hi everyone i have a question. In the figure, is the graph represented a heavy-tail normal distribution? i have: median=533 ;mode=547; maximum degree=845
Related
I am training an object detector for my own data using Tensorflow Object Detection API. I am following the (great) tutorial by Dat Tran https://towardsdatascience.com/how-to-train-your-own-object-detector-with-tensorflows-object-detector-api-bec72ecfe1d9. I am using the provided ssd_mobilenet_v1_coco-model pre-trained model checkpoint as the starting point for the training. I have only one object class.
I exported the trained model, ran it on the evaluation data and looked at the resulted bounding boxes. The trained model worked nicely; I would say that if there was 20 objects, typically there were 13 objects with spot on predicted bounding boxes ("true positives"); 7 where the objects were not detected ("false negatives"); 2 cases where problems occur were two or more objects are close to each other: the bounding boxes get drawn between the objects in some of these cases ("false positives"<-of course, calling these "false positives" etc. is inaccurate, but this is just for me to understand the concept of precision here). There are almost no other "false positives". This seems much better result than what I was hoping to get, and while this kind of visual inspection does not give the actual mAP (which is calculated based on overlap of the predicted and tagged bounding boxes?), I would roughly estimate the mAP as something like 13/(13+2) >80%.
However, when I run the evaluation (eval.py) (on two different evaluation sets), I get the following mAP graph (0.7 smoothed):
mAP during training
This would indicate a huge variation in mAP, and level of about 0.3 at the end of the training, which is way worse than what I would assume based on how well the boundary boxes are drawn when I use the exported output_inference_graph.pb on the evaluation set.
Here is the total loss graph for the training:
total loss during training
My training data consist of 200 images with about 20 labeled objects each (I labeled them using the labelImg app); the images are extracted from a video and the objects are small and kind of blurry. The original image size is 1200x900, so I reduced it to 600x450 for the training data. Evaluation data (which I used both as the evaluation data set for eval.pyand to visually check what the predictions look like) is similar, consists of 50 images with 20 object each, but is still in the original size (the training data is extracted from the first 30 min of the video and evaluation data from the last 30 min).
Question 1: Why is the mAP so low in evaluation when the model appears to work so well? Is it normal for the mAP graph fluctuate so much? I did not touch the default values for how many images the tensorboard uses to draw the graph (I read this question: Tensorflow object detection api validation data size and have some vague idea that there is some default value that can be changed?)
Question 2: Can this be related to different size of the training data and the evaluation data (1200x700 vs 600x450)? If so, should I resize the evaluation data, too? (I did not want to do this as my application uses the original image size, and I want to evaluate how well the model does on that data).
Question 3: Is it a problem to form the training and evaluation data from images where there are multiple tagged objects per image (i.e. surely the evaluation routine compares all the predicted bounding boxes in one image to all the tagged bounding boxes in one image, and not all the predicted boxes in one image to one tagged box which would preduce many "false false positives"?)
(Question 4: it seems to me the model training could have been stopped after around 10000 timesteps were the mAP kind of leveled out, is it now overtrained? it's kind of hard to tell when it fluctuates so much.)
I am a newbie with object detection so I very much appreciate any insight anyone can offer! :)
Question 1: This is the tough one... First, I think you don't understand correctly what mAP is, since your rough calculation is false. Here is, briefly, how it is computed:
For each class of object, using the overlap between the real objects and the detected ones, the detections are tagged as "True positive" or "False positive"; all the real objects with no "True positive" associated to them are labelled "False Negative".
Then, iterate through all your detections (on all images of the dataset) in decreasing order of confidence. Compute the accuracy (TP/(TP+FP)) and recall (TP/(TP+FN)), only counting the detections that you've already seen ( with confidence bigger than the current one) for TP and FP. This gives you a point (acc, recc), that you can put on a precision-recall graph.
Once you've added all possible points to your graph, you compute the area under the curve: this is the Average Precision for this category
if you have multiple categories, the mAP is the standard mean of all APs.
Applying that to your case: in the best case your true positive are the detections with the best confidence. In that case your acc/rec curve will look like a rectangle: you'd have 100% accuracy up to (13/20) recall, and then points with 13/20 recall and <100% accuracy; this gives you mAP=AP(category 1)=13/20=0.65. And this is the best case, you can expect less in practice due to false positives which higher confidence.
Other reasons why yours could be lower:
maybe among the bounding boxes that appear to be good, some are still rejected in the calculations because the overlap between the detection and the real object is not quite big enough. The criterion is that Intersection over Union (IoU) of the two bounding boxes (real one and detection) should be over 0.5. While it seems like a gentle threshold, it's not really; you should probably try and write a script to display the detected bounding boxes with a different color depending on whether they're accepted or not (if not, you'll get both a FP and a FN).
maybe you're only visualizing the first 10 images of the evaluation. If so, change that, for 2 reasons: 1. maybe you're just very lucky on these images, and they're not representative of what follows, just by luck. 2. Actually, more than luck, if these images are the first from the evaluation set, they come right after the end of the training set in your video, so they are probably quite similar to some images in the training set, so they are easier to predict, so they're not representative of your evaluation set.
