Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • 19:07

    mergify[bot] on master

    feat: chain uses dto end to end… (compare)

  • 19:07
    mergify[bot] closed #1340
  • 17:15
    mergify[bot] synchronize #1340
  • 17:15
    mergify[bot] labeled #1340
  • 15:53
    Bycob synchronize #1342
  • 15:09
    Bycob synchronize #1342
  • 15:08
    Bycob synchronize #1340
  • 14:34
    Bycob synchronize #1342
  • 14:26
    Bycob edited #1342
  • 14:25
    Bycob synchronize #1342
  • 13:58
    mergify[bot] review_requested #1342
  • 13:57
    Bycob labeled #1342
  • 13:57
    Bycob unlabeled #1342
  • 13:56
    Bycob synchronize #1342
  • Sep 18 13:16
    beniz labeled #1340
  • Sep 17 09:37
    mergify[bot] labeled #1343
  • Sep 17 09:37

    mergify[bot] on master

    fix(cpu): cudnn is now on by de… (compare)

  • Sep 17 09:37
    mergify[bot] closed #1343
  • Sep 17 07:39
    fantes labeled #1343
  • Sep 17 07:37
    mergify[bot] review_requested #1343
Emmanuel Benazera
@beniz
it picks gpuid or 0 as default
we're opened to suggestions regarding most useful options
cchadowitz-pf
@cchadowitz-pf
:+1: that sounds good, just trying to better understand how it works now to see how to best utilize it :) i think as is is probably perfectly fine, but will let you know if i have additional thoughts. thanks!
Bhavik samcom
@Bhavik_samcom_gitlab

Is there any way to pass the additional parameters with simsearch - indexing, like below
"data":[
"id": "<image_url>",
["id", "category", "<image_url>"]
...
]

I want to perform search of specific category, so these way it get simplified, but unable to figure out the way to do so. can you please guide here the possible ways as well?

