Torch Reviews

A wonderful machine learning library

Pros:

Note that the Lua version is unmaintained - only the Python and R libraries still receive maintenance. That said, Torch is a wonderful library for using tensors and other objects in machine learning. Very convenient tools and functions for implementing algorithms and whatnot.

Cons:

As said, the Lua version is unmaintained. The R version also is missing many features from PyTorch. The Python version, however, is top of the line.

My favorite deeplearning framework

Comments: Used this framework for machine learning applications mostly image classifications

Pros:

Since the introduction of pytorch framework, i have been using it more extensively compared to other frameworks like tensorflow and the rest. Pytorch has one of the most beautifully designed documentation page , it aslo comes with lots of pretrained convolutional neural networks architectures (CNN) which i use extensively.

Cons:

supports only nividia GPU for faster computations as cuda only supports nividia.

Torch LMS review

Comments: One of the best customer experiences I have had with any vendor. Until they were bought out.

Pros:

Easy to use for both admins and end users.

Cons:

The system isnt as robust as other learning management systems. It’s good for small companies who don’t have a lot of time and smaller audiences.

Best software for browsing and downloading youtube videos

Comments: It is good and fast web browser which can easily be use by any person.Downloading and Uploading are better with Torch.it is amazing using Torch especially youtube

Pros:

Torch is very helpful, I use torch when downloading videos from Youtube.It the best alternative to chrome for browsing. The software is user friendly.Privacy settings are awesome and they are fine organized to give a full secured browsing.

Cons:

when Updating, it consumes queit mobile data. Sometimes the browsing speed gets slow when using with VPN.

Deep Learning made easy with Pytorch

Comments: I enjoyed using pytorch to learn and build deep learning models

Pros:

Pytorch is quite easy to use. There is a large support community.

Cons:

Doesn't integrate well with scikitlearn. Its not made for production mainly for learning.

As Easy as Pytorch

Comments: Pytorch is an awesome machine/deep learning framework for both beginners and researchers. And it is also very easy to work with and understand.

Pros:

1. Pytorch is very easy to use, simple to understand especially for beginners. 2. Pytorch is developer friendly, and easy to debug. 3. Pytorch comes with a lot of pretrained models to work with

Cons:

Pytorch doesn't support any visualization tool such as tensorboard.

Torch- The fast and smart web browser

Comments: A good and fast web browser which can be easily used by any person.Downloads and Uploads are much better with Torch...

Pros:

The browsing speed of the browser is slightly better than other browsers. The interface is very easy to use and downloads and uploads can be done by little amount of time.Privacy settings are admireble and they are well organized to give a fully secured browsing.

Cons:

Updates consumes a lot of mobile data. Sometimes the browsing speed is slow when the user uses a VPN.

Torch is glamorous

Comments: My experience is very convenient and I use it every day in my professional work.

Pros:

Speed, design, easy to use, easy to download. It can also be used for professional use of the Internet.

Cons:

It's hard not to like something when it works very well, but something needs to be done on transparency when using it.

Torch

Comments: Good web browser with high browsing speed. And easy to use.

Pros:

Browsing speed is slightly good. Downloads are much faster than other web browsers. The software is easy to use because of the user friendly environment.

Cons:

Updates reqire lot amount of data. It blocks some websites. No adware protection. Background tasking consumes more memory from the RAM.

Another good free alternative for deep learning

Pros:

It incorporates most of the state-of-the-art research and papers algorithms for deep learning. The new PyTorch solves the main issue of this software (it uses Python instead of LUA)

Cons:

The main cons is having to program in LUA. Nowadays, Matlab and Python are the main programming languages. Having to program in LUA makes it hard to prototype for most of us.

Torch Review

Comments: My overall experience with Torch is not good as it does not meet my needs while implementing AI models.

Pros:

I like that Torch is being used in most of the advanced research and that I can find a full code written in Torch with these papers. Also, I like that Torch is always in update providing a range of deep learning architectures.

Cons:

Torch is a very isolated library that is too hard to use with other libraries and packages. It is hard to follow the deep learning pipeline using it as it needs some expertise during implementation. It is not suitable for beginners as it is very tiring to understand.

Torch- The alternative

Pros:

I find Torch the best alternative to chrome. I also like the "download videos" button. It is very helpful and I always use this browser when downloading videos from the internet.

Cons:

It could appeal a little bit chaotic for the ones used to chrome, but it's a matter of time to get used to it.

We leverage TORCH for analytics engine and tracking

Pros:

I like how you can filter your tracking by region and verticals which helps with reporting out the numbers on a monthly or quarterly basis.

Cons:

I like most of the features although sometimes our tracking gets a little busy with the amount of information that gets put in each code.

A efficient GPU first scientific computation library supports a great number of Machine Learning Alg

Pros:

Torch uses Lua which has great performance. Normal Lua code performance can reach 80% performance of C with transparent JIT optimization. Lua is easy to read and learn. It is concise. Besides, Lua has a direct interface to use C libraries. Since Torch is implemented with C, it is portable to different environments.

Cons:

Although Lua supports Linux, Mac and embedded systems, it needs LuaJIT environment when it runs. This makes Torch less used in industrial production.

Good machine learning framework used for various purposes

Comments: I've found several NLP or ASR based startups solely using PyTorch for their work.

Pros:

For speech related projects, the modules and repositories using PyTorch happen to be really helpful. Overall, PyTorch is one of the good options for image based works, too.

Cons:

Keras or TensorFlow users can sometimes find it as a hindrance to move towards PyTorch. In general, it can be hard to master although easy to get into.

HR program could be a bit more user friendly

Comments: I have only used it a little bit but so far so good. if they could adjust some placement of buttons I think it would be better.

Pros:

I havent used it alot but started at a new company that uses torch for their HR department. Seems to be okay so far.

Cons:

Could be a bit more user friendly. seems to be a bit of random buttons in places I wouldn't normally look.

An easy to use Open Source Machine Learning package

Pros:

It uses a dynamic computational graph (it is pythonic).

Cons:

It becomes complex when trying to integrate scykitlearn (a python library)

Excelente para deep learning

Comments: Tanto mi tesis como posterior empleo (no el actual) fue implementando deep learning con datasets bastante robustos, y necesitábamos una herramienta que nos permitiera realizar toda esta carga de trabajo sin problemas, muchas herramientas que probábamos se bloqueaban en medio del procesamiento y nos hacía perder mucho tiempo, Torch en cambio fue una herramienta muy confiable que merece ser reconocida por su rapidez, facilidad de uso y rendimiento.

Pros:

Cuando usé Torch, hacía uso de una GPU Nvidia ya que para Deep Learning tiene mejor desempeño debido a sus núcleos cuda. Torch tenía un excelente soporte para dicha GPU y jamás hubo algún tipo de problema respecto a incompatibilidades o algo similar.

Cons:

Tal vez que no es muy conocido, es sin duda una herramienta bastante útil y muy completa, no tengo quejas al respecto.