Commit581f669

committed

added stock prediction README & fixed kmeans tutorial code

1 parent1f46e8e commit581f669Copy full SHA for 581f669

File tree

+10

-3

lines changed

machine-learning
- kmeans-image-segmentation
  - kmeans_segmentation.py
- stock-prediction
  - README.md

+10

-3

lines changed

Lines changed: 9 additions & 2 deletions

Original file line number	Diff line number	Diff line change
`@@ -28,8 +28,11 @@`
`28`	`28`	`# convert back to 8 bit values`
`29`	`29`	`centers=np.uint8(centers)`
`30`	`30`
	`31`	`+# flatten the labels array`
	`32`	`+labels=labels.flatten()`
	`33`	`+`
`31`	`34`	`# convert all pixels to the color of the centroids`
`32`		`-segmented_image=centers[labels.flatten()]`
	`35`	`+segmented_image=centers[labels]`
`33`	`36`
`34`	`37`	`# reshape back to the original image dimension`
`35`	`38`	`segmented_image=segmented_image.reshape(image.shape)`
`@@ -38,10 +41,14 @@`
`38`	`41`	`plt.imshow(segmented_image)`
`39`	`42`	`plt.show()`
`40`	`43`
`41`		`-# disable only the cluster number 2`
	`44`	`+# disable only the cluster number 2 (turn the pixel into black)`
`42`	`45`	`masked_image=np.copy(image)`
	`46`	`+# convert to the shape of k clusters`
	`47`	`+masked_image=masked_image.reshape((-1,k))`
`43`	`48`	`masked_image[labels==2]= [0,0,0]`
`44`	`49`
	`50`	`+# convert back to original shape`
	`51`	`+masked_image=masked_image.reshape(image.shape)`
`45`	`52`	`# show the image`
`46`	`53`	`plt.imshow(masked_image)`
`47`	`54`	`plt.show()`

Lines changed: 1 addition & 1 deletion

Original file line number	Diff line number	Diff line change
`@@ -2,5 +2,5 @@`
`2`	`2`
`3`	`3`	`To run this:`
`4`	`4`	-`pip3 install -r requirements.txt`
`5`		-- Please read the tutorial before using this, edit`parameters.py` for your needs and run`train.py`. This will start training using the parameters you specified, you can use`tensorboard` on`logs` folder to visualize your training process.
	`5`	+- Please read the[tutorial](https://www.thepythoncode.com/article/stock-price-prediction-in-python-using-tensorflow-2-and-keras) before using this, edit`parameters.py` for your needs and run`train.py`. This will start training using the parameters you specified, you can use`tensorboard` on`logs` folder to visualize your training process.
`6`	`6`	- Once you trained your model, use`test.py` to evaluate and test your model.

Comments

(0)