ViT 模型的 HuggingFace Inference API 问题 - “图像特征提取”错误

问题描述 投票:0回答:1

我的 Vision Transformer (ViT) 模型 rshrott/vit-base-renovation2 的推理 API 遇到问题。

https://huggingface.co/rshrott/vit-base-renovation2

当我尝试使用 API 时,收到以下错误:

{
“error”: "HfApiJson(Deserialize(Error(“unknown variant image-feature-extraction, expected one of audio-classification, audio-to-audio, audio-source-separation, automatic-speech-recognition, feature-extraction, text-classification, token-classification, question-answering, translation, summarization, text-generation, text2text-generation, fill-mask, zero-shot-classification, zero-shot-image-classification, conversational, table-question-answering, image-classification, image-segmentation, image-to-text, text-to-speech, … visual-question-answering, video-classification, document-question-answering, image-to-image, depth-estimation, line: 1, column: 318)))”
}

有趣的是,当我直接在 Python 中使用 Transformers 管道时,模型按预期工作:

from transformers import pipeline
from PIL import Image
import requests

pipe = pipeline(model=“rshrott/vit-base-renovation2”)
url = 'https://example.com/image.jpeg'
image = Image.open(requests.get(url, stream=True).raw)
preds = pipe(image)

此代码运行没有任何问题并返回预期的预测。但是,通过推理 API 使用同一模型时会遇到错误。我怀疑可能存在与预期任务类型相关的配置问题,但我不确定如何解决它。

为什么会出现此错误以及如何修复它?我已经检查了型号卡和配置,但我似乎无法找到“图像特征提取”的来源或原因。

machine-learning huggingface-transformers inference image-classification
1个回答
0
投票

我不知道到底发生了什么,但我遇到了同样的问题,并重新训练了我的模型,但名称不同,并且没有模型名称,就像这样

args = TrainingArguments(
    f"NameOfYourModel",
    remove_unused_columns=False,
    evaluation_strategy = "epoch",
    save_strategy = "epoch",
    learning_rate=5e-5,
    per_device_train_batch_size=batch_size,
    gradient_accumulation_steps=4,
    per_device_eval_batch_size=batch_size,
    num_train_epochs=4,
    warmup_ratio=0.1,
    logging_steps=10,
    load_best_model_at_end=True,
    metric_for_best_model="accuracy",
    push_to_hub=True,
)

并且成功了! 抱歉,没有关于此问题的更详细信息^_^

© www.soinside.com 2019 - 2024. All rights reserved.