issues
search
OFA-Sys
/
Chinese-CLIP
Chinese version of CLIP which achieves Chinese cross-modal retrieval and representation generation.
MIT License
4.5k
stars
464
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
下载模型出错
#113
kydbj
closed
1 year ago
1
把vit-b-16模型转成ONNX格式，报：Exporting the operator 'aten::unflatten' to ONNX opset version 13 is not supported.
#112
JustinZou1
closed
1 year ago
9
同一batch中出现同一类的情况
#111
michelleweii
closed
1 year ago
4
如何继续之前的训练
#110
T0305
closed
1 year ago
5
LMDB格式文件 不区分train test valid
#109
kukuhaza
closed
1 year ago
2
关于图的描述词
#108
duyz1234
opened
1 year ago
6
请问demo中的图文检索的图片库是哪个，我希望能构建一个类似的
#107
6Roy
closed
1 year ago
5
请问demo中的图到图检索是如何实现的？
#106
yourfathermyson
closed
1 year ago
1
请问demo中的图到图检索是如何实现的？
#105
yourfathermyson
closed
1 year ago
1
你好，我在VOC2007无法重现论文中zero-shot的结果，可以帮忙看下吗？谢谢！
#104
pandaupc
opened
1 year ago
4
请问 Chineselip finetune的策略是用的CLIP Models are Few-shot Learners里面few-shot的策略吗
#103
caidou05
opened
1 year ago
1
关于图像描述任务的数据预处理
#102
kukuhaza
closed
1 year ago
5
Adaption to Pytorch2.0
#101
DtYXs
closed
1 year ago
0
关于数据集的构造
#100
Sally6651
closed
1 year ago
5
执行bash命令报错
#99
Sally6651
closed
1 year ago
2
运行报错
#98
Sally6651
closed
1 year ago
18
bash run_scripts/muge_finetune_vit-b-16_rbt-base.sh ${DATAPATH}时出错
#97
duyz1234
closed
1 year ago
21
混合精度训练模型反向传播失效
#96
Hakeyi
opened
1 year ago
3
Chinese-CLIP是如何修改context length的？How does Chinese-CLIP change the context length?
#95
DengXianqi
closed
1 year ago
2
训练脚本运行报错main.py: error: unrecognized arguments: --local-rank=1
#94
6Roy
opened
1 year ago
4
请问两阶段有效性验证的曲线图下面的200B-1000B代表什么意思？
#93
ZYX511
closed
1 year ago
1
关于文本表征的token疑问
#92
ygfrancois
closed
1 year ago
2
modelscope中文图文检索应用
#91
sugarandgugu
closed
1 year ago
0
zeroshot classification RN50 mnist 结果无法复现，RN50其他数据正确，mnist在其他模型上的也跟报的结果一致，但mnist在RN50上只有35.44的结果，希望帮忙看下，谢谢！
#90
pandaupc
closed
1 year ago
3
单个图片和单个文本计算的相似性是一个30+左右的数值，如何判断是否相关呢
#89
yier2333
closed
1 year ago
1
请问如何把图片转换中文描述呢？
#88
minbo366-ai
closed
1 year ago
1
如何将模型部署到移动端
#87
SZN712
opened
1 year ago
1
少样本 finetune
#86
raymond1123
closed
1 year ago
2
使用 fp16 进行 finetune 时，精度不符合预期
#85
Tramac
closed
1 year ago
1
你好，上周与你们沟通后，提交了我做的在zero-shot上的结果，你们最近是否有分析，希望能有进一步交流
#84
pandaupc
closed
1 year ago
5
如何对encoder后图像embedding进行降维？
#83
letiantony
closed
1 year ago
1
如何得到attention map
#82
UknowSth
closed
1 year ago
5
预训练时，是否有使用任何“预”预训练模型？
#81
Tramac
closed
1 year ago
3
Flickr30k-CN text2image结果
#80
pandaupc
closed
1 year ago
1
issue 零样本分类结果 #77 :提交修改zero-shot imageclassification 推理过程后与原始结果对比
#79
pandaupc
closed
1 year ago
0
fix(data): update lmdb pairs and imgs path
#78
Tramac
closed
1 year ago
0
零样本分类结果
#77
pandaupc
closed
1 year ago
5
About COCO-CN dataset, how should I process it?
#76
Suncheng2022
closed
1 year ago
2
No module named 'torch._C._distributed_rpc'; 'torch._C' is not a packageModuleNotFoundError
#75
yourfathermyson
closed
1 year ago
1
报错main.py: error: unrecognized arguments: --local-rank=0
#74
1026468976
closed
1 year ago
8
Support gradient accumulation
#73
DtYXs
closed
1 year ago
0
训练模型时，他是输入一个图片对应多个query，还是随机输入不同的query和图片，如果是前者的话，我保持图片数量不变，增加query应该不会增加太多训练时间
#72
shenghuangxu
closed
1 year ago
1
pillow版本问题
#71
huangyangke
closed
1 year ago
1
为什么OFA模型漏字情况严重
#70
OKC13
closed
1 year ago
1
To make my own lmdb dataset, the base64 image saved should be resized beforehand，right?
#69
zhouwei5113
closed
1 year ago
1
finetune数据量多的时候，生成lmdb的速度会越来越慢
#68
wangpx0906
opened
1 year ago
2
tensorrt容器
#67
huangyangke
opened
1 year ago
1
onnx bert model attention_mask
#66
qiankl
closed
1 year ago
2
finetune后模型非常大，如何对模型进行压缩
#65
wangpx0906
closed
1 year ago
2
导入CLIP模型Finetune，clip_model.encode_text()输出Nan
#64
Qiaojiao-225
closed
1 year ago
9
Previous
Next