Comments (4)
thomwolf
commented on May 7, 2024
1
Ok I've added FP16 support (see updated readme)
from transformers.
thomwolf
commented on May 7, 2024
Yes probably. I am testing fp16 right now. If it works well I will push it to the repo.
from transformers.
howardhsu
commented on May 7, 2024
Thanks for this quick updates.
from transformers.
Ashish-Gupta03
commented on May 7, 2024
I'm not able to work with FP16 for pytorch BERT code. Particularly for BertForSequenceClassification, which I tried and got the issue
Runtime error: Expected scalar type object Half but got scalar type Float for argument #2 target
when I enabled fp16.
Also when using
logits = logits.half() labels = labels.half()
then the epoch time also increased.
from transformers.
Related Issues (20)
- Mamba Models - MambaForMaskedLM and MambaForSequenceClassification HOT 3
- Resize(feature_extractor.size) is a dictionary, not an int or sequence HOT 3
- Timestamps are broken for whisper large with WhisperForConditionalGeneration HOT 1
- Sampling after LogitsProcessor ignores scores if previous token was `eos` HOT 2
- `get_image_features` of CLIPModel returns inconsistent embeddings for single image vs batch images HOT 1
- AttributeError: 'DataParallel' object has no attribute 'device' when trying the Lora-for-sequence-classification-example HOT 1
- Language Modelling Example HOT 7
- (Willing to PR) Make `tokenizer.padding_side` an argument instead of only being a field HOT 1
- Standard Generation and Prompt Lookup Decoding (PLD) generations are not consistent HOT 4
- ValueError: Unrecognized configuration class <class 'transformers.models.idefics2.configuration_idefics2.Idefics2Config'> for this kind of AutoModel: AutoModelForCausalLM. HOT 4
- Slowdown in Training Speed Due to SDPA Mask Fix in Version 4.40.0 HOT 2
- Idefics-2-base model fine-tuning throws indexing error HOT 5
- image-segmentation: AttributeError: read HOT 3
- transformers offline model loading is not working from version 4.40.0 for models without safetensors HOT 1
- Bad request: Can't load config for 'None'. Make sure that: - 'None' is a correct model identifier listed on 'https://huggingface.co/models' - or 'None' is the correct path to a directory containing a config.json file HOT 3
- Trainer with resume_from_checkpoint does not work with multiple Peft Adapters HOT 1
- `generate` method does not work with `use_cache=False` in `Llama-2` model with `model.config._attn_implementation = "eager"` HOT 3
- add `stream` to pipeline parameters HOT 5
- Trainer/accelerate doesn't save model when using FSDP with SHARDED_STATE_DICT
- In attempting to install Apple's new OpenELM, huggingface's transformers appears to want us to ask for permission from FaceBook/Meta? HOT 6
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.