Comments (2)
Hi @sevinjyolchuyeva , num_return_sequences is not the solution here. It's meant for generation but this problem is not purely generative. To get the probability of the class tokens, make sure each class label is encoded as single id. Then to get get probability you'll need to generate yourself without using the .generate
method and get the final logits. Then let's say we are predicting "true" or "false", to get probability apply softmax only on the "true" and "false" tokens (in your case, 8 tokens).
Hope this helps. I'm not entirely sure how this will work, but worth giving it a try
from exploring-t5.
Hi, @patil-suraj :),
Thank you so much for your comment. I have understood what you said. I will try it.
from exploring-t5.
Related Issues (15)
- Fine-tune Any Models? HOT 1
- difference between decoder_input_ids and lm_labels HOT 2
- load checkpoints and general fine tuning advice HOT 3
- TypeError: function() argument 1 must be code, not str HOT 8
- AttributeError: 'Trainer' object has no attribute 'proc_rank'
- TypeError: cannot unpack non-iterable NoneType object
- T5FineTuner issue "in training_epoch_end avg_train_loss = torch.stack([x["loss"] for x in outputs]).mean() " HOT 4
- t5 training notebook issue HOT 3
- RuntimeError: Input, output and indices must be on the current device HOT 1
- t5-large does not recognize my GPU HOT 1
- Adding ByT5 notebook HOT 2
- fill = pipeline('fill-mask', model='tamil_bert', tokenizer='tamil_bert') HOT 1
- I modify a lot to adapt to the new version of pytorch_lightning
- AttributeError: can't set attribute
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 exploring-t5.