Let's implement ResNet

[@alexviveslliset1464]

Baseblock is not being used, why is that? Doesn't that mean this is not correct?

[@nurali2525]

I guess model was not saved after training completed, right?

[@AyushChaurasia]

yup. it wasn't saved :)

[@SuperJg007]

in bottleneck's second conv2d layer, the height and width won't change right? cuz of the formula
((x-k+2p)/s)+1
=( (x - 3 + 2)/1) +1
=x

[@AyushChaurasia]

Yes that's right

[@deraktdar]

what is the baseBlock for? You don't seem to use it.

[@deraktdar]

oh I guess baseBlock and bottleNeck are replaceable.

[@wolfisraging]

Thanks for uploading

[@dapengzhang4473]

Hi, is it the same theory to build Resnet101? Where do I need to make changes? Thank you

[@sachinkun21]

Great work bro. Thanks for sharing this..
I have a question. Can you please explain what is the difference between identity block and Conv2D block wrt tf implementation code? Is identity block same as the bottleneck block you created in pytorch?

[@sayeedchowdhury11]

At line 75, shouldn't it be self.inplanes != planes * block.expansion as the condn? I think there is a mistake

[@AyushChaurasia]

Here we're setting the planes value not checking it.

[@sarthaknarayan2159]

Sorry for the trouble and thanks for the help

[@AyushChaurasia]

I'm happy that you understood the concept 😊

[@autripat]

You don't type the line of code, but paste it from somewhere? How do you do that?

[@AyushChaurasia]

Its quite simple..I wrote the code beforehand and tested it.. Then I used undo command to delete all the code line by line...and then used the redo command to again make it appear.

[@studiousgirl3202]

i dont have pip3 installed. Can i replace it with pip and proceed?

[@AyushChaurasia]

You can easily update pip to pip3. There are various tutorials available on the internet. But if you want to stick with pip, you can.

[@phuccoiinkorea3341]

Damn I'm still on week 2. I'll rewatch this later

[@AyushChaurasia]

Sure..you can watch it whenever you want because implementation of resnets is not discussed in the course..best of luck 😊

[@Karan-ow4wl]

What does self.dim_change(res) mean?

[@vigneshvenkataraman8]

could you explain this??

[@Karan-ow4wl]

@@vigneshvenkataraman8 it calls the self.dim_change function below.

[@vigneshvenkataraman8]

@@Karan-ow4wl thx a lot

[@hasibzunair]

Nice!

[@AyushChaurasia]

Thanks

[@dexter-2633]

How do we use the trained model for testing or in production mode?

[@dimplelidar7668]

Low volume. Pls increase the volume

[@AyushChaurasia]

I'll make sure the next time

[@neelabhjatin4660]

Seriously you have gotta stop coughing.

[@AyushChaurasia]

I guess that problem is resolved in later videos 😉