Question 2: if you have not changed that part in the config file mobilenet_v1_coco-model, all your images (both for training and testing) are rescaled to 300x300 pixels at the start of the network, so your preprocessings don't matter.
Question 3: no it's not a problem at all, all these algorithms were designed to detect multiple objects in images.
Question 4: Given the fluctuations, I'd actually keep training it until you can see improvement or clear overtraining. 10k steps is actually quite small, maybe it's enough because your task is relatively easy, maybe it's not enough and you need to wait ten times that to have significant improvement...
I download the following graph-cut code:
https://github.com/shaibagon/GCMex
I compiled the mex files, and ran it for pre-defined image in the code (which is rgb image)
I wanna optimize the image segmentation results,
I have probability map of the image, which its dimension is (width,height, 5). Five probability distribution over the image dimension are stacked together. each relates to one the classes.
My problem is which parts of code should according to the probability image.
I want to define Data and Smoothing terms based on my application.
My question is:
1) Has someone refined the code according to the defining different energy function (I wanna change Unary and pair-wise formulation).
2) I have a stack of 3D images. I wanna define 6-neighborhood system, 4 neighbors in current slice and the other two from two adjacent slices. In which function and part of code can I do the refinements?
Thanks
I have some data that tells me the amount of hours water is available for particular towns.
You can see it here
I want to use train a Multilayer Perceptron based on that data, to take a set of coordinates and indicate the approximate number of hours for which that coordinate will have water.
Does this make sense?
If so, am I correct in saying, there has to be two input layers? One for lat and one for long. And the output layer should be the number of hours.
Would love some guidance.
I would solve that differently:
Just create an ArrayList of WaterInfo:
WaterInfo contains lat,lon, waterHours.
Then for a given coordinate search the closest WaterInfo in the list.
Since you have not many elements, just do a brute force search, to find the closest.
You further can optimize, to find the three closest WaterInfo points, and calculate the weithted average of WaterHours. As weight you use the air distance from current position to Waterinfo position.
To answer your question:
"Does this makes sense"?
From the goal to get a working solution: NO!
Ask yourself, why do you want to use MLP for this task.
Further i doubt that using two layers for lat / long makes sense.
A coordinate (lat/lon) is one point on the world, so that should be one layer in the model. You can convert the lat/lon coord to a cell identifier: Span a grid over Brazil; with cell width 10 or 50km; now convert a lat/long coordinate to a cellId: Like E4 on a chess board, you will calculate one integer value representing the cell. (There are other solutions to get an unique number, too, choose one you like)
Now you have a modell geoCellID -> waterHours, which better represents the real world situation.
I want to detect the best rototraslation matrix between two set of points.
The second set of points is the same of the first, but rotated, traslated and affecteb by noise.
I tried to use least squared method by obviously the solution is usually similar to a rotation matrix, but with incompatible structure (for example, where i should get a value that represents the cosine of an angle i could get a value >1).
I've searched for the Constrained Least Squared method but it seems to me that the constrains of a rototraslation matrix cannot be expressed in this form.
In this PDF i've stated the problem more formally:
http://dl.dropbox.com/u/3185608/minquad_en.pdf
Thank you for the help.
The short answer: What you will need here is "Principal Component Analysis".
Apply this to both sets of points centered at their respective centers of mass. The PCA will effectively give you a rotation matrix for each aligned to the data set principal components. Multiplying the inverse matrix of the original set by the new rotation will give you a matrix that takes the old (centered) set to the new. Inverse translations and translations can similarly be applied to the rotation to create a homogeneous matrix that maps the one set to the other.
The book PRINCE, Simon JD. Computer vision: models, learning, and inference. Cambridge University Press, 2012.
gives, in Appendix "B.4 Reparameterization", some info about how to constrain a matrix to be a rotation matrix.
It seems to me that your problem has also a solution based on SVD: see the Kabsch algorithm also described by Olga Sorkine-Hornung and Michael Rabinovich in
Least-Squares Rigid Motion Using SVD and, more practically, by Nghia Kien Ho in FINDING OPTIMAL ROTATION AND TRANSLATION BETWEEN CORRESPONDING 3D POINTS.
I have a set of vectors in multidimensional space (may be several thousands of dimensions). In this space, I can calculate distance between 2 vectors (as a cosine of the angle between them, if it matters). What I want is to visualize these vectors keeping the distance. That is, if vector a is closer to vector b than to vector c in multidimensional space, it also must be closer to it on 2-dimensional plot. Is there any kind of diagram that can clearly depict it?
I don't think so. Imagine any twodimensional picture of a tetrahedron. There is no way of depicting the four vertices in two dimensions with equal distances from each other. So you will have a hard time trying to depict more than three n-dimensional vectors in 2 dimensions conserving their mutual distances.
(But right now I can't think of a rigorous proof.)
Update:
Ok, second idea, maybe it's dumb: If you try and find clusters of closer associated objects/texts, then calculate the center or mean vector of each cluster. Then you can reduce the problem space. At first find a 2D composition of the clusters that preserves their relative distances. Then insert the primary vectors, only accounting for their relative distances within a cluster and their distance to the center of to two or three closest clusters.
This approach will be ok for a large number of vectors. But it will not be accurate in that there always will be somewhat similar vectors ending up at distant places.