ModelVérifiéExterne

Pep2Tcr GRATCR

This repository contains relevant information of our research article 'Leveraging contrastive learning for conditional and robust TCR sequence generation'.

Sourcehuggingface.co/keiwoo/Pep2Tcr-GRATCR/tree/43e325036b944641df718fc7dac9dddee9f4a375

TypeOtherModèle de basekeiwoo/RoBERTpepVersionRevision 43e3250Éditeur@keiwooCClasse

Source du modèle

Type: Other
Taille du fichier: 682 MB
Modèle de base: keiwoo/RoBERTpep
Version: vRevision 43e3250
Mis à jour: 4 oct. 2026
Éditeur: @keiwoo
Téléchargements: 10
Tâches: Génération de texte
Source: Hugging Face

Description de la source

This repository contains relevant information of our research article 'Leveraging contrastive learning for conditional and robust TCR sequence generation'.

Transformers

You can use Pep2Tcr trained on GRATCR dataset with Transformers. To get started, install the necessary dependencies to setup your environment:

pip install -U transformers torch

2. Run the generation

from transformers import EncoderDecoderModel, AutoTokenizer  
import torch

torch.cuda.manual_seed_all(42)
device = torch.device('cuda:1')

model_name = 'keiwoo/Pep2Tcr-GRATCR'
model = EncoderDecoderModel.from_pretrained(
    model_name, 
    dtype=torch.float16,    # If your device does not support `BF16` precision mode
                            # (GPU based on NVIDIA's Ampere and its subsequent architectures), you should comment this line.
    ).to(device)
model.eval()
tokenizer = AutoTokenizer.from_pretrained(model_name, do_lower_case=False)

sample_kwargs = {
    'do_sample': True,
    'top_k': 9,
    'top_p': 0.93,
    'temperature': 0.9,
    'num_return_sequences': 16,
    'repetition_penalty': 1.1,
    'no_repeat_ngram_size': 3,
    'max_new_tokens': 24,
}
beam_kwargs = {
    'num_beams': 16,
    'num_return_sequences': 16,
    'repetition_penalty': 1.1,
    'no_repeat_ngram_size': 3,
    'max_new_tokens': 24,
}
inputs = "AAGIGILTV"
tokenized = tokenizer(" ".join(inputs), return_tensors="pt", ).to(device)
outputs = model.generate(**tokenized, **sample_kwargs) # If you want different sequences every generation, uncomment this line. 
# outputs = model.generate(**tokenized, **beam_kwargs) # If you want same sequences every generation (most possible sequence the model think), uncomment this line.

tcr = tokenizer.batch_decode(outputs, skip_special_tokens=True)
tcr = [i.replace(' ', '') for i in tcr]
print(tcr)
# ['CASSQEGLAGGTDTQYF', 'CASSQEGLAGGYNEQFF', 'CASSQDPGLAGGYNEQFF', ...]

3. Reproduce the model

Please refer to https://github.com/keiwoo/Pep2Tcr

Sources

1 source

Hugging FacePrincipalVersion 43e325036b944641df718fc7dac9dddee9f4a375

huggingface.co/keiwoo/Pep2Tcr-GRATCR/tree/43e325036b944641df718fc7dac9dddee9f4a375

Vérifié 4 oct.

Artefacts du modèle

1 artefact

model.safetensors

safetensors · 682 MB · SHA-256 717bf1ec1232…efb6 · Hugging Face

Télécharger

Extraits de sources

3 extraits

Readme11 juil.huggingface.co/keiwoo/Pep2Tcr-GRATCR

--- license: apache-2.0 base_model: - keiwoo/RoBERTpep - keiwoo/RoBERTtcr pipeline_tag: text-generation --- This repository contains relevant information of our research article *'Leveraging contrastive learning for conditional and robust TCR sequence generation'*. #### Transformers You can use Pep2Tcr trained on [GRATCR dataset](https://doi.org/10.1109/JBHI.2024.3514089) with Transformers. To get started, install the necessary dependencies to setup your environment: ``` pip install -U transformers torch ``` ### 2. Run the generation ``` from transformers import EncoderDecoderModel, AutoTokenizer import torch torch.cuda.manual_seed_all(42) device = torch.device('cuda:1') model_name = 'keiwoo/Pep2Tcr-GRATCR' model = EncoderDecoderModel.from_pretrained( model_name, dtype=torch.float16, # If your device does not support `BF16` precision mode # (GPU based on NVIDIA's Ampere and its subsequent architectures), you should comment this line. ).to(device) model.eval() tokenizer = AutoTokenizer.from_pretrained(model_name, do_lower_case=False) sample_kwargs = { 'do_sample': True, 'top_k': 9, 'top_p': 0.93, 'temperature': 0.9, 'num_return_sequences': 16, 'repetition_penalty': 1.1, 'no_repeat_ngram_size': 3, 'max_new_tokens': 24, } beam_kwargs = { 'num_beams': 16, 'num_return_sequences': 16, 'repetition_penalty': 1.1, 'no_repeat_ngram_size': 3, 'max_new_tokens': 24, } inputs = "AAGIGILTV" tokenized = tokenizer(" ".join(inputs), return_tensors="pt", ).to(device) outputs = model.generate(**tokenized, **sample_kwargs) # If you want different sequences every generation, uncomment this line. # outputs = model.generate(**tokenized, **beam_kwargs) # If you want same sequences every generation (most possible sequence the model think), uncomment this line. tcr = tokenizer.batch_decode(outputs, skip_special_tokens=True) tcr = [i.replace(' ', '') for i in tcr] print(tcr) # ['CASSQEGLAGGTDTQYF', 'CASSQEGLAGGYNEQFF', 'CASSQDPGLAGGYNEQFF', ...] ``` ### 3. Reproduce the model Please refer to [https://github.com/keiwoo/Pep2Tcr](https://github.com/keiwoo/Pep2Tcr)

Note4 oct.huggingface.co/keiwoo/Pep2Tcr-GRATCR

Source context: 10 downloads · 0 likes · Pipeline text-generation · Repo keiwoo/Pep2Tcr-GRATCR

Note18 juil.huggingface.co/keiwoo/Pep2Tcr-GRATCR

Source context: 105 downloads · 0 likes · Pipeline text-generation · Repo keiwoo/Pep2Tcr-GRATCR