"Missing 'proposalResponses' parameter in transaction request"

I recently started learning fabric and wrote [a small demo](https://github.com/JessKXWL/python-fabric-demo), but there were errors when uploading data using postman. When I upload the following data, the code will not make mistakes.But when the data I upload is 15M, an error will be reported,"Missing 'proposalResponses' parameter in transaction request"。[here is the data](https://github.com/JessKXWL/python-fabric-demo/blob/main/record.json)
```
{
"identifier": "client0_to_rsu4_epoch1",
"data": {
"enc_vectors": {
"conv1.weight": "conv1.weightxxxxxxxxxxx",
"conv1.bias": "conv1.biasxxxxx",
"conv2.weight": "conv2.weightxxx",
"conv2.bias": "conv2.biasxxxxxx",
"fc1.weight": "fc1.weightxxxxx",
"fc1.bias": "fc1.biasxxxx",
"fc2.weight": "fc2.weightxxxxxxx",
"fc2.bias": "fc2.biasxxxx"
},
"dec_aes_key": "chBXLZA03PXYzs43UArhGfGsSn7TACdldFIqYPTO3/FjS73h50IX4RSoALE13fqkROEray0lsLiShIN/48exerGDEaL0BN6O0gGpq55WwNtRUqrsOvGzwisFLJVGK3d5nCyo+Hw8QwjO4fwfQs2ynpeubP1+Zr/TIx09p4NDUHM=",
"loss_slice": 0.10846090443590854
}
}
```
The following is the data displayed in the docker log
```
docker logs dev-peer1.org1.example.com-privacy_cc-1.0
invoke is runningadd
进入add函数
反序列化中
根据Id获取值
添加数据
序列化数据
保存数据
发送事件
```

I wonder if this problem occurs because the uploaded data is too large.

I tried to change PreferredMaxBytes from 512KB to 15MB, but it didn't work.

Is there any way for me to upload such large data?