Hi, great code. I tried to look on your colab code and run it. I was using GPU in google colab, but instead of that, it was really slow. So first I wanted to know, if the code is good and running good, so I changed number of epochs to 50, just to see a result.
My 1. question: Can I run a code first just on 50 epochs to see if it is good and then run it on 1000 epochs? Or there will be a huge difference?
Then I got stuck on a problem in the part - Generate embeddings and visualize using T-SNE.
Exactly in this line of code - "data = DataReader(config, train=False)" and I got this error: "TypeError: init() got an unexpected keyword argument 'train'". I cant find any argument "train" in DataReader. So my 2. question is: Is there a mistake or do I need to change something?
And my 3. question, this code looks so amazing but is also really big/long. Is it possible to use this code also for any other dataset and I just need to change something?
Thank you for your time! Have a nice day :)
Hi, great code. I tried to look on your colab code and run it. I was using GPU in google colab, but instead of that, it was really slow. So first I wanted to know, if the code is good and running good, so I changed number of epochs to 50, just to see a result.
My 1. question: Can I run a code first just on 50 epochs to see if it is good and then run it on 1000 epochs? Or there will be a huge difference?
Then I got stuck on a problem in the part - Generate embeddings and visualize using T-SNE.
Exactly in this line of code - "data = DataReader(config, train=False)" and I got this error: "TypeError: init() got an unexpected keyword argument 'train'". I cant find any argument "train" in DataReader. So my 2. question is: Is there a mistake or do I need to change something?
And my 3. question, this code looks so amazing but is also really big/long. Is it possible to use this code also for any other dataset and I just need to change something?
Thank you for your time! Have a nice day :)