GPU bench-marking with image classification | Deep Learning Tutorial 17 (Tensorflow2.0, Python)
23:58
Customer churn prediction using ANN | Deep Learning Tutorial 18 (Tensorflow2.0, Keras & Python)
40:41
Tensorboard Introduction | Deep Learning Tutorial 16 (Tensorflow2.0, Keras & Python)
Вставка
- Опубліковано 27 вер 2024
- Often it becomes necessary to see what's going on inside your neural network. Tensorboard is a tool that comes with tensorflow and it allows you to visualize neural network as well as how it trains itself. This tool is very helpful in debugging issues too. We will use the notebook that we created in previous videos for recognizing handwritten digits and visualize accuracy,loss with every epoch using tensorboard. We will also look at the visual graphs of neural network along with some internal computations.
Do you want to learn technology from me? Check codebasics.io/ for my affordable video courses.
🔖 Hashtags 🔖
#tensorboard #tensorboardtutorial #tensorboardgraph #deeplearningtutorial #tensorboardpytorch #pytorchtensorboard
Next Video: • GPU bench-marking with...
Previous video: • Chain Rule | Deep Lear...
Deep learning playlist: • Deep Learning With Ten...
Machine learning playlist : • Machine Learning Tutor...
Code for notebook: github.com/cod...
Website: codebasics.io/
Facebook: / codebasicshub
Twitter: / codebasicshub
Check out our premium machine learning course with 2 Industry projects: codebasics.io/courses/machine-learning-for-data-science-beginners-to-advanced
The problem you are facing at 9:20 is because you did not make a separate log for each run. You ran the code multiple times with the log_dir name being the same. You should append a unique number (I append time) to make the log-name of each run unique. If you don't, then data from every new run is saved on the previous log itself. And it does not overwrite your old data of that log (it's like adding a sentence to an already written paragraph).
At the end of the first run, you were at say, 90% accuracy then you re-run the code (without changing the log-name) and the accuracy of your first epoch goes back to 1%. That's why you see that zig-zag in the accuracy and loss graphs. It appended the 1% info in front of 90% in the log_dir. (not to mention that it saved the x-axis data too making the graph of 1% start from step-1 on the x-axs)
if you are facing error in tensorboard dashboard, try it in powershell prompt , move to same directory add that same commad, it worked for me !
unexpected very detailed graph is displayed on my tensorboard
I'm Bing watching ur deep learning videos ❤❤
It was an OK video, however and for some reason, I cannot get Tensorboard to work from Git Bash, only using Jupyter Notebooks. A bit more focus there would have been nicer.
Thank you for declaring hyperparameters and parameters. I always confused them
Waiting sir.. For your tutorial
Very soon!
Exciting 🤩
ERROR: Could not find `tensorboard`. Please ensure that your PATH
contains an executable `tensorboard` program, or explicitly specify
the path to a TensorBoard binary by setting the `TENSORBOARD_BINARY`
environment variable.
This error coming when i run the command
%load_ext tensorboard
%tensorboard --logdir logs/fit
How to fix that?
Thank u so much sir.. Brilliant representation...
thanks
This is really useful, thank you sir.
Glad it was helpful!
Sir is there any way we can get information about dataset used from TensorBoard?
Thank u sir.. I am doing project on CNN.. Using tensorflow
All the best
Thank you! BTW, did you also do a video on time-series forecasting?
Sir ,you should see my graph...it was like noodles...so I give it up..😶
Tq sir i waiting for u r videos
How the histogram is will be giving us more information for debugging..
thanks master
I am getting an error here please. When I try running the callback, in the end I get an error that TypeError: can only concatenate str (not "list") to str. I don't know why this is happening though I run the same code.
aren't the actual values the unsmoothed (faded) graph? So you should always set smoothing to 0
This is going to be really helpful to understand what goes on under the hood
yup
Nice tutorial. Is it possible to somehow view these charts during execution? For very long executions, it will be helpful to view the progress while the training is underway. Is it OK to use 'Sigmoid' on a Dense layer with 10 neurons? My understanding is that Softmax is more appropriate for multi-class classification with more than 2 classes.
Softmax works better for multiclassification and its advisable
bash: tensorboard: command not found i got this error how to fix this one
Same problem
@@muhammadusama1000 activate ur conda env
you can also use-: py -m tensorboard
Try running your neural network py file on the command prompt instead on Jupyter Notebook, and after it has run, enter the command: python -m tensorboard.py --logdir logs/
@@alokjha6920 add path of anaconda/scripts to environment variables
Hello sir can you make a video on how to add meta data to tensor flow models to use it in android as tflite
yes
codebasics Thanks
I am getting one error while launching tensorboard saying No dashboards are active for the current data set.
Probable causes:
You haven’t written any data to your event files.
TensorBoard can’t find your event files.
Same error , sir can you please help us
Please upload a video on Adam optimizer...
yes that is in my plans
I am unable to access, getting error ERROR: Failed to start
'C:\\Users\\Vinay.Chauhan\\Anaconda3\\Lib\\site-packages\\tensorboard'
(set by the `TENSORBOARD_BINARY` environment variable): [WinError 5]
Access is denied. Please help me!!
open anoconda with administrative
Run JupyterNotebook (anaconda) as administrator
If it still doesn't work:
Delete the cell:
"%load_ext tensorboard
%tensorboard --logdir logs/"
Replace with:
"pip uninstall -y tensorboard tb-nightly
pip install -U tb-nightly"
Finally replace the cell by:
"%reload_ext tensorboard
%tensorboard --logdir logs/"
And go to "localhost:6006/"
@@pseudounknow5559 THANKS
How to kill tensorboard port in jupyter notebook
Can we get a job at fang with bachlers
yes
sir i dont have gitbash is it necessary to install
go with google collab
No you can use Cmd and Powershell
SGD is not necessarily better than Adam or vice verse, based on the plots. The reason is that you trained using Adam based on weights of previous SGD runs.
lol
>tensorboard --logdir logs/
2021-04-21 15:39:25.725464: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library 'cudart64_101.dll'; dlerror: cudart64_101.dll not found
2021-04-21 15:39:25.876844: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
Please disable cuda and GPU in the configuration because it seems your computer does not have it.. it'll work finely after that
Waiting sir.. For your tutorial
without giving any proper Idea u inserted tensor board.
10:47
Can you please detail a bit on mages in Tesorboard ? How to visialize the worngly predicted images in TB...
Is it normal that i dont have the same values on the jupyter console and on the tensorboard charts ? epoch 1/5 Accuracy on console is 0.8746 and on tensorboard it is 0.922 ...... epoch 5/5 on console it is 0.9832 and on tensorboard it is 0.9834. There are not very different however not the same, is it a problem anf if so why ?
Does someone have the same issue ?
My tensor board gives me the following error:
"No dashboards are active for the current data set.
You haven’t written any data to your event files.
TensorBoard can’t find your event files."
did you fix the error?
@@imane-hx1kp yes I did. The problem was the version of tensorflow. I downgraded it and the problem resolved. Many thanks for asking. I am still following your lovely tutorials.
as i watched the nex video on ANN i realised that i have intel iris xe gpu and not nividia and the time it takes to train the model with only cpu took me so much time, do you have any alternatives?@@shafagh_projects