Faster-RCNN finetuning with PyTorch. Object detection using PyTorch. Custom dataset. Wheat detection

In this video, we are going to see how can we fine tune a pretrained faster-rcnn model using PyTorch. The dataset we will be using is the wheat detection dataset.
The video has three objectives:
1. Explain the main portion of the code for fine tuning faster-rcnn.
2. How does the faster-rcnn finetuning look like and what layers are affected as a result?
3. See the whole code for fine tuning faster-rcnn.
We will see how the last layer of the faster-rcnn module is affected due to fine tuning. We also see how the Non Max Suppression (NMS) also works in the context of fine tuning faster-rcnn.
The main code for faster-rcnn is:
model = torchvision.models.detection.fasterrcnn_resnet50_fpn(pretrained = True)
num_classes = 2
in_features = model.roi_heads.box_predictor.cls_score.in_features
model.roi_heads.box_predictor = FastRCNNPredictor(in_features, num_classes)

[@brunonogueirarenzo2306]

Great content, Datum!

[@sachinkapase9218]

Great explanation. please make the video to calculate mAP for this.

[@vijayalaxmiise1504]

Sir, video has clarified many doubts. Thanks a lot

[@studshelper9901]

now found the diamond from coal field, i used to detect text from image Thanks a lot😀😀

[@karlm9584]

I receive an error during setting up the targets for the model training in the for loop - for d in data: ... targ["boxes"] = d[1]["boxes"].to(device) causes an error. On inspection, len(d) returns 1 so the index of d[1] is invalid. How to solve this error please? Any attempts to solve it results in the input data being incompatible with the model

[@tilakrajchoubey5534]

will it work if my images are of different lengths??

[@SumanMondal-qz4qd]

in the last cell where you have drawn the output image you didn't draw the boxes form the output you used the prior bounding boxes which are meant to use for training and validation.... I found this misleading

[@atamsujiwanto1840]

how if we use 3 labels, sir? can you explain the code for 3 labels?

[@rihababbessy5707]

Is that works with tensorflow ?

[@raehanfelda8956]

can you make video about how to tuning yolov8 hyperparameter using raytune?

[@rv0_0]

link to the code please

[@AbhiramMirthipati]

How to get the accuracy of the model get printed ?

[@luisleal4169]

great video! chan you share the code and dataset? do you have any github repo?

[@user-dv2fk4qs4w]

How can I contact you? I want to clarify some points

[@bholuramgurjar900]

Good job! can you share the code?

[@_maha_]

how you evaluate the model

[@nra6925]

can you make tutorial for retinet model?

[@nasimaislambithi9510]

uploads the code link pls

[@user-yj9jp4bz7v]

hi

[@geekroy1728]

bro posts only videos, but not github links🤣🤣🤣