X2Paddle

X2Paddle copied to clipboard

.pb 转paddle模型失败，附.pb模型

5

 .pb模型：链接：https://pan.baidu.com/s/1oSJ0ps-aFiVyDLC0rul6jA 提取码：asdf 模型转换命令： x2paddle --framework=tensorflow --model=frozen_model.pb --save_dir=pd_model 问题： 1）形状对不上 2）_BatchNorm_Const_2 receives a shape [0], but the expected shape is [64]. 应用场景：类似pix2pix，用于图像生成，风格迁移 paddlepaddle 2.2.1 x2paddle 1.3.5 谢谢

zuixiaosanlang

onnx模型转paddle模型报错

4

模型来源： pytorch1.6 模型说明： JDE多目标跟踪模型，其相应repo为 https://github.com/lyk125/Towards-Realtime-MOT 模型文件： https://pan.baidu.com/s/1Ifgn0Y_JZE65_qSrQM2l-Q 转换过程出错提示如下： Traceback (most recent call last): File "F:/pythonworkplace/jdeTransform/ts1.py", line 38, in input_examples=[torch.tensor(input_data)]) File "D:\Anaconda3\envs\pytorch2paddle\lib\site-packages\x2paddle-1.0.1-py3.6.egg\x2paddle\convert.py", line 201, in pytorch2paddle File "D:\Anaconda3\envs\pytorch2paddle\lib\site-packages\x2paddle-1.0.1-py3.6.egg\x2paddle\op_mapper\dygraph\pytorch2paddle\pytorch_op_mapper.py", line 43, in...

yinyimoshushi

ONNX模型转换报错，Couldn't find x2paddle_Split__658_0_p3 in previous layers, the layers should be make by topological sort

2

## 问题描述 从ONNX模型转换到的时候报错 - 错误信息 ```bash AssertionError: Couldn't find x2paddle_Split__658_0_p3 in previous layers, the layers should be make by topological sort ``` - 转换时用的命令行 ```bash x2paddle --framework=onnx --model=pose_landmarks_saved_model_lite.onnx --save_dir=pose_landmarks_saved_model_lite ```...

YY-GAO

Failed when convert deepspeech pb model to paddle model

5

Hi,i want to use paddle to do inference of deepspeech pb model in the tensorrt-way. Firstly, i get the checkpoint in https://github.com/mozilla/DeepSpeech and convert it to pb model,then i want...

lfgogogo

tensorflow2.x转paddle

1

x2paddle支持tensorflow2.x转paddle嘛

xiaobai-python

# Create A Good Pull Request > 下面的文字请保留在PR说明的最后面，并在提完PR后，根据实际情况勾选确认以下情况 Please check the follow step before merging this pull request - [x] Python code style verification - [x] Review all the code...

wjj19950828

yolov5模型torch转onnx测试通过，onnx转paddle又出了问题

1

模型来源：YoloV5 Pytorch版 模型类型：目标检测 运行环境：windows10 onnx=1.6.0 paddlepaddle-gpu=2.1.0(paddlepaddle cpu 1.8.5也试了) x2paddle==1.3.4 应用场景：Torch转onnx转paddle然后计划用paddle用paddle inference部署 torch转onnx顺利调通，转换的时候，opset_version=11 转换得到的onnx文件已经通过了下面这段onnx官网的测试用例输出The model is valid import onnx model = onnx.load(r"D:\Code\github_project\count_detection_project\torchyolo.onnx") try: onnx.checker.check_model(model) except onnx.checker.ValidationError as e: print('The model is...

Stave604671

根据微信公众号转换openpose的手势关键点报错

1

模型来源： caffe 模型说明： 手势关键点检测 模型文件： https://github.com/CMU-Perceptual-Computing-Lab/openpose/tree/master/models/hand 转换过程出错提示如下： ``` paddle.__version__ = 2.0.0-alpha0 Now translating model from caffe to paddle. Total nodes: 107 Nodes converting ... Converting node 107 ... Nodes converted....

vicwer

pytorch2paddle转ResNet18出错

1

模型来源：pytorch的 ResNet18模型 模型说明：图像分类模型，其repo为https://github.com/ozanciga/self-supervised-histopathology/releases/tag/tenpercent 转换过程出错提示如下： ``` ========= 1 OPs are not supported yet =========== ========== aten::linear ============ Traceback (most recent call last): File "D:\zhuyuang\projects\multi-classification\torchTransformPaddle.py", line 43, in pytorch2paddle(module=model, File "C:\Users\lenovo\.conda\envs\paddle\lib\site-packages\x2paddle\convert.py", line...

Yihouyihou

[菁英计划]torch.utils.data.DataLoader.md中有笔误

2

https://github.com/PaddlePaddle/X2Paddle/blob/develop/docs/pytorch_project_convertor/API_docs/utils/torch.utils.data.DataLoader.md 这里最后一行好像self写错了

zhoucz97