Instructions to use lxzcpro/demo_cls_T5 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use lxzcpro/demo_cls_T5 with PEFT:
from peft import PeftModel from transformers import AutoModelForSequenceClassification base_model = AutoModelForSequenceClassification.from_pretrained("Rostlab/prot_t5_xl_uniref50") model = PeftModel.from_pretrained(base_model, "lxzcpro/demo_cls_T5") - Notebooks
- Google Colab
- Kaggle
Model Card for Model-demo-1.2B
This model is used for a demo classification task
Task type
Protein-level Classification
Model input type
AA Sequence
Label meanings
0: A
1: B
LoRA config
- r: 8
- lora_dropout: 0.0
- lora_alpha: 16
- target_modules: ['SelfAttention.o', 'SelfAttention.k', 'EncDecAttention.o', 'EncDecAttention.v', 'SelfAttention.q', 'EncDecAttention.k', 'EncDecAttention.q', 'SelfAttention.v']
- modules_to_save: ['classifier']
Training config
- optimizer:
- class: AdamW
- betas: (0.9, 0.98)
- weight_decay: 0.01
- learning rate: 0.001
- epoch: 1
- batch size: 4
- precision: 16-mixed
- Downloads last month
- 4
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for lxzcpro/demo_cls_T5
Base model
Rostlab/prot_t5_xl_uniref50