Update README.md

README.md

@@ -10,7 +10,7 @@ tags:
 ---
 
 
-# TransHLA2.0-IM
+# TriStageHLA-IM
 
 A minimal Hugging Face-compatible PyTorch model for peptide–HLA binding classification using ESM with optional LoRA and cross-attention. There is no custom predict API; inference follows the training path: tokenize peptide and HLA pseudosequence with the ESM tokenizer, pad or truncate to fixed lengths (default peptide=16, HLA=36), run a forward pass as `logits, features = model(epitope_ids, hla_ids)`, then apply softmax to get the binding probability.
 
@@ -31,10 +31,10 @@ pip install torch transformers peft
 import torch
 import torch.nn.functional as F
 from transformers import AutoModel, AutoTokenizer
-model_id = "SkywalkerLu/TransHLA2.0-IM"
+model_id = "SkywalkerLu/TriStageHLA-IM"
 model = AutoModel.from_pretrained(model_id, trust_remote_code=True).to(device).eval()
 ```
-## How to use TransHLA2.0-IM
+## How to use TriStageHLA-IM
 ```python
 import torch
 import torch.nn.functional as F
@@ -44,7 +44,7 @@ from transformers import AutoModel, AutoTokenizer
 device = "cuda" if torch.cuda.is_available() else "cpu"
 
 # Load model (replace with your model id if different)
-model_id = "SkywalkerLu/TransHLA2.0-IM"
+model_id = "SkywalkerLu/TriStageHLA-IM"
 model = AutoModel.from_pretrained(model_id, trust_remote_code=True).to(device).eval()
 
 # Load tokenizer used in training (ESM2 650M)
@@ -98,7 +98,7 @@ from transformers import AutoModel, AutoTokenizer
 device = "cuda" if torch.cuda.is_available() else "cpu"
 
 # Load model and tokenizer
-model_id = "SkywalkerLu/TransHLA2.0-IM"  # replace with your model id if different
+model_id = "SkywalkerLu/TriStageHLA-IM"  # replace with your model id if different
 model = AutoModel.from_pretrained(model_id, trust_remote_code=True).to(device).eval()
 tok = AutoTokenizer.from_pretrained("facebook/esm2_t33_650M_UR50D")