Hello. I am new to data science. I have learnt how to use many methods such as the ensemble method, logistic regression, knn and so on. I have started learning neural network but I can't install Tensorflow on my 32 bits system (since Tensorflow is not compatible with 32 bits). I don't have money to get another pc for now. I tried to use google cloud in hope that I may have access to Tensorflow on it but Google cloud was hopelessly slow. It couldn't even complete the running of my first line of code throughout the day! My first line of code was 'import numpy as np'.
\nMy question is, is there a way round this? Is it possible to somehow install Tensorflow in cloud and access it on my pc? I want to keep learning until I am able to purchase another laptop (64 bits). I will appreciate any suggestions from you guys. Thanks.
\nNote: I read something on how to instal Tensorflow from source on stack exchange but the writer says the effort will not be worth it as one would not be able to do much with it because of space requirements.
\nNote: I don't know much about the cloud yet. All I know is that it can be a place to store large files.
", "messages": [ { "id": 988084, "postDate": "2020-08-27T19:05:57.220Z", "content": "Thanks @Ezeanyi. I just learnt I could use the notebook here at Kaggle since it has all the packages pre-installed. It is also running fast.
", "rawMarkdown": "Thanks @Ezeanyi. I just learnt I could use the notebook here at Kaggle since it has all the packages pre-installed. It is also running fast.", "votes": 1 }, { "id": 988042, "postDate": "2020-08-27T18:10:29.387Z", "content": "I faced the same challenge and overcome it by compiling Tensorflow 1.13.2 from source in a 32 bit Ubuntu partition with single core processor 2.4 GHz, 4 GB ram with 4 GB swap space. I don't know if you are familiar with linux, if you are I will suggest creating a Linux partition as your deep learning workstation.
\nInformation I gathered while looking for workaround to the issue specified its the last version of Tensorflow that can run on 32 bit. You can google to find useful info to help you build Tensorflow from source. Check https://stackoverflow.com/questions/33634525/tensorflow-on-32-bit-linux
Hello. I am new to data science. I have learnt how to use many methods such as the ensemble method, logistic regression, knn and so on. I have started learning neural network but I can't install Tensorflow on my 32 bits system (since Tensorflow is not compatible with 32 bits). I don't have money to get another pc for now. I tried to use google cloud in hope that I may have access to Tensorflow on it but Google cloud was hopelessly slow. It couldn't even complete the running of my first line of code throughout the day! My first line of code was 'import numpy as np'.
\nMy question is, is there a way round this? Is it possible to somehow install Tensorflow in cloud and access it on my pc? I want to keep learning until I am able to purchase another laptop (64 bits). I will appreciate any suggestions from you guys. Thanks.
\nNote: I read something on how to instal Tensorflow from source on stack exchange but the writer says the effort will not be worth it as one would not be able to do much with it because of space requirements.
\nNote: I don't know much about the cloud yet. All I know is that it can be a place to store large files.
", "rawMarkdown": "Hello. I am new to data science. I have learnt how to use many methods such as the ensemble method, logistic regression, knn and so on. I have started learning neural network but I can't install Tensorflow on my 32 bits system (since Tensorflow is not compatible with 32 bits). I don't have money to get another pc for now. I tried to use google cloud in hope that I may have access to Tensorflow on it but Google cloud was hopelessly slow. It couldn't even complete the running of my first line of code throughout the day! My first line of code was 'import numpy as np'.\n\nMy question is, is there a way round this? Is it possible to somehow install Tensorflow in cloud and access it on my pc? I want to keep learning until I am able to purchase another laptop (64 bits). I will appreciate any suggestions from you guys. Thanks. \n\nNote: I read something on how to instal Tensorflow from source on stack exchange but the writer says the effort will not be worth it as one would not be able to do much with it because of space requirements.\n\nNote: I don't know much about the cloud yet. All I know is that it can be a place to store large files." } ], "comments": [ { "id": 988084, "author_name": "samuel adeshina", "author_url": "", "post_date": "2020-08-27T19:05:57.220000", "content": "Thanks @Ezeanyi. I just learnt I could use the notebook here at Kaggle since it has all the packages pre-installed. It is also running fast.
", "votes": 1, "replies": [] }, { "id": 988042, "author_name": "Ezeanyi", "author_url": "", "post_date": "2020-08-27T18:10:29.387000", "content": "I faced the same challenge and overcome it by compiling Tensorflow 1.13.2 from source in a 32 bit Ubuntu partition with single core processor 2.4 GHz, 4 GB ram with 4 GB swap space. I don't know if you are familiar with linux, if you are I will suggest creating a Linux partition as your deep learning workstation.
\nInformation I gathered while looking for workaround to the issue specified its the last version of Tensorflow that can run on 32 bit. You can google to find useful info to help you build Tensorflow from source. Check https://stackoverflow.com/questions/33634525/tensorflow-on-32-bit-linux