# squeezenet with Deepstream

**URL:** <https://forums.developer.nvidia.com/t/squeezenet-with-deepstream/56439>\
**Category:** General\
**Created:** [December 21, 2017, 5:42pm UTC](https://forums.developer.nvidia.com/t/squeezenet-with-deepstream/56439 "2017-12-21T17:42:10Z")\
**Posts on this page:** 6\
**Page:** 1

<div class="post-metadata">

**Author:** ![Ravik](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@Ravik](https://forums.developer.nvidia.com/u/Ravik)\
**Post date:** [December 21, 2017, 5:42pm UTC](https://forums.developer.nvidia.com/t/squeezenet-with-deepstream/56439/1 "2017-12-21T17:42:10Z")

</div>

Can you provide an example for squeezenet or Resnet 50 with Deepstream on Nvidia JetsonTX2

Regards

---

<div class="post-metadata">

**Author:** ![AastaLLL](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/aastalll/32/14043_2.png) [@AastaLLL](https://forums.developer.nvidia.com/u/AastaLLL)\
**Post date:** [December 22, 2017, 6:02am UTC](https://forums.developer.nvidia.com/t/squeezenet-with-deepstream/56439/2 "2017-12-22T06:02:18Z")

</div>

Hi,

It’s easy to launch a Caffe-based model on Jetson TX2.

Here is our tutorial for launching Caffe model with TensorRT on Jetson:

> **[GitHub - dusty-nv/jetson-inference: Hello AI World guide to deploying deep-learning...](https://github.com/dusty-nv/jetson-inference)**
>
> Hello AI World guide to deploying deep-learning inference networks and deep vision primitives with TensorRT and NVIDIA Jetson.

Thanks.

---

<div class="post-metadata">

**Author:** ![AastaLLL](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/aastalll/32/14043_2.png) [@AastaLLL](https://forums.developer.nvidia.com/u/AastaLLL)\
**Post date:** [December 22, 2017, 6:03am UTC](https://forums.developer.nvidia.com/t/squeezenet-with-deepstream/56439/3 "2017-12-22T06:03:35Z")

</div>

More, please noticed that DeepStream is not available for Jetson yet.

---

<div class="post-metadata">

**Author:** ![Ravik](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@Ravik](https://forums.developer.nvidia.com/u/Ravik)\
**Post date:** [December 22, 2017, 6:11am UTC](https://forums.developer.nvidia.com/t/squeezenet-with-deepstream/56439/4 "2017-12-22T06:11:52Z")

</div>

Hi AstaLLL,

We trained couple of custom models, we are using DeepDetect for inference, we have some challenges when using deepstream.

We are in touch with engineering thanks

---

<div class="post-metadata">

**Author:** ![Ravik](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@Ravik](https://forums.developer.nvidia.com/u/Ravik)\
**Post date:** [January 24, 2018, 6:14pm UTC](https://forums.developer.nvidia.com/t/squeezenet-with-deepstream/56439/5 "2018-01-24T18:14:39Z")

</div>

Hi AastaLLL,

it seems some of the detection layers we used are not supported by Deepstream, is it possible to commercially engage Nvidia engineering team ?

---

<div class="post-metadata">

**Author:** ![kayccc](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/kayccc/32/14043_2.png) [@kayccc](https://forums.developer.nvidia.com/u/kayccc)\
**Post date:** [January 29, 2018, 7:16am UTC](https://forums.developer.nvidia.com/t/squeezenet-with-deepstream/56439/6 "2018-01-29T07:16:37Z")

</div>

> [@](#):
>
> Hi AastaLLL,
> 
> it seems some of the detection layers we used are not supported by Deepstream, is it possible to commercially engage Nvidia engineering team ?

Hi RaviKiranK,

Please contact with our regional sales team for your requirement.

Thanks