PyTorch Reviews & Product Details - Page 2

What do you like best about PyTorch?

The best thing I like about PyTorch is that it is very simple and easy to code and provides numerous trained functions and models. And If you are really stuck somewhere the documents will really help you, they are very clear. And it is an open-source library so can be used anywhere we like. Review collected by and hosted on G2.com.

What do you dislike about PyTorch?

Being an open-source library it provides a lot of things still when it comes to production for large-scale models, it is a bit ineffective and sometimes can face a problem during scaling. Review collected by and hosted on G2.com.

What do you like best about PyTorch?

Open source, free, easy to use and optimized framework for deep learning model developments. choice of data types and selection of model architecture is very easy for beginners in the field of AI. A lot of examples and free tutorials are available. Another advantage over other frameworks is PyTorch provide dynamic graphing. Review collected by and hosted on G2.com.

What do you dislike about PyTorch?

scalability issue, also developing and integrating into an application is a little difficult. also, only C++ API is provided with PyTorch. Deploying the developed model on the mobile platform is difficult. Review collected by and hosted on G2.com.

What do you like best about PyTorch?

PyTorch's way of writing a module and the seamless integration of various layers/ architectures makes it versatile not only for Research and development but also for production. Review collected by and hosted on G2.com.

What do you dislike about PyTorch?

There's not much to dislike about the framework. Just a bit more diversified support from the community should help. The community is great nevertheless. A faster compilation as compared to some of its peers can be an essential step up for it. Review collected by and hosted on G2.com.

What do you like best about PyTorch?

Documentation and simplicity. And so much state of the art implementations are in pytorch. Graph manipulation are good and all thinks are intuitive. Review collected by and hosted on G2.com.

What do you dislike about PyTorch?

I work only in research, but I feel pytorch is far from a fast development real world applications. Review collected by and hosted on G2.com.

What do you like best about PyTorch?

What I liked the most about Pytorch Library is using GPU or CPUs, and it distributes the computational task among multiple CPUs, which makes the development faster Review collected by and hosted on G2.com.

What do you dislike about PyTorch?

The least liked part about Pytorch is it doesn't have much support available for developers in the community for the occurred error to be solved. Review collected by and hosted on G2.com.

What do you like best about PyTorch?

I like the convenience in debugging and various implementations are available for using pre-trained networks. Building large, complex deep learning architecture is easier with Pytorch. Review collected by and hosted on G2.com.

What do you dislike about PyTorch?

One thing I can think of as disadvantage having smaller user community. Although it does not have much community support as compared to Tensorflow, it's growing. Cannot think of anything that I dislike. I use Pytorch in my daily work and it has always been my favorite. Review collected by and hosted on G2.com.

What do you like best about PyTorch?

Pytorch provides useful abstractions to the downstream tasks of deep learning model development. It is heavily being used in the deep learning community as it is more pythonic in nature and hence is easy to learn and implement Review collected by and hosted on G2.com.

What do you dislike about PyTorch?

Pytorch is popular but for production setup tensorflow still remains a common choice and has more mature deep learning libraries and strong visualisations. Review collected by and hosted on G2.com.

What do you like best about PyTorch?

1. Its easy of development

2. Its ease of going to deeper levels of reviewing & tuning hyperparameters and vectors

3. Easy debugging

4. Parallellism and very fast Review collected by and hosted on G2.com.

What do you dislike about PyTorch?

1. Less users as it new

2. Lacks visualization like tensorboard Review collected by and hosted on G2.com.

What do you like best about PyTorch?

THere's a great interface and support features. Pytorch allows my organisation to quickly develop and deploy models. Review collected by and hosted on G2.com.

What do you dislike about PyTorch?

I'd like it to be faster to open and startup. Review collected by and hosted on G2.com.

What do you like best about PyTorch?

It is flexible and pythonic. The documentation is very comprehensive. Overall it is very easy to use. Review collected by and hosted on G2.com.

What do you dislike about PyTorch?

It does not have the most mature serving framework, but the Pytorch team is working on beefing up this part of Pytorch. Review collected by and hosted on G2.com.