What is missing from MATLAB?
I'm curious, is there something you wish to do with MATLAB but you can't, maybe something you can do with other similar software but can't with MATLAB?
For new entries, please use the follow-up thread here. Please do not post new answers in this thread.
470 Comments
Time Descending
Compare sldd files and publish a HMTL report using Matlab script:
I can compare .slx, .m, etc. files using the visdiff function and publish a report using the publish function like this:
compare = visdiff('File1.slx', 'File1_changed.slx');
publish(compare, 'format', 'html');
But I can´t do the same with .sldd files:
compare = visdiff('File1.sldd', 'File1_changed.sldd');
publish(compare, 'format', 'html');
Error using visdiff
Unable to compare 'C:\basic\File1.sldd' and 'C:\basic\File1_changed.sldd' without opening the Comparison Tool.
Why if I can generate an HTML report using the Comparision Tool in the same way, both for .slx and .sldd files?
I want to use the SymbolicTransformer function of python GPlearn
Why does the paid matlab not have some functions on free python? I can't understand
How to visualize the changes in dlarray/weight distribution with histograms in deep learning?
Histogram displays how the trend of tensor (weight, bias, gradient, etc.) changes during the training process in the form of histogram. Developers can adjust the model structures accurately by having an in-depth understanding of the effect of each layer.
I very much hope that the official version will strengthen the readstruct function in the future! Lack of complete uniformity to support more format requirements.
Missing: ability to set multiple bits within the same word, by bit number.
Currently, if you try something like
bitset(A, [5 3])
then you get back a vector of values, setting each of the bits in turn, instead of setting all those bits within a single word.
You have to resort to things like
bitor(A, sum(bitset(0, [5 3])))
- action recognition: https://github.com/open-mmlab/mmaction
- video understanding, X3D: https://github.com/facebookresearch/SlowFast ，https://github.com/kenshohara/3D-ResNets-PyTorch
- reID: https://github.com/JDAI-CV/fast-reid
- track: https://github.com/ZQPei/deep_sort_pytorch，https://votchallenge.net/challenges.html
- Metric Learning: https://github.com/KevinMusgrave/pytorch-metric-learning
- image Classification: efficient,https://github.com/lukemelas/EfficientNet-PyTorch
- object detection: Centernet, https://github.com/xingyizhou/CenterNet , efficientDet, https://github.com/google/automl/tree/master/efficientdet，detectron2，https://github.com/facebookresearch/detectron2 ,PointRCNN,https://github.com/sshaoshuai/PointRCNN
- face recognition: insightface, https://github.com/deepinsight/insightface, sphereface,https://github.com/wy1iu/sphereface, arcface,https://github.com/ronghuaiyang/arcface-pytorch
- Semantic segmentation, instance segmentation，https://github.com/facebookresearch/maskrcnn-benchmark
- deeplearning preprocess data: https://github.com/NVIDIA/DALI
- StyleGAN: https://github.com/NVlabs/stylegan
- DeepNetwork Architecture：Mnasnet，https://github.com/tensorflow/tpu/tree/master/models/official/mnasnet
- deepnetwork inference performance: https://github.com/microsoft/onnxruntime ,https://github.com/zuoqing1988/ZQCNN
and so on....
The above are influential applications of deep learning in various aspects, but it is difficult to reproduce in matlab. Although Matlab2019b version supports automatic differentiation mechanism, it is still difficult to implement algorithms in matlab. The efficiency of the differentiation mechanism is not high, and many operators do not support it. I tried to implement the more famous yolov3/v4 algorithm with the latest MATLAB2020a version, but it is still not satisfactory
In summary, my personal suggestions are like my personal answer above, and I hope that future versions can improve a lot!
以上都为深度学习在各个方面有影响力的应用，但是在matlab中复现困难，虽然Matlab2019b版本支持自动微分机制，但仍然不易在matlab实现算法，微分机制效率不高，很多operators也不支持。。。
总之，我的个人建议就像上面的个人回答建议一样，我希望将来的版本可以有所改善！
Oh, good, thank you for your reminder, I thought I didn’t publish it yesterday!☺😀😀
You said that already; https://www.mathworks.com/matlabcentral/answers/1325-what-is-missing-from-matlab#answer_469935
It disappeared from the bottom of the list because it got voted for and moved upwards.
- action recognition: https://github.com/open-mmlab/mmaction
- video understanding, X3D: https://github.com/facebookresearch/SlowFast ，https://github.com/kenshohara/3D-ResNets-PyTorch
- reID: https://github.com/JDAI-CV/fast-reid
- track: https://github.com/ZQPei/deep_sort_pytorch，https://votchallenge.net/challenges.html
- Metric Learning: https://github.com/KevinMusgrave/pytorch-metric-learning
- image Classification: efficient,https://github.com/lukemelas/EfficientNet-PyTorch
- object detection: Centernet, https://github.com/xingyizhou/CenterNet , efficientDet, https://github.com/google/automl/tree/master/efficientdet，detectron2，https://github.com/facebookresearch/detectron2 ,PointRCNN,https://github.com/sshaoshuai/PointRCNN
- face recognition: insightface, https://github.com/deepinsight/insightface sphereface,https://github.com/wy1iu/sphereface, arcface,https://github.com/ronghuaiyang/arcface-pytorch
- Semantic segmentation, instance segmentation，https://github.com/facebookresearch/maskrcnn-benchmark
- deeplearning preprocess data: https://github.com/NVIDIA/DALI
- StyleGAN: https://github.com/NVlabs/stylegan
- DeepNetwork Architecture：Mnasnet，https://github.com/tensorflow/tpu/tree/master/models/official/mnasnet
- deepnetwork inference performance: https://github.com/microsoft/onnxruntime ,https://github.com/zuoqing1988/ZQCNN
and so on....
