Bitcoin Price Prediction Using Machine Learning And Python

Bitcoin Price Prediction Using Machine Learning And Python

Please Subscribe !

⭐Please Subscribe !⭐

⭐Support the channel and/or get the code by becoming a supporter on Patreon:

⭐Websites:

⭐Helpful Programming Books
► Python (Hands-Machine-Learning-Scikit-Learn-TensorFlow):

► Learning Python:

►Head First Python:

► C-Programming :

► Head First Java:

#Bitcoin #MachineLearning #BitcoinPrediction

[stanleydubois]

It makes sense, BTC and crypto is off helping to regulate, rather that pretend it won't ever happen. Stick with BTC/ETH, Most people don't like change but after the change is made they grow used to it and it becomes a non issue usually because their fears never materialize. And benefits they were unaware with before turn out to be far more beneficial. The projects that initiated the process of regulation have not been ruined, they got involved in setting guidelines and helping the regulators understand the crypto space. I’d get involved more knowing that I have made over 17.5` btc from day-trade with Paulson Dmitriev in few weeks.

[RishabhGupta]

this isn't a classification problem. accuracy metric @ 18:15 doesn't make any sense. For regression we need a metric like MSE to understand how far off our predictions were.

Apart from this as someone else pointed out, when you split the dataset into train and test you are shuffling the data, and this renders the time series data absolutely useless

On top of this, I don't understand why this video has so many likes, the model makes no sense. We are essentially trying to fit a curve to a 'prices vs prices' graph, and this model doesn't capture any notion of time in it.

[story_teller_]

The video was really good.

But some clarification: For the last 30 days prediction we had done for future 30 days prediction_days_array at 21:18, and we dont have the actual price data to compare from the dataset. And at last you had made comparison with the wrong data.

[pattty]

right for anyone that's trying to get it to work, you need to add this to the slitting of data function.
x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.2, shuffle=False) <---- suffle=False
will make sure the data isn't shuffled around because the order is important to the prediction

[xlm-oneringtobringthemall]

Thank you - well done.
Where did you get the data from?

[BB-kofh]

Don't you have to shift the predicted values down by 30 at the end, in order for todays value to correspond to prior predicted 30 day value.
Currently, it seems that your comparing todays value with the predicted 30day value at the end.
Your vids are awesome.
Thanks

[jurijmonzon]

How do I run the prediction without test data? I mean I don't have those values yet.

[IranGenius]

This code does not work. because in splitting the test and train data it shuffles them and learning is based on shuffled data that is not true for time series data like stock prices. So the accuracy that is calculating in the code is not true.

[naderbouchnag]

Great work :) please is there a video that explains how to predict bitcoin price using the twitter sentiment as a variable ( Keras)?

[salmanhaider]

Admin please also provide the datasets also in each of your video it will be very helpful

[gtqa]

It is a bit weird to give a prediction using only the predicted attribute for the model...Isn't it?

[kenleung]

could you tell us how to download bitcoin csv

[whaThirukkovalluruSripriya]

I am getting NAN from 1st row after df['predction] Line ...plz help me

[BekBrace]

Would this have the same results on visual studio code?

[diegoeduardo]

This is video is perfect! do you have any video that do the predict stock with sentiment ? I would like know how this will work.

[anthonyamendolare]

Your videos are awesome. I am gonna watch all of them haha. Thank you!

[carloeugenioleviaguirrecha]

can someone explain me why you have to shift the data?

[abdulghafarmohammadi]

Does someone have the excel file pls???

[carloeugenioleviaguirrecha]

why you have to drop many times the prediction_days?

[BlokeGaming]

This prediction was way off.. Your other one, for one day predictions in Apple stock markets worked better on BTC and ETH than this did on either.. But I couldn't figure out how to feed that prediction back into the code and keep predicting based on prior predictions. (I'm not great with python or these libraries)
I tested both BTC and ETH, I got nothing near the right numbers and I tried the suggested (shuffle = False) when splitting too, less random but still nothing near close.. Sad :( I really want to see Log periodic power law prediction in any code running or something that uses Fibonacci sequencing to predict..