Update README.md

README.md

@@ -1,10 +1,12 @@
+---
+datasets:
+- HuggingFaceH4/MATH-500
+---
 # EPIC Router Family
 
 This repository hosts the public checkpoints for the EPIC router models. Each
 checkpoint learns to pick the best reasoning configuration (method, aggregator,
-sample count, etc.) given a natural-language math question. The underlying
-training and evaluation code lives in the EPIC GitHub project; these weights are
-ready-to-use drop-in artifacts for that codebase.
+sample count, etc.) given a natural-language math question.
 
 
 ## Available Versions
@@ -26,10 +28,12 @@ Each checkpoint contains:
 
 ## Quickstart (Python)
 
-Install dependencies:
+Install the package locally:
 
 ```bash
-pip install huggingface_hub torch sentence-transformers
+git clone https://github.com/nguyenngocbaocmt02/epic.git
+cd epic_project
+pip install -e .
 ```
 
 Load a checkpoint and route a question:
@@ -37,8 +41,8 @@ Load a checkpoint and route a question:
 ```python
 from huggingface_hub import hf_hub_download
 import torch
-from router.models import RouterScoringModel, MiniLMQuestionEncoder, QuestionProjector
-from data_schemas.reasoning import ReasoningConfig
+from epic.router.models import RouterScoringModel, MiniLMQuestionEncoder, QuestionProjector
+from epic.data_schemas.reasoning import ReasoningConfig
 
 REPO_ID = "baonn/epic"
 VERSION = "1.0"  # or 0.5 / 0.75 / 0.25
@@ -70,11 +74,11 @@ reasoning_configs = [
     ReasoningConfig.deserialize(payload) for payload in checkpoint["methods"]
 ]
 
-question = "How many positive divisors does 3600 have?"
+questions = ["x + 20 = 30 then x = ?", "How many positive divisors does 3600 have?"]
 with torch.no_grad():
-    logits = model([question])
-    method_index = torch.argmax(logits, dim=1).item()
+    logits = model(questions)
+    method_indices = torch.argmax(logits, dim=1).tolist()
 
-selected_config = reasoning_configs[method_index]
-print("Recommended config:", selected_config.serialize(include_samples=True))
+print("Recommended config for question 1:", reasoning_configs[method_indices[0]].serialize(include_samples=True))
+print("Recommended config for question 2:", reasoning_configs[method_indices[1]].serialize(include_samples=True))
 ```