Комментарии:
Great tutorial. One of the best. Thanks.
ОтветитьI learned now python c embedded c java to do it
My problem I don't have powerful gpu.
Basically you will trained your bot files . Cuda, keras tf extension then link with python and execute
Anyway you are a very good engineer 👍
I think a good pace to learn this is 10 per day
ОтветитьWhat should I do if I don't have a MacBook and don't have the option to hit command m to add that "heading" note? kind of unfair you are teaching on apple device.
ОтветитьI thought that I own the AI model I trained.
But actually ... why should somebody owns it instead?!!
I'm learning image classification for my science fair project, and this is the ONLY tutorial I could find that taught me what was happening and how to do it in enough detail. I'm 3 hours in and excited to figure it all out. Is there anything I should know about the differences in the newer versions of TensorFlow and/or Python?
ОтветитьThis deep learning course is incredible. It would be really helpful in solving problem in our final year project. Thank you so much for all this content.
ОтветитьTf version 2.12.0 had an error at the 4 hr 36 minute mark-ish. Had to use model.fit(tf.expand_dims(X, axis=1), y, eopochs=10)
ОтветитьThanks Dan for this wonderful tutorial!
Ответитьloved this course so much!!
I actually bought it on udemy LOVED IT
thanks Daniel and any one who helped in the making of this course !!!!
should I learn tensorflow or PyTorch?
Ответитьthe best till now
ОтветитьThis is the best course on tensorflow that I've seen so far!
ОтветитьAfter going through so many complicated and advanced tutorials (PS :- Didn't understand all the stuff there, this video literally made every concept clear). HATS OFF to such great creators like you sir !. Thank you so very much. Please make more advanced long tutorials so that we can become experts as well. Thank you !
ОтветитьI would like to thank you for the qualtity of the video but also and maybe most important for the support you give with guithub and the UPDATES. I was running a problem of dimensions..and after thinking of giving up I went to your gituhb to run your code to check what was wrong with me....SURPRISE!!!! you updated the code for newest versioon of tf. and i had to expand dim when I fit the model. Thank you so much for all your work. GG to you!!!
Ответить4h07min...Having a shape of 3 for the input is the contrary as a normal dataframe no? In tensor the feature is in index whereas in a data frame the feature is the header of the column and each lkine is an observation.... Am I right when I am highllighting this???
ОтветитьIn video 32, why cant I run: model.fit(X, y, epocks=5)
Instead, when I ran: model.fit(tf.expand_dims(X,axis=1), y, epochs=5), it went oke
discord link?
ОтветитьBest video I watched on whole internet if you want to start learning Tensorflow or deep learning
ОтветитьDear Daniel, I wanted to thank you for the effort you put into making this video. I was learning machine learning when I came across your videos. I used it a lot and my learning process increased greatly. I owe you❤
ОтветитьHey Daniel, I'm studying neuroscience at uni currently and I just wanted to say thanks for the great tf series. I'm doing data analysis for the lab I'm working in and have been trying to implement some deep learning techniques to speed things up. This channel has been the most helpful resource by far for my own exploration of ML, so keep on, brother, and thanks again.
ОтветитьBought your course from Udemy, its huge, but its the best course on earth.
ОтветитьI am from India...Sir!....u r teaching well sir!
ОтветитьThanks a lot man, its 2023 and this video is the best you could get anywhere
ОтветитьHow can I support ?
ОтветитьIn the linear regression model I believe there is a theoretical error, adding a level without any activation function makes no sense because everything is equivalent to a single linear level, can you explain to me why the network improves anyway?
Is there something I'm missing?
What you are doing is wonderful and we all appreciate this effort. I have a question to ask, if you could help me with the first regression model as it's giving me error during model fitting and saying that ndim should at least be 2 but it's 1 in your case and the model isn't working. I tried to reshape the model and it worked but then I'm not sure how to do prediction on dimension 2 array.
ОтветитьAmazing tutorial;🥰 Thankyou; Greetings from Austria Tyrol
ОтветитьAs per D Burke, Let me jog down what I learn from trying to use neural network model using tensorflow on the famous iris data.
First since there are 4 inputs and 1 out put with 3 classes, this is a multi classification problem
When using the model.add format of building my tensorflow, the input layer is entered separately as tf.keras.layers.InputLayer(input_shape(inputDim,)
the output layer of course uses a corresponding number of neurons, in the iris case 3 outputs,
for multi-classification problems use sparse_categorical_crossentropy
when defining the metrics, for a reason I don't understand yet, tf.keras.metrics then selecting either Accuracy or categorical_accuracy not working. Instead a simple assignment of metrics=['accuracy'] works fine
Run evaluation on: model.evaluate (X_test.iloc[-20:], y_test.iloc[-20:])
then used the remaining to predict and check the result. For example I used the last 10 values to predict and compare. It returned only one error.
def irisPredictor(i): return df.iloc[np.where(df['target']==i)]['target_name'].unique()
for i in np.argmax(model.predict(X_test.iloc[-10:], verbose=False), axis=1): print(irisPredictor(i))
df.iloc[y_test.iloc[-10:].index]['target_name']
Hi Daniel,
For the decision boundary in video64, we could import DecisionBoundaryDisplay from sklearn. In my machine learning journey I try to find a package/library before creating it from scratch. Is that ok or u suggest we do everything from scratch?
Thank you for your help.
Bought this on Udemy. Worth the purchase.
ОтветитьRead About Islam man❤
ОтветитьJust wanna say thanks
ОтветитьHii sir actually i need a suggestion about which book i choose for learning Ai Ml including deep learning that cover both theories and implementation.
I watched all video but i need book like who can help in learning and as well helpful in giving practical knowledge
What is the pre-requisite before coding this ?
P.s- i know C++ and java
Just in case anybody else on mac m2 or m1 was very frustrated at the model throwing an error when in notebook 01 when we are compiling and fitting the model. change the line optimizer=tf.keras.optimizers.SGD() to optimizer=tf.keras.optimizers.legacy.SGD(). It reverts back to older XLA architecture becuase the new architecture isnt yet compatible with new mac silicon.
Ответитьcant thank you enough , just goes to show if you have a good teacher like Daniel, even machine learning , Ai and high end concepts become clear in the seekers mind , god bless bro , keep up the good work , thank you :))
ОтветитьJust completed this course, there is a certain poise in how you teach man. Amazing! will be learning more from you in future!
Ответить