pytorch RuntimeError: Expected tensor for argument #1 'indices' to have scalar type Long;

最新推荐文章于 2024-01-08 19:40:40 发布
原创 最新推荐文章于 2024-01-08 19:40:40 发布 · 1.4w 阅读 · 1 ·
CC 4.0 BY-SA版权
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
文章标签：

#pytorch

博客指出在pytorch中，将np.int64的数传给embedding时会报错，报错内容为期望参数为Long类型，实际为CPUType。解决办法是对该数进行数值转换，转为python内置int类型，可使用np.asscalar(np.int16(a))实现。

摘要生成于 C知道 ，由 DeepSeek-R1 满血版支持， 前往体验 >

错误:将一个np.int64的数传给embedding
报错内容
RuntimeError: Expected tensor for argument #1 ‘indices’ to have scalar type Long; but got CPUType instead (while checking arguments for embedding)

解决：对这个数进行数值转换，转为python内置int类型即可，np.asscalar(np.int16(a))即可。

Expected tensor for argument #1 ‘indices’ to have scalar type Long；but got torch.IntTensor instead
qq_35668469的博客
08-18 1万+
Expected tensor for argument #1 ‘indices’ to have scalar type Long;but got torch.IntTensor instead训练的批量样本数据输入值需要是long值的Tensor数据，而不是int值的Tensor数据.
Expected tensor for argument #1 ‘indices‘ to have scalar type Long； but got torch.cuda.FloatTensor i
weixin_37817539的博客
03-08 5705
Expected tensor for argument #1 'indices' to have scalar type Long; but got torch.cuda.FloatTensor instead (while checking arguments for embedding) 我遇到这个问题时是由于我nn.Embedding()的输入并非torch.int64(torch.LongTensor)，解决方法是使用torch.LongTensor()或者是dtype=int64，这是证明
transformer 4 RuntimeError: Expected tensor for argument #1 ‘indices‘ to have scalar type Long
叶子的博客
01-16 4937
在使用transformer 4.0时，报错误提示RuntimeError: Expected tensor for argument #1 'indices' to have scalar type Long;该问题主要时由于tensor的类型导致的，解决方法是在相应报错行的前一行对数据类型进行转换。假设输入数据为x，那么增加行为“x = torch.tensor(x).to(torch.int64)”。
【报错解决】RuntimeError: Expected tensor for argument #1 ‘indices‘ to have one of the following scalar
小白冲鸭的博客
01-08 4087
直接换一个函数生成tensor，就是不再使用Tensor()函数了，而是使用torch.tensor()函数，也就不会报错了。在Tensor()函数后面加.long()转换成long类型，就不会再报错了。
RuntimeError: Expected tensor for argument #1 ‘indices‘ to have scalar type Long； but got CUDAType
anshiquanshu的专栏
02-22 2762
错误如下：PyTorch查询embedding的时候，报错了 ~/anaconda3/lib/python3.6/site-packages/torch/nn/modules/module.py in __call__(self, *input, **kwargs) 491 result = self._slow_forward(*input, **kwargs) 492 else: --> 493 result =
RuntimeError: Expected tensor for argument #1 'indices' to have scalar type Long; but got CUDAType
月夜长影
10-15 1万+
错误如下：PyTorch查询embedding的时候，报错了 ~/anaconda3/lib/python3.6/site-packages/torch/nn/modules/module.py in __call__(self, *input, **kwargs) 491 result = self._slow_forward(*input, **kwargs) ...
RuntimeError: Expected tensor for argument #1 'indices' to have one of the following scalar types: Long, Int; but got torch.cuda.FloatTensor instead (while checking arguments for embedding)
03-29
- “Expected tensor for argument #1 'indices' to have scalar type Long; but got torch.cuda.IntTensor instead (while checking arguments for embedding)”[^1]。 - “Expected tensor for argument #1 '...
from transformers import AutoTokenizer, AutoModelForVision2Seq import torch # 加载预训练模型和 tokenizer model_path = "E:\\Qwen\\qwen2.5vl" # 转义路径中的反斜杠 tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True) model = AutoModelForVision2Seq.from_pretrained(model_path, trust_remote_code=True) # 设置模型为评估模式 model.eval() # 假设图像输入是一个 tensor，形状取决于模型的具体要求 image_input = torch.randn(1, 3, 224, 224) # 示例输入图像，batch_size=1, channel=3, height=224, width=224 # 文本输入部分 text_inputs = tokenizer("This is a sample text input", return_tensors="pt", padding=True, truncation=True) input_ids = text_inputs["input_ids"] attention_mask = text_inputs["attention_mask"] import torch.onnx # 定义输出文件名 onnx_model_path = "qwen2_5_vl.onnx" # 导出模型 dynamic_axes = { 'input_ids': {0: 'batch_size'}, # 动态维度 'attention_mask': {0: 'batch_size'}, 'output': {0: 'batch_size'} } try:torch.onnx.export( model, (image_input.float(), input_ids.long(), attention_mask), onnx_model_path, export_params=True, # 存储训练参数 opset_version=13, # ONNX 算子集版本 do_constant_folding=True, # 优化常量 input_names=['image_input', 