--- language: - en tags: - information retrieval - embedding model - visual information retrieval metrics: - recall pipeline_tag: feature-extraction license: apache-2.0 --- # Memex: OCR-free Visual Document Embedding Model as Your Personal Librarian The model only takes images as document-side inputs and produce vectors representing document pages. Memex is trained with over 200k query-visual document pairs, including textual document, visual document, arxiv figures, plots, charts, industry documents, textbooks, ebooks, and openly-available PDFs, etc. Its performance is on a par with our ablation text embedding model on text-oriented documents, and an advantages on visually-intensive documents. Our model is capable of: - Help you read a long visually-intensive or text-oriented PDF document and find the pages that answer your question. - Help you build a personal library and retireve book pages from a large collection of books. - It has only 2.8B parameters, and has the potential to run on your PC. - It works like human: read and comprehend with **vision** and remember **multimodal** information in hippocampus. ![Memex Archtechture](images/memex.png) # News - 2024-08-17: 👊 We open-sourced [cleaned version of training codebase](https://github.com/RhapsodyAILab/MiniCPM-V-Embedding-v0-Train) for MiniCPM-Visual-Embedding, which supports **deepspeed zero stage 1,2** and **large batchsize** like `4096` for full-parameter training to turn VLMs into dense retrievers. We also developed methods to filter training datasets and generating queries using unlablled datasets. We supports **multi-nodes, multi-GPUs** high-efficiency **evaluation** on large retrieval datasets. With such efforts, we support up to `20B` VLM contrastive learning with `4096` batch size. We have tested that one can train a VLM dense retriever with only **1 GPU, but with batch size of `4096`**. - 2024-07-14: 🤗 We released **online huggingface demo**! Try our [online demo](https://huggingface.co/spaces/bokesyo/minicpm-visual-embeeding-v0-demo)! - 2024-07-14: 😋 We released a **locally deployable Gradio demo** of `Memex`, take a look at [pipeline_gradio.py](https://huggingface.co/RhapsodyAI/minicpm-visual-embedding-v0/blob/main/pipeline_gradio.py). You can build a demo on your PC now! - 2024-07-13: 💻 We released a **locally deployable command-line based demo** for users to retireve most relavant pages from a given PDF file (could be very long), take a look at [pipeline.py](https://huggingface.co/RhapsodyAI/minicpm-visual-embedding-v0/blob/main/pipeline.py). - 2024-06-27: 🚀 We released our first visual embedding model checkpoint on [huggingface](https://huggingface.co/RhapsodyAI/minicpm-visual-embedding-v0). - 2024-05-08: 🌍 We [open-sourced](https://github.com/RhapsodyAILab/minicpm-visual-embedding-v0) our training code (full-parameter tuning with GradCache and DeepSpeed zero-stage2, supports large batch size across multiple GPUs with zero-stage1) and eval code. # Deploy on your PC **Please make sure you have at least 32GB memory on your PC.** - Apple M1/M2/M3 with 32GB memory. - x86 CPU with 32GB memory. - x86 CPU with 32GB memory + Nvidia GPU with 16GB memory. ### Install dependencies Use pip to install all dependencies: ``` Pillow==10.1.0 timm==0.9.10 torch==2.1.2 torchvision==0.16.2 transformers==4.36.0 sentencepiece==0.1.99 numpy==1.26.0 ``` ### Download model weights and modeling file Use one of the following methods: - Download with git clone. ```bash git lfs install git clone https://huggingface.co/RhapsodyAI/minicpm-visual-embedding-v0 ``` - Download with huggingface-hub. ```bash pip install huggingface-hub huggingface-cli download --resume-download RhapsodyAI/minicpm-visual-embedding-v0 --local-dir minicpm-visual-embedding-v0 --local-dir-use-symlinks False ``` ### Launch demo Install `gradio` first. ```bash pip install gradio ``` Adapt the code in `pipeline_gradio.py` according to your device. - For M1/M2/M3 users, please make sure `model = model.to(device='mps', dtype=torch.float16)` then run `PYTORCH_ENABLE_MPS_FALLBACK=1 python pipeline_gradio.py`. - For x86 CPU users, please remove `model = model.to(device)` then run `python pipeline_gradio.py`. - For x86 CPU + Nvidia GPU users, please make sure `model = model.to('cuda')` then run `python pipeline_gradio.py`. - If you encountered an error, please open an issue [here](https://huggingface.co/RhapsodyAI/minicpm-visual-embedding-v0/discussions), we will respond soon. # For research purpose To run the model for research purpose, please refer the following code: ```python from transformers import AutoModel from transformers import AutoTokenizer from PIL import Image import torch device = 'cuda:0' # Load model, be sure to substitute `model_path` by your model path model_path = '/local/path/to/model' tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True) model = AutoModel.from_pretrained(model_path, trust_remote_code=True) model.to(device) # Load image to PIL.Image object image_1 = Image.open('/local/path/to/images/memex.png').convert('RGB') image_2 = Image.open('/local/path/to/images/us2020.png').convert('RGB') image_3 = Image.open('/local/path/to/images/hard_negative.png').convert('RGB') # User query query_instruction = 'Represent this query for retrieving relavant document: ' query = 'Who was elected as president of United States in 2020?' query_full = query_instruction + query # Embed image documents with torch.no_grad(): p_reps = model(text=['', '', ''], image=[image_1, image_2, image_3], tokenizer=tokenizer).reps # Embed text queries with torch.no_grad(): q_reps = model(text=[query_full], image=[None], tokenizer=tokenizer).reps # [B, s, d] # Calculate similarities scores = torch.matmul(q_reps, p_reps.T) print(scores) # tensor([[-0.0112, 0.3316, 0.2376]], device='cuda:0') ``` # Todos - [x] Release huggingface space demo. - [ ] Release the evaluation results. - [ ] Release technical report. # Limitations - This checkpoint is an alpha version, and may not be strong in your tasks, for bad case, please create an issue to let us know, many thanks! - The modeling script `modeling_minicpmv` on `huggingface` is not standard yet, the inference code could be further improved. - The inference speed is low, because vision encoder uses `timm`, which does not yet support `flash-attn`. - The model performs not well on Chinese and other non-English information retrieval tasks. # Citation If you find our work useful, please consider cite us: ```bibtex @misc{RhapsodyEmbedding2024, author = {Rhapsody Group}, title = {Memex: OCR-free Visual Document Embedding Model as Your Personal Librarian}, year = {2024}, howpublished = {\url{https://huggingface.co/RhapsodyAI/minicpm-visual-embedding-v0}}, note = {Accessed: 2024-06-28} } ``` Thanks to MiniCPM-V-2.0 `arxiv.org/abs/2408.01800`, without which there won't be `minicpm-visual-embedding`.