Emmanuel Benazera
@beniz
hi @Bhavik_samcom_gitlab the spirit is rather to storz metadata outside of the similarity index. This is because by design we consider DD is not concerned with structured storage. Let me know if this is not clear or if I misunderstood your question
Bhavik samcom
@Bhavik_samcom_gitlab
Hi @beniz Thanks for demonstrating the vision.
I understand, is there not even an option to store just a simple id that will be associated with the image (without other metadata)? I am asking because the only thing returned is an image url which is not necessarily unique or a key I can use in my database for querying.
Emmanuel Benazera
@beniz
yes the id is stored of course and its the URL, let me know @Bhavik_samcom_gitlab if you have issues getting it back from the API. So the idea is yiu control it via the URL that is a UUID.
You can keep a matching table outside of DD between URLs and you internal identifier.
YaYaB
@YaYaB
Hey DD's team :)
I am trying to play a bit with tensorrt.
Do you have somewhere the compatible models from caffe?
I already used the googlenet, resnet18 and ssd however it does not seem to work for:
  • resnext
    TensorRT does not support in-place operations on input tensors in a prototxt file.
    [2021-03-17 09:55:52.917] [imageserv_resnet] [error] mllib internal error: Error while parsing caffe model for conversion to TensorRT
  • se_resnext
    [2021-03-17 09:42:18.603] [imageserv_resnet] [info] Bias weights are not set yet. Bias weights can be set using setInput(2, bias_tensor) API call.
    [2021-03-17 09:42:18.606] [imageserv_resnet] [info] Bias weights are not set yet. Bias weights can be set using setInput(2, bias_tensor) API call.
    [2021-03-17 09:42:18.606] [imageserv_resnet] [info] Bias weights are not set yet. Bias weights can be set using setInput(2, bias_tensor) API call.
    [2021-03-17 09:42:18.606] [imageserv_resnet] [info] Bias weights are not set yet. Bias weights can be set using setInput(2, bias_tensor) API call.
    [2021-03-17 09:42:18.606] [imageserv_resnet] [info] Bias weights are not set yet. Bias weights can be set using setInput(2, bias_tensor) API call.
    could not parse layer type Axpy
    [2021-03-17 09:42:18.606] [imageserv_resnet] [error] mllib internal error: Error while parsing caffe model for conversion to TensorRT
  • efficient model
    [2021-03-17 09:33:45.133] [imageserv_resnet] [info] Bias weights are not set yet. Bias weights can be set using setInput(2, bias_tensor) API call.
    could not parse layer type Swish
    [2021-03-17 09:33:45.138] [imageserv_resnet] [error] mllib internal error: Error while parsing caffe model for conversion to TensorRT
Emmanuel Benazera
@beniz
Hi @YaYaB the caffe to trt parser may not support all layers. As for the in-place operations in the resnext architecture (without se_), it may not come from DD right ?
Efficientnet is not efficient at all in practice, as a side note, you better not use it.
YaYaB
@YaYaB
For resnext I saw some people able to use it with trt from tensorflow. Maybe the caffe implementation in the proto may be problematic. Well noted for EfficientNet!
Do you plan on integrating external tensorrt models, for instance from torch, etc. ?
Emmanuel Benazera
@beniz
torch to trt is hard from C++, it is more or less easily done by loading the weights back with a Python script that describes an identical architecture, e.g. easy from torchvision.
YaYaB
@YaYaB
Ok thanks for the info, cheers!
Romain Guilmont
@rguilmont
Hello there :) Hope all DD team is doing great !
I'm starting the prometheus exporter ( i'll soon give you details @beniz with link to the repo/docker image ). I just have a doubt on 2 metrics: total_transform_duration_ms and total_predict_duration_ms. Does predict duration includes the transform duration ? It looks like it does.
Emmanuel Benazera
@beniz
hi @rguilmont yes it does.
Romain Guilmont
@rguilmont
perfect, thanks !
Romain Guilmont
@rguilmont
image.png
Here's a draft of Grafana dashboard that uses prometheus metrics from deepdetect exporter
Emmanuel Benazera
@beniz
beautiful :)
Romain Guilmont
@rguilmont
Hey guys ! I have noticed a memory leak ( ram, not gpu memory ) on latest 0.15 DeepDetect. Before investigating more, is it something you're already aware of ?
Emmanuel Benazera
@beniz
hello, probably not, you can explain it here or in an issue.
Romain Guilmont
@rguilmont
I'll do an issue, i tried to identify clearly which kind of requests caused the leak and i was not able to yet.
Romain Guilmont
@rguilmont
jolibrain/deepdetect#1260 here's the issue. Unfortunately it's not perfect but i hope it can help you to pin-point the issue
tinco
@tinco:matrix.org
[m]
hi! I'm looking to setup a low code system for training object detection models, and it seems deep detect might fit the bill
would it be easy to integrate new architectures that are currently not explicitly supported by deepdetect? for example detectron2 on caffe2?
Emmanuel Benazera
@beniz
hi @tinco we used to support detectron2 with caffe2, it's deprecated now. If your task is object detection, DD comes with plenty of other battle-tested architectures.
tinco
@tinco:matrix.org
[m]
ah ok, I'm not super up to speed on what's the latest and greatest, I'm mostly hoping to enable our team to run experiments with different models themselves
Emmanuel Benazera
@beniz
DD has light models for simple problems/embedded/high fps applications, and larger models for more complicated problems.
tinco
@tinco:matrix.org
[m]
for the past couple years we've been working with a proprietary system that first worked with resnet and now yolov4, and our researcher is training it to segment buildings into components
Emmanuel Benazera
@beniz
so object detection + segmentation ?
tinco
@tinco:matrix.org
[m]
yeah, that's why detectron2 appealed to me, they've got this cover photo with really neat segmentation
Emmanuel Benazera
@beniz
what we'd do with DD is detection with refinedet_512 then apply a segmenter to every object, using a chain, but that's different than detectron2 that does both in a single pass.
tinco
@tinco:matrix.org
[m]
ah right
so why did you deprecate detectron2, was it not used a lot or does it perform less well than the alternatives more most use cases?
Emmanuel Benazera
@beniz
because pytorch did basically deprecate caffe2
tinco
@tinco:matrix.org
[m]
ahh alright, thanks for catching me up haha :D
Emmanuel Benazera
@beniz
Also, very don't see may semantic segmentation models with our customers. I believe this is due to labeling costs. We like to automate the labeling steps ;)
tinco
@tinco:matrix.org
[m]
we're paying students haha, and our customers are actually paying for the labeling already, we're trying to optimize the process
Emmanuel Benazera
@beniz
sure
tinco
@tinco:matrix.org
[m]
so what's your business model, do you sell consultancy around deepdetect, or is it a tool you use to implement machine learning at your clients?
Emmanuel Benazera
@beniz
yes, we are mostly a service company, we serve large corps mostly on complex problems, when there's no product on the shelves, or when there's not much litterature on whether a problem can be solved with ML/DL/RL. DD is the tool that embeds everything we have solved, and that goes into production.
tinco
@tinco:matrix.org
[m]
very cool, thanks for sharing!
Emmanuel Benazera
@beniz
no worries, we've got several requests for yolo models recently, so they might make it into the framework soon.
As for semantic seg, the path for us will be through our torch C++ backend, here again depending on requests and usage.
you can open issues on github for feature requests
tinco
@tinco:matrix.org
[m]
hey, so I just noticed that yolov5 is in pytorch, does that mean the model could just be dropped through a model repository in deep detect, or will there be some code needed as well?
Emmanuel Benazera
@beniz
almost... we've looked at it recently, and the ultralytics repo has code that makes it a bit more tricky, typically there's a bbox filtering step that they did put outside the model, which is weird, and that would need to recoded, that's for the detail.
we've got the request several times now, so we'll try to have an answer to yolov5 :)