 4-bit model
from transformers.
LysandreJik
commented on September 12, 2024
Thanks for your issue @camenduru! cc @SunMarc is the right person to help you out.
from transformers.
SunMarc
commented on September 12, 2024
Hey @camenduru, it is not possible load a model with a torch_dtype = torch.float8_e4m3fn. The accepted values are torch.float32, torch.bfloat16 and torch.float16. Did you find a tutorial that does this ?
from transformers.
camenduru
commented on September 12, 2024
Hey @SunMarc, thanks for the answer. Why is it not possible?
from transformers.
Related Issues (20)
- Obtaining an Exception "KeyError: 'labels'" while fine-tuning Whisper HOT 14
- Failed to load universal_checkpoint with deepspeed integreation HOT 2
- it works for me when adding the following two codes in the model requests code modeule HOT 1
- Some bug in object detect when train in macbook HOT 7
- Tensor size mismatch when trying to run RT-DETR on multiple gpus HOT 3
- Add SDPA implementation to more models HOT 5
- Different result when using model.generate or translation pipeline (and TypeError when argument 'forced_bos_token_id' is set in pipeline) HOT 1
- Supporting Padding in llava processor HOT 1
- Errors with nn.RMSNorm in DeepSpeed HOT 2
- static cache: RuntimeError: cannot mutate tensors with frozen storage HOT 2
- ONNX export may fail for Hiera due to `math.sqrt` and python type `int` casts HOT 2
- Bug in Llama3.1 8B Instruct Tokenizer HOT 2
- Doc bug: wrong token argument name for Tokenizer.from_pretrained() HOT 4
- Is it possible to infer the model separately through encoder.onnx and decoder.onnx HOT 4
- Replace `torch.tensor()` with `torch.from_numpy()` when processing numpy arrays HOT 4
- TFSwinModel: AttributeError: module 'keras.utils' has no attribute 'unpack_x_y_sample_weight' HOT 1
- No warningmessage when load models with a gamma or beta parameter HOT 4
- MultiTask Classification and label_names on Trainer
- PR #31502 bumps up PyTorch requirement to >=2.4 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 transformers.