fix: draft

one_md_per_day_format/piscine/Week3/w3day03.md

@@ -242,9 +242,3 @@ model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accur
 model.fit(X_train_sc, y_train_multi_class, epochs = 1000, batch_size=20)
 ```
 
-
-
-
-# Exercise 6 GridSearch
-
-https://medium.com/@am.benatmane/keras-hyperparameter-tuning-using-sklearn-pipelines-grid-search-with-cross-validation-ccfc74b0ce9f

one_md_per_day_format/projects/project3/project3.md

@@ -0,0 +1,12 @@
+# Computer vision
+
+MNIST dataset 
+
+Andrew Ng course 
+
+Manage image with python 
+
+
+Use a pre-trained CNN: Transformers
+
+Optional: Hack the CNN https://medium.com/@ageitgey/machine-learning-is-fun-part-8-how-to-intentionally-trick-neural-networks-b55da32b7196