-
I found that the first feature layer output is 19*19 and last two layers both give a 1*1 size output, dose it correct? Why choosing such a small size comparing to the original structure?
-
This error is prompting when I run
python generate_tfrecord.py --csv_input=data/train_labels.csv --output_path=data/train.record
I am trying to build a custom object detector using Single Shot Det…
-
-
I have already worked with converting Yolov3 to .tmfile (uint8) and monitored the inference time on VIM3 Khadas board. Similarly, I have converted mobilenet.caffemodel model into uint8 format.
For i…
-
cannot find load function for operator: Cast
Create graph failed
,errno: 0
-
Dear ChunML,
I tested your code ssd-tf2. I look good job.
But, i need deploy on Coral TPU with small memory. So, i want decrease size model. Can you add option SSD with MOBILENET backbone?
Thank y…
-
I tried to run the train using mobilenet implementation. I modified the nets_factory script by adding ssd_300_mobilenet2 to the dictionaries but I get the following error
**File "train.py", line 6,…
-
Hi
I have setup and installed `cuda 11.6` on a `Tesla T4 GPU` on a AWS machine. Below is the screenshot of `nvidia smi`
![Capture](https://user-images.githubusercontent.com/20199336/227714241-3…
-
I followed all the steps, you have mentioned for converting TensorFlow SSD model v2 1.14 into TensorRT bin.
I updated the buildengine.py file and graphsurgeon converter for new Jetpack support.
…
-
SSD-MobileNetV1 | ssd_mobilenet_v1_10.onnx | 10 | reshape: Wrong number of elements for reshape | https://github.com/onnx/models/blob/main/vision/object_detection_segmentation/ssd-mobilenetv1/model/…