salman-h-khan/PL-ZSD_Release
This repository provides training and evaluation code for paper titled "Polar Loss for Zero-Shot Object Detection." (Arxiv version) and "Improved Visual-Semantic Alignment for Zero-Shot Object Detection" (accepted in AAAI 2020)
PythonMIT
Issues
- 0
Question about vocabulary_w2v.txt and word_w2v.txt
#29 opened by JKZhan - 0
How was the file word_w2v.txt generated?
#28 opened by xkhnhms - 0
How to predict objects of a custom category?
#27 opened by xkhnhms - 0
- 4
- 0
Train or test with cpu
#25 opened by xiaofeng-c - 0
One query regarding docker image
#24 opened by SUSHIL1312 - 5
How to generate a word_w2v.txt file?
#11 opened by bassyess - 0
Generate vocabulary file for Pascal VOC dataset
#23 opened by Rahn80643 - 1
Problem in running gzsd.py file
#22 opened by SUSHIL1312 - 4
- 1
evaluation of accuracy
#19 opened by aoko-sys - 0
Custom dataset
#21 opened by manyaafonso - 0
Error while running gzsd.py
#20 opened by SodaCoder - 3
- 0
Error when run training code: tensorflow/core/grappler/optimizers/dependency_optimizer.cc:666] Iteration = 0, topological sort failed with message: The graph couldn't be sorted in topological order.
#18 opened by EasitMickly - 1
Error when loading weights, I want to train it on another dataset, and the num_classes is not 65
#14 opened by Johumliu - 2
for GZSD,when training,classification (Concatenate) (None, None, 65) ? why not (None, None, 80) ??
#15 opened by lixiangamy - 1
The num_classes during training and testing
#10 opened by weiyaowu-md - 1
A warning during training
#9 opened by TessGreymane - 1
train/val curve
#13 opened by lsq357 - 1
Have you tried applying the politary loss function to the traditional classification problem instead of the zero shot object classification problem? I would like to know whether the effect will be better or not.
#5 opened by Bigwode - 2
Can you share train code? I want to know how to train this model. Thank you!
#6 opened by TessGreymane - 1
- 1
- 1
can not find keras_resnet.models
#3 opened by yangdb - 1
train code?
#2 opened by asdfqwer2015 - 2
can you share the training code
#4 opened by TessGreymane