'input_ids', 'attention_mask'], # 输入名称 output_names=['output'], # 输出名称 dynamic_axes={ 'image_input': {0: 'batch_size'}, 'input_ids': {0: 'batch_size'}, 'attention_mask': {0: 'batch_size'}, 'output': {0: 'batch_size'} } ) except RuntimeError as e: print(f"错误发生在: {e}") # 注意查看错误栈中提到的具体操作名称 print(f"Model has been converted to ONNX format and saved at {onnx_model_path}") 错误发生在: Expected tensor for argument #1 'indices' to have one of the following scalar types: Long, Int; but got torch.FloatTensor instead (while checking arguments for embedding) Model has been converted to ONNX format and saved at qwen2_5_vl.onnx
最新发布
07-02
错误信息 `Expected tensor for argument #1 'indices' to have scalar types: Long, Int; but got torch.FloatTensor instead` 表明： 1. **嵌入层(embedding layer)输入类型不匹配**：PyTorch的嵌入层要求输入是...
Expected tensor for argument #1 'indices' to have scalar type Long; but got CUDAFloatTensor instead
weixin_33671935的博客
11-10 2949
运行代码时出现此问题，参数类型的错误 传的参数应该是torch.longtensor类型 使用a=torch.LongTensor()定义即可 转载于:https://www.cnblogs.com/yangruicvpr/p/9939497.html
Expected tensor for argument #1 ‘indices‘ to have scalar type Long；but got torch.IntTensor instead
あずにゃん梓喵的博客
07-27 2818
日萌社 人工智能AI：Keras PyTorch MXNet TensorFlow PaddlePaddle 深度学习实战（不定时更新） 报错：RuntimeError: Expected tensor for argument #1 'indices' to have scalar type Long; but got torch.Int...
宏基因组学和未培养微生物
01-28
宏基因组学和未培养微生物，陈金华，贺闽，自然界中约有99％的微生物在当前的条件下不可培养，这使微生物资源的开发利用受到了限制。宏基因组技术不通过微生物纯培养而直接�
RuntimeError: Expected tensor for argument #1 ‘indices‘ to have scalar type Long； but got CUDAFloatT
周迪新的博客
04-27 960
使用torchsummary 中的 summary函数时，模型中有Embedding层的时候，会报错。原因是Embedding 需要整型的Tensor。 这么使用即可： from torchsummary import summary summary(model, (1, 100, 8), dtypes=[torch.LongTensor], device='cpu') pip install 的torchsummary 不行，没有dtypes这个参数。 所以你下载这里提供的 torchsummary.p
RuntimeError: Expected tensor for argument #1 ‘indices‘ to have scalar type Long； but got torch.IntT
binqiang2wang
07-13 4883
RuntimeError: Expected tensor for argument #1 ‘indices’ to have scalar type Long; but got torch.IntTensor instead (while checking arguments for embedding) RuntimeError: Expected tensor for argument #1 'indices' to have scalar type Long; but got torch.IntTe
RuntimeError: Expected tensor for argument #1 ‘input‘ to have the same device as tensor for argument
qq_46382359的博客
08-10 1651
RuntimeError: Expected tensor for argument #1 ‘input’ to have the same device as tensor for argument #2 ‘weight’; but device 3 does not equal 0 (while checking arguments for cudnn_convolution) 这个问题是提示使用多个GPU出错。 但是我只使用了一块GPU呀？ device = torch.device(“cuda:3”
【Bug解决】RuntimeError: Expected tensor for argument #1 ‘input‘ to have the same device as tensor for..
学习 & 分享 ~
05-31 4086
单机多卡跑网络模型遇到以下错误： 报错内容：从报错内容可以看出， 和 不在同一个 GPU 上，输入数据在 device 0 上（cuda:0），模型权重在 device 2 上（cuda:2）。首先定位到报错的代码行，打印一下网络输入数据（）和网络权重（）分别所在的设备： 打印输出结果，数据在 cuda 0，而网络权重在 cuda 2，和报错信息提示是一致的。关于该错误出现的原因，很可能是模型定义的问题，没有继承 nn.Module 方法，因为 DataParallel 只能对 nn.Module 及继
torch在处理输入数据时出现错误：Expected tensor for argument #1 'indices' to have scalar type Long; but got CUDAFl
weixin_44007959的博客
12-19 7974
在jupyter notebook中运行代码时出现此问题，参数类型的错误 传的参数应该是torch.longtensor类型 原来是直接的input_ids = torch.tensor(…) 使用input_ids = torch.LongTensor() 定义即可 ...
RuntimeError Expected tensor for argument #1 'input' to have the same device as tensor for argument
Jemary_的博客
10-23 7771
RuntimeError: Expected tensor for argument #1 'input' to have the same device as tensor for argument #2 'weight'; but device 1 does not equal 0 (while checking arguments for cudnn_convolution) 记录一下bu...
Expected tensor for argument #1 ‘indices‘ to have one of the following scalar types: Long, Int； but
热门推荐
WGS.
01-20 2万+
RuntimeError: Expected tensor for argument #1 'indices' to have one of the following scalar types: Long, Int; but got torch.FloatTensor instead (while checking arguments for embedding) 解决： 转成long类型才能作为nn.embedding的输入
评论 21
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值