Tensorboardx: Unable to execute add_graph()

[chinmay5]

While trying to use the feature of aff_graph(), I get the error

`AssertionError Traceback (most recent call last)
in ()
3 model = Meso4()
4 w = SummaryWriter(comment='Meso4')
----> 5 w.add_graph(model, (dummy_input, ),)

~/anaconda3/lib/python3.5/site-packages/tensorboardX/writer.py in add_graph(self, model, input_to_model, verbose, **kwargs)
564 print('add_graph() only supports PyTorch v0.2.')
565 return
--> 566 self.file_writer.add_graph(graph(model, input_to_model, verbose))
567 else:
568 # Caffe2 models do not have the 'forward' method

~/anaconda3/lib/python3.5/site-packages/tensorboardX/pytorch_graph.py in graph(model, args, verbose, omit_useless_nodes)
216 return graph
217
--> 218 assert LooseVersion(torch.__version__) >= LooseVersion("1.0.0")
219
220 with torch.onnx.set_training(model, False):

AssertionError: `

I am using Pytorch version 0.4.1 post 2 while the torchvision version is 1.6 (I upgraded it using pip install torchvisionX --upgrade). Can someone please help me out here

[lanpa]

Please use torchvisionX v1.4 for legacy pytorch.

[chinmay5]

Thank you for the quick response. Just one followup though, when I try to execute the program, I just get an output of Error occurs, No graph saved Is there some way to get a more detailed response in here

[lanpa]

Welcome, so which pytorch/tensorboardX combination are you using now? I didn't do much error handling for add_graph, please post your minimal reproducible code if possible so I can make improvement based on that.

[chinmay5]

Hi. I was able to solve the issue. It was more on my end and I needed to tweak some parameters. I think simply using version 1.4 of tensorboardX was sufficient to solve the issue. Thank you so much

[fengz63]

Hi. I was able to solve the issue. It was more on my end and I needed to tweak some parameters. I think simply using version 1.4 of tensorboardX was sufficient to solve the issue. Thank you so much

Thank you for solving my problem!