You can use it with not only Python but also C++. It indicates that we can implement ML, DL and AI tools in future in faster compiling languages like C++, Java and C#, which will have a moderate learning curve with lesser system strain. Review collected by and hosted on G2.com.

It does not work well when you have to train a very small amount of data. On using small amount of data, you may find it out that PyTorch is not an optimal choice. Review collected by and hosted on G2.com.

The best thing about PyTorch is it is very developer-friendly and It is faster compared to another key frameworks like tensor flow. PyTorch is very helpful in terms of coding. Review collected by and hosted on G2.com.

What I disliked the most about PyTorch is support on error parts is not much available over the internet and the official documentation can be a little better for understanding. Review collected by and hosted on G2.com.

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.

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.

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.

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.

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.

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.

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.

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 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.

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.

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.

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.

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.

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.

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.

1. Less users as it new

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