Upernet Convnext Tiny

🚀 UPerNet模型卡片

UPerNet是用於圖像分割的模型，藉助segmentation-models-pytorch庫，可方便地加載預訓練模型進行推理。

🚀 快速開始

加載預訓練模型

點擊下面的按鈕在Colab中打開示例：

步驟如下

1. 安裝所需依賴：

pip install -U segmentation_models_pytorch albumentations

2. 運行推理代碼：

import torch
import requests
import numpy as np
import albumentations as A
import segmentation_models_pytorch as smp

from PIL import Image

device = "cuda" if torch.cuda.is_available() else "cpu"

# Load pretrained model and preprocessing function
checkpoint = "smp-hub/upernet-convnext-tiny"
model = smp.from_pretrained(checkpoint).eval().to(device)
preprocessing = A.Compose.from_pretrained(checkpoint)

# Load image
url = "https://huggingface.co/datasets/hf-internal-testing/fixtures_ade20k/resolve/main/ADE_val_00000001.jpg"
image = Image.open(requests.get(url, stream=True).raw)

# Preprocess image
np_image = np.array(image)
normalized_image = preprocessing(image=np_image)["image"]
input_tensor = torch.as_tensor(normalized_image)
input_tensor = input_tensor.permute(2, 0, 1).unsqueeze(0)  # HWC -> BCHW
input_tensor = input_tensor.to(device)

# Perform inference
with torch.no_grad():
    output_mask = model(input_tensor)

# Postprocess mask
mask = mask.argmax(1).cpu().numpy()  # argmax over predicted classes (channels dim)

📦 安裝指南

安裝所需依賴：

pip install -U segmentation_models_pytorch albumentations

💻 使用示例

基礎用法

import torch
import requests
import numpy as np
import albumentations as A
import segmentation_models_pytorch as smp

from PIL import Image

device = "cuda" if torch.cuda.is_available() else "cpu"

# Load pretrained model and preprocessing function
checkpoint = "smp-hub/upernet-convnext-tiny"
model = smp.from_pretrained(checkpoint).eval().to(device)
preprocessing = A.Compose.from_pretrained(checkpoint)

# Load image
url = "https://huggingface.co/datasets/hf-internal-testing/fixtures_ade20k/resolve/main/ADE_val_00000001.jpg"
image = Image.open(requests.get(url, stream=True).raw)

# Preprocess image
np_image = np.array(image)
normalized_image = preprocessing(image=np_image)["image"]
input_tensor = torch.as_tensor(normalized_image)
input_tensor = input_tensor.permute(2, 0, 1).unsqueeze(0)  # HWC -> BCHW
input_tensor = input_tensor.to(device)

# Perform inference
with torch.no_grad():
    output_mask = model(input_tensor)

# Postprocess mask
mask = mask.argmax(1).cpu().numpy()  # argmax over predicted classes (channels dim)

🔧 技術細節

模型初始化參數

model_init_params = {
    "encoder_name": "tu-convnext_tiny.in12k_ft_in1k",
    "encoder_depth": 5,
    "encoder_weights": None,
    "decoder_channels": 512,
    "decoder_use_norm": "batchnorm",
    "in_channels": 3,
    "classes": 150,
    "activation": None,
    "upsampling": 4,
    "aux_params": None
}

📚 詳細文檔

數據集

數據集名稱：ADE20K

更多信息

• 庫地址：https://github.com/qubvel/segmentation_models.pytorch
• 文檔地址：https://smp.readthedocs.io/en/latest/

本模型已使用 PytorchModelHubMixin 推送到模型中心。

📄 許可證

本項目採用MIT許可證。