🚀 世萌验证码识别模型
世萌验证码识别模型可用于图像转文本任务,具体为图像描述生成。该模型基于vit - gpt2微调,使用了60000张图片的训练集,能有效识别验证码。
🚀 快速开始
本模型可在Transformers
库中使用,以下是使用示例代码:
from transformers import VisionEncoderDecoderModel, ViTImageProcessor, AutoTokenizer
import torch
from PIL import Image
model = VisionEncoderDecoderModel.from_pretrained("AIris-Channel/vit-gpt2-verifycode-caption")
feature_extractor = ViTImageProcessor.from_pretrained("AIris-Channel/vit-gpt2-verifycode-caption")
tokenizer = AutoTokenizer.from_pretrained("AIris-Channel/vit-gpt2-verifycode-caption")
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model.to(device)
max_length = 16
num_beams = 4
gen_kwargs = {"max_length": max_length, "num_beams": num_beams}
def predict_step(image_paths):
images = []
for image_path in image_paths:
i_image = Image.open(image_path)
if i_image.mode != "RGB":
i_image = i_image.convert(mode="RGB")
images.append(i_image)
pixel_values = feature_extractor(images=images, return_tensors="pt").pixel_values
pixel_values = pixel_values.to(device)
output_ids = model.generate(pixel_values, **gen_kwargs)
preds = tokenizer.batch_decode(output_ids, skip_special_tokens=True)
preds = [pred.strip() for pred in preds]
return preds
pred=predict_step(['ZZZTVESE.jpg'])
print(pred)
✨ 主要特性
- 图像描述生成:可实现图像到文本的转换,适用于图像描述任务。
- 基于微调:基于vit - gpt2进行微调,能更好地适应验证码识别任务。
- 训练数据丰富:使用了60000张图片的训练集,保证了模型的准确性和泛化能力。
💻 使用示例
基础用法
from transformers import VisionEncoderDecoderModel, ViTImageProcessor, AutoTokenizer
import torch
from PIL import Image
model = VisionEncoderDecoderModel.from_pretrained("AIris-Channel/vit-gpt2-verifycode-caption")
feature_extractor = ViTImageProcessor.from_pretrained("AIris-Channel/vit-gpt2-verifycode-caption")
tokenizer = AutoTokenizer.from_pretrained("AIris-Channel/vit-gpt2-verifycode-caption")
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model.to(device)
max_length = 16
num_beams = 4
gen_kwargs = {"max_length": max_length, "num_beams": num_beams}
def predict_step(image_paths):
images = []
for image_path in image_paths:
i_image = Image.open(image_path)
if i_image.mode != "RGB":
i_image = i_image.convert(mode="RGB")
images.append(i_image)
pixel_values = feature_extractor(images=images, return_tensors="pt").pixel_values
pixel_values = pixel_values.to(device)
output_ids = model.generate(pixel_values, **gen_kwargs)
preds = tokenizer.batch_decode(output_ids, skip_special_tokens=True)
preds = [pred.strip() for pred in preds]
return preds
pred=predict_step(['ZZZTVESE.jpg'])
print(pred)
📄 许可证
本模型使用的是Apache-2.0
许可证。
属性 |
详情 |
模型类型 |
图像到文本模型(基于vit - gpt2微调) |
训练数据 |
60000张图片 |