Inferensi Phi-3-Vision secara Lokal
February 1, 2026 ยท View on GitHub
Phi-3-vision-128k-instruct memungkinkan Phi-3 tidak hanya memahami bahasa, tetapi juga melihat dunia secara visual. Melalui Phi-3-vision-128k-instruct, kita dapat menyelesaikan berbagai masalah visual, seperti OCR, analisis tabel, pengenalan objek, mendeskripsikan gambar, dan lain-lain. Kita bisa dengan mudah menyelesaikan tugas yang sebelumnya memerlukan banyak data pelatihan. Berikut adalah teknik terkait dan skenario aplikasi yang dikutip oleh Phi-3-vision-128k-instruct
0. Persiapan
Pastikan pustaka Python berikut sudah terpasang sebelum digunakan (disarankan Python 3.10+)
pip install transformers -U
pip install datasets -U
pip install torch -U
Disarankan menggunakan CUDA 11.6+ dan menginstal flatten
pip install flash-attn --no-build-isolation
Buat Notebook baru. Untuk menyelesaikan contoh, disarankan Anda membuat konten berikut terlebih dahulu.
from PIL import Image
import requests
import torch
from transformers import AutoModelForCausalLM
from transformers import AutoProcessor
model_id = "microsoft/Phi-3-vision-128k-instruct"
kwargs = {}
kwargs['torch_dtype'] = torch.bfloat16
processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(model_id, trust_remote_code=True, torch_dtype="auto").cuda()
user_prompt = '<|user|>\n'
assistant_prompt = '<|assistant|>\n'
prompt_suffix = "<|end|>\n"
1. Menganalisis gambar dengan Phi-3-Vision
Kita ingin AI dapat menganalisis isi gambar kita dan memberikan deskripsi yang relevan
prompt = f"{user_prompt}<|image_1|>\nCould you please introduce this stock to me?{prompt_suffix}{assistant_prompt}"
url = "https://g.foolcdn.com/editorial/images/767633/nvidiadatacenterrevenuefy2017tofy2024.png"
image = Image.open(requests.get(url, stream=True).raw)
inputs = processor(prompt, image, return_tensors="pt").to("cuda:0")
generate_ids = model.generate(**inputs,
max_new_tokens=1000,
eos_token_id=processor.tokenizer.eos_token_id,
)
generate_ids = generate_ids[:, inputs['input_ids'].shape[1]:]
response = processor.batch_decode(generate_ids,
skip_special_tokens=True,
clean_up_tokenization_spaces=False)[0]
Kita bisa mendapatkan jawaban yang relevan dengan menjalankan skrip berikut di Notebook
Certainly! Nvidia Corporation is a global leader in advanced computing and artificial intelligence (AI). The company designs and develops graphics processing units (GPUs), which are specialized hardware accelerators used to process and render images and video. Nvidia's GPUs are widely used in professional visualization, data centers, and gaming. The company also provides software and services to enhance the capabilities of its GPUs. Nvidia's innovative technologies have applications in various industries, including automotive, healthcare, and entertainment. The company's stock is publicly traded and can be found on major stock exchanges.
2. OCR dengan Phi-3-Vision
Selain menganalisis gambar, kita juga bisa mengekstrak informasi dari gambar. Ini adalah proses OCR yang sebelumnya memerlukan penulisan kode yang rumit.
prompt = f"{user_prompt}<|image_1|>\nHelp me get the title and author information of this book?{prompt_suffix}{assistant_prompt}"
url = "https://marketplace.canva.com/EAFPHUaBrFc/1/0/1003w/canva-black-and-white-modern-alone-story-book-cover-QHBKwQnsgzs.jpg"
image = Image.open(requests.get(url, stream=True).raw)
inputs = processor(prompt, image, return_tensors="pt").to("cuda:0")
generate_ids = model.generate(**inputs,
max_new_tokens=1000,
eos_token_id=processor.tokenizer.eos_token_id,
)
generate_ids = generate_ids[:, inputs['input_ids'].shape[1]:]
response = processor.batch_decode(generate_ids,
skip_special_tokens=False,
clean_up_tokenization_spaces=False)[0]
Hasilnya adalah
The title of the book is "ALONE" and the author is Morgan Maxwell.
3. Perbandingan beberapa gambar
Phi-3 Vision mendukung perbandingan beberapa gambar. Kita bisa menggunakan model ini untuk menemukan perbedaan antar gambar.
prompt = f"{user_prompt}<|image_1|>\n<|image_2|>\n What is difference in this two images?{prompt_suffix}{assistant_prompt}"
print(f">>> Prompt\n{prompt}")
url = "https://hinhnen.ibongda.net/upload/wallpaper/doi-bong/2012/11/22/arsenal-wallpaper-free.jpg"
image_1 = Image.open(requests.get(url, stream=True).raw)
url = "https://assets-webp.khelnow.com/d7293de2fa93b29528da214253f1d8d0/news/uploads/2021/07/Arsenal-1024x576.jpg.webp"
image_2 = Image.open(requests.get(url, stream=True).raw)
images = [image_1, image_2]
inputs = processor(prompt, images, return_tensors="pt").to("cuda:0")
generate_ids = model.generate(**inputs,
max_new_tokens=1000,
eos_token_id=processor.tokenizer.eos_token_id,
)
generate_ids = generate_ids[:, inputs['input_ids'].shape[1]:]
response = processor.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
Hasilnya adalah
The first image shows a group of soccer players from the Arsenal Football Club posing for a team photo with their trophies, while the second image shows a group of soccer players from the Arsenal Football Club celebrating a victory with a large crowd of fans in the background. The difference between the two images is the context in which the photos were taken, with the first image focusing on the team and their trophies, and the second image capturing a moment of celebration and victory.
Penafian:
Dokumen ini telah diterjemahkan menggunakan layanan terjemahan AI Co-op Translator. Meskipun kami berupaya untuk mencapai akurasi, harap diperhatikan bahwa terjemahan otomatis mungkin mengandung kesalahan atau ketidakakuratan. Dokumen asli dalam bahasa aslinya harus dianggap sebagai sumber yang sahih. Untuk informasi penting, disarankan menggunakan terjemahan profesional oleh manusia. Kami tidak bertanggung jawab atas kesalahpahaman atau penafsiran yang keliru yang timbul dari penggunaan terjemahan ini.