Sentence Similarity
Transformers
Safetensors
PEFT
sentence-transformers
English
lora
reinforcement-learning
domain-adaptation
sentence-embeddings
curriculum-learning
multi-task-learning
rag
information-retrieval
cross-domain
Eval Results (legacy)
Instructions to use EphAsad/DomainEmbedder with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use EphAsad/DomainEmbedder with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("EphAsad/DomainEmbedder", dtype="auto") - PEFT
How to use EphAsad/DomainEmbedder with PEFT:
Task type is invalid.
- sentence-transformers
How to use EphAsad/DomainEmbedder with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("EphAsad/DomainEmbedder") sentences = [ "That is a happy person", "That is a happy dog", "That is a very happy person", "Today is a sunny day" ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [4, 4] - Notebooks
- Google Colab
- Kaggle
File size: 1,103 Bytes
485908e | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 | {
"version": "DomainEmbedder-v2.6",
"step": 4000,
"base_model": "FireDevourerEmbedder-RL-v3.6",
"avg_reward": 1.5269566774368286,
"accuracy": 92.5,
"all_scores": {
"avg_reward": 1.5269566774368286,
"accuracy": 92.5
},
"timestamp": "2026-02-10T02:01:18.178319",
"method": "TRUE LoRA (PEFT) + SUPERVISED RL + CURRICULUM",
"config": {
"lora_rank": 16,
"lora_alpha": 32,
"lora_target_modules": [
"query",
"value"
],
"rl_algorithm": "Supervised A2C Policy Gradient",
"rl_total_steps": 5000,
"rl_gamma": 0.99,
"rl_entropy_coef": 0.1,
"correctness_bonus": 1.0,
"correctness_penalty": 0.5,
"curriculum_learning": true,
"domains": [
"medical",
"legal",
"code",
"finance",
"scientific"
]
},
"files": {
"base_model": "FireDevourerEmbedder-RL-v3.6.pt",
"rl_policy": "rl_policy.pt",
"lora_adapters": {
"medical": "medical_lora/",
"legal": "legal_lora/",
"code": "code_lora/",
"finance": "finance_lora/",
"scientific": "scientific_lora/"
}
}
} |