Project 8. Gold Price Prediction using Machine Learning with Python | Machine Learning Projects 

Glad you like them!😇

yeah

This problem will happen because the function corr() cannot calculate the correlation for the columns that are declared as non-numerical data, in this example the Date column has type object. I've solved the problem by following the below step-by-step: # checking which columns contain numerical data print(gold_data.dtypes) # select only the columns with numerical values numerical_data = gold_data.select_dtypes(include=['float64', 'int64']) # calculate the correlation correlation = numerical_data.corr() print(correlation) # constructing a heatmap to understand the correlation plt.figure(figsize=(8,8)) sns.heatmap(correlation, cbar=True, square=True, fmt='.1f', annot=True, annot_kws={'size':8}, cmap='Blues')

@@thiagoalves4080 you also try with drop the date column

correlation = gold_data.corr(numeric_only=True) # just copy paste this...

I am also facing the same problem

correlation = gold_data.corr(numeric_only=True) # just copy paste this...

ya excellent question, maybe we have to not sure though

Just drop the date column and then run correlation. correlation = df.drop(['Date'],axis=1).corr()

hi! deployment will be covered after deep learning course

you can use forecasting models like LSTM

@@Siddhardhan i tried that but i only get graph i want actual values

import datetime gold_data['Date'] = pd.to_datetime(gold_data['Date']) use this code

Yeah. This is because the corr() function while calculating correlation does not accept Strings(dates are strings in the dataset). So, u can solve this issue by dropping the Date column from the dataset before running that line. Add this cell before that line: gold_data = gold_data.drop('Date',axis=1) and execute again. Now the error will go away

my pleasure 😇

I'll work on that after completing this machine learning course.

yeah, but the value is very low... that means there is not a significant correlation...

hi! you can try a time series prediction kind of an approach for gold price prediction. for car price prediction, you can definitely add the brands as feature for training. my goal while making the videos will be to clearly explain the implementation and the process in it. I may miss a thing or two. so you can definitely improvise a lot.

@@Siddhardhan Thanks a lot sir for your reply

yes, also random forest is used as classifier i think regression is more suitable here. Yes?

I have given the link for the dataset in the description of this video.

You can also search for it in Kaggle

if you completely understand the Project, if you practice the code by yourself, then you can add.

thanks! may be in the future 😄

hi! mail your error with screenshot and your Google colab link to datascience2323@gmail.com

did you get it ?

you can calculate mean absolute error and other metrics.

@@Siddhardhan can you make a video on stock price prediction with rnn lstm and other complex algorithms

hi! refer my other project videos to learn how to make a predictive system.

@@Siddhardhan thanks got it

Yes, you can

you can use mean Absolute error as well. these are metrics for Regression problems

Hi! I have explained it in the video. Kindly check. They are silver price, currency pair value, etc.

@@Siddhardhan Does the data mean that gold price is around 120 in 2018?

Might be. It is in us dollars. I am not sure what is the specific metric. You can refer this for details about the dataset: www.kaggle.com/altruistdelhite04/gold-price-data

Hi! this dataset is very small.. it won't be suitable to do with RNN. but yes. u can definitely try.

hi! you can definitely try.

hi! I have given the Kaggle link for this project in the description of this video... you can get more information there... kindly check.

hi! I am not that active in Kaggle recently