About EvalCrafter Env

[fjscfy]

Hello, I'd like to ask if it's possible to run your code metrics without using Docker, or could you share the requirements.txt for your environment?

[Yaofang-Liu]

Thanks for your suggestion, we're working on it.

[daeunni]

@Yaofang-Liu May I ask are there updates? thanks :)

[Yaofang-Liu]

We tried to do so, but it seems that there are too many packages, which cause many incompatible issues when using Conda or pip for installation. I will upload the requirements.txt right away and take a deep dive into it later.

Feel free to let me know if you have any good suggestions. Many thanks :-)

[Yaofang-Liu]

Now, you could consider using requirements.txt or environment.yml to set up the environment.

[Edwardyan1112]

Now, you could consider using requirements.txt or environment.yml to set up the environment.

I am using requirements.txt and environment.yml to build environment, but I found there are many package version conflict. Is there any way to update the correct requirements?

[Yaofang-Liu]

Now, you could consider using requirements.txt or environment.yml to set up the environment.

I am using requirements.txt and environment.yml to build environment, but I found there are many package version conflict. Is there any way to update the correct requirements?

Many thanks for raising such a good issue. Environment setup with Conda is supported now please check here .

[Edwardyan1112]

Now, you could consider using requirements.txt or environment.yml to set up the environment.

I am using requirements.txt and environment.yml to build environment, but I found there are many package version conflict. Is there any way to update the correct requirements?

Many thanks for raising such a good issue. Environment setup with Conda is supported now please check here .

Thanks for reply. But there are still many conflicts in environment.yml. For example:

The conflict is caused by:
The user requested keras==2.13.1
tensorflow 2.12.0 depends on keras<2.13 and >=2.12.0

[Yaofang-Liu]

Now, you could consider using requirements.txt or environment.yml to set up the environment.

I am using requirements.txt and environment.yml to build environment, but I found there are many package version conflict. Is there any way to update the correct requirements?

Many thanks for raising such a good issue. Environment setup with Conda is supported now please check here .

Thanks for reply. But there are still many conflicts in environment.yml. For example:

The conflict is caused by: The user requested keras==2.13.1 tensorflow 2.12.0 depends on keras<2.13 and >=2.12.0

It should be okay now. Many thanks for raising this issue!

[Edwardyan1112]

Now, you could consider using requirements.txt or environment.yml to set up the environment.

I am using requirements.txt and environment.yml to build environment, but I found there are many package version conflict. Is there any way to update the correct requirements?

Many thanks for raising such a good issue. Environment setup with Conda is supported now please check here .

Thanks for reply. But there are still many conflicts in environment.yml. For example:

The conflict is caused by: The user requested keras==2.13.1 tensorflow 2.12.0 depends on keras<2.13 and >=2.12.0

It should be okay now. Many thanks for raising this issue!

Thanks a lot! There is no problem in environment.yml now.