dataframe package does not work

[cigoic]

Hi,

I noticed that there is dataframe package installed. But it does not work properly when calling import. Thanks.

import "github.com/kniren/gota/dataframe"

1:8: could not import github.com/kniren/gota/dataframe (open /lgo/pkg/github.com/kniren/gota/dataframe.a: no such file or directory)

[yunabe]

"github.com/kniren/gota/dataframe" works in my environment.

This usually means "github.com/kniren/gota/dataframe" does not exist in your GOPATH or we can not build "github.com/kniren/gota/dataframe" because it's broken.

Can you confirm you can execute go install github.com/kniren/gota/dataframe in your environment?

[cigoic]

Okay, I got it.

I added it to Dockerfile and be able to use the dataframe package right now. Thanks.

[yunabe]

I added it to Dockerfile and be able to use the dataframe package right now.

What did you add to Dockerfile? It's helpful for me to understand the common pitfalls to improve the documentation of this project.

Thanks,
yunabe

[cigoic]

hi,

I simply add "github.com/kniren/gota/..." after the statement of installing 3'rd party's packages.

RUN go get -u <original statement> github.com/kniren/gota/...
RUN lgo install <original statement> github.com/kniren/gota/...

I encountered some issue while installing pip, therefore, I made some changes as follows:
RUN curl https://bootstrap.pypa.io/get-pip.py | python3 && pip3 install......

I also refered gophernotes' dockerfile, adding data science packages on my own docker image.
You can find the revised Dockfile here:
https://github.com/cigoic/go_public/blob/master/docker/lgo_py3/Dockerfile

Thanks for providing lgo. I just start using it recently.

By the way, I did run official Tensorflow Go API's example via lgo.
But the kernel dies easily. Do you have any ideal about this?

Thanks.
Jim

[yunabe]

Thanks!

I encountered some issue while installing pip, therefore, I made some changes as follows:

I added a workaround for pip v10+ubuntu issue in this commit. It might be helpful to you too.

By the way, I did run official Tensorflow Go API's example via lgo.
But the kernel dies easily. Do you have any ideal about this?

I haven't tried Tensorflow Go API. Can you file another file and describe the details of the problme you saw and your environment?

Thanks,
yunabe