Sentence Similarity
sentence-transformers
Safetensors
Transformers
English
Chinese
qwen2_vl
image-text-to-text
mteb
Qwen2-VL
vidore
custom_code
Eval Results (legacy)
Instructions to use Alibaba-NLP/gme-Qwen2-VL-2B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use Alibaba-NLP/gme-Qwen2-VL-2B-Instruct with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("Alibaba-NLP/gme-Qwen2-VL-2B-Instruct", trust_remote_code=True) 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] - Transformers
How to use Alibaba-NLP/gme-Qwen2-VL-2B-Instruct with Transformers:
# Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("Alibaba-NLP/gme-Qwen2-VL-2B-Instruct", trust_remote_code=True) model = AutoModelForMultimodalLM.from_pretrained("Alibaba-NLP/gme-Qwen2-VL-2B-Instruct", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
training batch size
#3
by yyy111 - opened
hi, thank you for your awesome work. There are some doubts about the training batch size in the paper. Does it refer to per device batch size or total batch size? And how many A100 80G are required for training?
Looking forward to your reply!
Same question. It says using 8 A100 during traininig, but not clarify the batchsize refer to.