Stock Price Prediction And Forecasting Using Stacked LSTM- Deep Learning

Video quality	The size	Download

Информация о Stock Price Prediction And Forecasting Using Stacked LSTM- Deep Learning

Название	:	Stock Price Prediction And Forecasting Using Stacked LSTM- Deep Learning
Продолжительность	:	36.33
Дата публикации	:
Просмотров	:	423 rb

x and y must have same first dimension, but have shapes (30,) and (10, 1)brhow can i slove this error ?
Comment from : SK Shamim Aktar

🎯 Key Takeaways for quick navigation:brbr00:22 📊 The video discusses stock market prediction using Stacked LSTM (Long Short Term Memory) deep learning modelsbr01:04 ❗ Caution: The presenter advises against using the model for actual investments as it's for demonstration purposes and not guaranteed to make profitsbr02:56 📈 The video tutorial uses Python libraries like Pandas and TensorFlow to collect and analyze stock market databr05:30 📈 Data preprocessing involves collecting historical stock data, dividing it into training and testing datasets, and scaling the data between 0 and 1 using Min-Max scalingbr09:20 🧩 Proper data division in time series analysis involves ensuring that training data is ordered by date to account for temporal dependenciesbr13:11 📉 Data pre-processing for time series forecasting involves creating sequences of data points for input and output, considering the number of time steps to predict future valuesbr21:01 🧠 Stacked LSTM model requires reshaping the input data into a three-dimensional format to train on sequential databr21:57 📊 The video demonstrates building a stacked LSTM model for stock price prediction using Keras and compiling it with Mean Squared Error as the loss functionbr22:38 💡 The video discusses using the Adam optimizer for deep learning modelsbr23:02 🖥️ You don't necessarily need a GPU to execute deep learning code; it can be done on regular laptops with i5 or i7 processorsbr25:23 📊 Mean Squared Error (MSE) can be used to measure the performance of a machine learning model, and it can be calculated for both training and test databr26:43 📈 The video demonstrates plotting predictions for both training and test databr30:08 🤖 When predicting future data, you should use previous data (eg, 100 days) to forecast upcoming valuesbr35:54 🤝 You can improve the model's accuracy by trying different variations, such as using bi-directional LSTM, experimenting with longer input sequences, or predicting further into the futurebrbrMade with HARPA AI
Comment from : Monalisa Burma

Sir can we implement this in Android Studio for multiple companies ?
Comment from : kuttyreddy pramodini

Your initial music almost made me deaf What nonsense!
Comment from : Sanjida Jahan

Can not access Tingoo website
Comment from : Vaishnavi Korgaonkar

Hi , im having trouble running the api key as it is coming invalid decimal literal , could anyone please help
Comment from : Aditya Sinha

will the test set have data leakage problem? as it is feature-scaled before splitting
Comment from : Toby Wong

Sir can you pls share the code for reference
Comment from : Lovish Sharma

I think you should not be using same data for validation and testing, because it has already seen the validation data during training phase
Comment from : Bhupathi Reddy

This is so Good tutorial detailing is so good
Comment from : shiv jaiswal

pandas datareader isnt pre installed and get dta tiingo isnt workung
Comment from : Rishab Nandi

you're not making predictions off of predicitons, you're doing a regression, not forecasting What you've done at the end to forecast the future, you get a flat-line, that's how you should have evaluated the test-data In real-world, you've to make predictions on predictions It's okay, these things are too abstract So, as they say, stock pricing is volatile
Comment from : Sangram Kesari Ray