Repository navigation
any suggestion for onnx exporting? #35
Description
wkkuniquegmail
Issue body actions
No description provided.
Reactions are currently unavailable
Activity
I have export it to onnx, seems no obstacles to do this:
onnxexp glance -m vits_baker_v1.onnx vits!? Namespace(model='vits_baker_v1.onnx', subparser_name='glance', verbose=False, version=False) Exploring on onnx model: vits_baker_v1.onnx Model summary on: vits_baker_v1.onnx ------------------------------------------- ir version: 7 opset_import: 11 producer_name: pytorch doc_string: all ops used: Gather,Constant,Mul,Transpose,Shape,Cast,Range,Unsqueeze,Less,Conv,Concat,Reshape,Div,MatMul,Sub,Add,ConstantOfShape,Slice,Pad,Equal,Where,Softmax,Pow,ReduceMean,Sqrt,Relu,Split,Erf,RandomNormalLike,Not,Greater,And,Expand,ScatterND,NonZero,GatherND,CumSum,Softplus,ReduceSum,GatherElements,Neg,Exp,Ceil,Clip,ReduceMax,If,Tanh,Sigmoid,LeakyRelu,ConvTranspose ------------------------------------------- Summary: Name Shape Input/Output ----------- ------- -------------- tst [1, -1] INPUT tst_lengths [1] INPUT output [-1] OUTPUTReacted by AaronWong and Junseo (Chato) KoReacted by Prajwal Rao, JAEBOK KIM and Aloento@AronWang @Aloento I have updated the exportation code here: https://github.com/jinfagang/lark
I have reconstructed VITS training now it can be more easily training on other backbones.
Reacted by Aloento@jinfagang sad, 404
Will be release after some code refactoring. Meanwhile, join QQ group: 526506088 for more discussion about speech model
I have export it to onnx, seems no obstacles to do this:
onnxexp glance -m vits_baker_v1.onnx vits!? Namespace(model='vits_baker_v1.onnx', subparser_name='glance', verbose=False, version=False) Exploring on onnx model: vits_baker_v1.onnx Model summary on: vits_baker_v1.onnx ------------------------------------------- ir version: 7 opset_import: 11 producer_name: pytorch doc_string: all ops used: Gather,Constant,Mul,Transpose,Shape,Cast,Range,Unsqueeze,Less,Conv,Concat,Reshape,Div,MatMul,Sub,Add,ConstantOfShape,Slice,Pad,Equal,Where,Softmax,Pow,ReduceMean,Sqrt,Relu,Split,Erf,RandomNormalLike,Not,Greater,And,Expand,ScatterND,NonZero,GatherND,CumSum,Softplus,ReduceSum,GatherElements,Neg,Exp,Ceil,Clip,ReduceMax,If,Tanh,Sigmoid,LeakyRelu,ConvTranspose ------------------------------------------- Summary: Name Shape Input/Output ----------- ------- -------------- tst [1, -1] INPUT tst_lengths [1] INPUT output [-1] OUTPUT@jinfagang What about the speed?
update ?
obstacles
what is about multi-speaker instance?
It is ok to convert the pytorch model into onnx. but inference speed on GPU with onnx is far more slower then that with pytorch model.
