Update README.md

Author: Frederik Reiter

README.md

@@ -1,5 +1,7 @@
 # Handwriting Recognition
 
+NOT MAINTAINED!
+
 ## Introduction
 In this repository I used the [NIST Special Database 19](https://www.nist.gov/srd/nist-special-database-19) and [Tensorflow](https://www.tensorflow.org/) to create a convolutional neural network, which recognizes handwritten digits. It differentiates between 47 classes: All uppercase letters, all numbers and a few lower case letters.I was able to achieve an accuracy of about 87%. If you have any questions, please feel free to open an issue, I will respond to all issues and do my best to help.
 
@@ -37,4 +39,4 @@ First I reconstruct the same graph that I used for training, then load the saved
 I made a dictionary that simply has the letter or number as value for each label. Refer to predict.py for more information.
 
 ## Notes
-This Readme doesn't go into very much detail on how the code works. I will answer issues as detailed as I can if you have specific questions, but if you don't know anything about Tensorflow, I can't help you that much. Again, I can really recommend [this course on udemy.com](https://www.udemy.com/complete-guide-to-tensorflow-for-deep-learning-with-python). I am not affiliated with udemy or the creator of this course, but I acquired most of mz know-how from that course. 
+This Readme doesn't go into very much detail on how the code works. I will answer issues as detailed as I can if you have specific questions, but if you don't know anything about Tensorflow, I can't help you that much. Again, I can really recommend [this course on udemy.com](https://www.udemy.com/complete-guide-to-tensorflow-for-deep-learning-with-python). I am not affiliated with udemy or the creator of this course, but I acquired most of mz know-how from that course.