The above are influential applications of deep learning in various aspects, but it is difficult to reproduce in matlab. Although Matlab2019b version supports automatic differentiation mechanism, it is still difficult to implement algorithms in matlab. The efficiency of the differentiation mechanism is not high, and many operators do not support it. I tried to implement the more famous yolov3/v4 algorithm with the latest MATLAB2020a version, but it is still not satisfactory
以上都为深度学习在各个方面有影响力的应用，但是在matlab中复现困难，虽然Matlab2019b版本支持自动微分机制，但仍然不易在matlab实现算法，微分机制效率不高，很多operators也不支持。。。
Thank you for your reply, I have been following the latest development progress of "deeplearning toolbox", although there are a lot of new features listed, but most of these are superficial improvements, the substance of the support “operator” is very limited, for the experienced in-depth researcher is still very insufficient.
I'm not familiar enough with Deep Learning Toolbox and the functionality you've requested to determine how much (if any) of what you've requested has been implemented, but the Deep Learning Toolbox Release Notes lists many items for each of the past couple releases. Perhaps you can scan through and determine which of your requests have been implemented.
Ability to use a multidimensional logical array to substitute for multiple dimensions while still specifying other dimensions.
For example if you build an ROI that is going to be 2d, and you cannot currently do things like
array(mask2d, :)
array(mask2d, :, :)
to talk about the rgb elements associated with the mask.
In the particular case of ROI mask you can repmat(mask, 1, 1, 3) but there are plausibly times where you might want to do something like
array(2, mask)
with you wanting to substitute a 2d logical array for the second and third dimension.
- Native support for custom keyboard shortcuts, i.e. let me point to a function to execute when a key-combo is pressed. Currently I can put them in the "Favorites" and add them to quick-access and order them for Alt+1, Alt+2, etc, which is alright, not great if you want more than 2 or 3. There is also theEditorMacro from File Exchange, which is good, but would be nice if it was baked-in to Matlab so I could more easily share helper functions with colleagues.
- Better git support. VS Code is light-years ahead here, for example.
- Extended syntax highlighting. Built-in keywords are highlighted, let's add a color for variables, functions calls, and class constructors/static methods. The example screenshots here are a nice example of how much more readable code is with functions highlighted.
- Bring the improvements of function parameter help text from live scripts to the regular .m files
In release R2021b the Editor now automatically displays contextual hints for arguments, property values, and syntaxes. See the Release Notes for more information. There are some additional editor related changes also listed in the Environment section of those Release Notes, just clear the search term "suggestions" to see them.
" Back in the 90's when I used the Brief programmer's editor,..."
Amen, brother! to all of the above. Anything based on the Windoes CUA is fatally broke before leaving the station.
Besides the features above, column cut/paste, bookmarks, etc., etc., ... the list is nearly endless besides the facility to customize at will what you didn't like or needed improved or simply just wanted to add. Theoretically one could do custom integration in VS editor but the level of complexity is so high as to make it essentially a fulltime occupation to be able to learn it well enough to do anything useful.
Simply even keystroke mapping is broke because you can't get away from the stinkin' context switch with Alt and Ctrl that aren't mappable so all of a sudden you're back at menu level instead of in the file edit context. Pitiful stuff! :(
And, unfortunately, the situation hasn't really improved significantly as far as the editing functionality itself; only the integration of debugging and the accroutrements of syntax coloring, etc., Not that those aren't significant and very useful in their own right and the coding hints are especially good for newbies (if they would only pay attention to them as we see here in Answers they generally don't, just as folks mostly don't read the documentation more than a cursory scan...).
I've retired from active consulting, but while still writing significant code, my general development pattern was to actually write the code using Brief or a workalike outside the ML environment and only if needed more than cursory test/debug, then and only then load into the integrated editor for that purpose to set breakpoints.
Being able to compile a desktop standalone app made in app designer that works on a 32 bit Windows.
Although I agree with your conclusion, I do want to comment on the level of obsolescence of 32 bit.
In early 2016 I bought a 2-in-1 and didn't pay enough attention. When trying to update to Win10x64 I discovered that the CPU itself was 32 bit, and the 64 bit option wasn't the default. This was from a reputable brand. So it is not that obsolete. I would even suggest computers like that are a prime target for cross-compilation, because small apps will run fine, but complex calculations would be an issue.
This will never be possible. 32 bit support is gone and it will never return: it presented too many limitations. The current minimum memory for MATLAB is the same as the maximum memory for 32 bit.
I did some checking a couple of years ago. I could only find one Intel 32 bit x86 CPU that had been manufactured in the previous decade, and manufacturing of it had stopped several years before I checked. I checked the major laptop and notebook manufacturers and could only find one Intel or AMD based notebook that was manufactured with that 32 bit CPU, with the manufacturer having switched to x64 completely after about 18 months.
My mother's 15+ year old system has an x64 and it was always a low budget system: by 15 years ago it was no longer economical to put x86 cpus into low budget computers.
If your hardware is x86 then you are talking about the kind of computers that charities will no longer accept to send to needy people, because the systems are so obsolete. You can get better systems literally given away on Craigslist or equivalent.
I would recommend against Mathworks putting any effort into this idea.
Sign in to participate
