Loading
0 Follower
0 Following
darthgera123
Pulkit Gera

Location

Hyderabad, IN

Badges

3
2
2

Connect

Activity

Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Mon
Wed
Fri

Ratings Progression

Loading...

Challenge Categories

Loading...

Challenges Entered

Measure sample efficiency and generalization in reinforcement learning using procedurally generated environments

Latest submissions

No submissions made in this challenge.

A benchmark for image-based food recognition

Latest submissions

See All
failed 60620

5 Problems 15 Days. Can you solve it all?

Latest submissions

See All
graded 74284
graded 74283
graded 74258

Recognise Handwritten Digits

Latest submissions

No submissions made in this challenge.

Online News Prediction

Latest submissions

See All
graded 60330
graded 60329

Crowdsourced Map Land Cover Prediction

Latest submissions

See All
graded 67492
graded 60084

Latest submissions

No submissions made in this challenge.

5 Problems 15 Days. Can you solve it all?

Latest submissions

See All
graded 67356
graded 67322
graded 67310

5 PROBLEMS 3 WEEKS. CAN YOU SOLVE THEM ALL?

Latest submissions

No submissions made in this challenge.

Predict Power Consumption

Latest submissions

See All
graded 67496

Predict Wine Quality

Latest submissions

See All
graded 67498

Student Evaluation

Latest submissions

See All
graded 67497

Predict if an online advertisement will be clicked or not

Latest submissions

See All
graded 67499
Participant Rating
Participant Rating

MASKD

MMdetection unable to form final test file

Almost 4 years ago

I am trying to use MMdetection 2.0 for MASKD object detection. However, I am facing difficulty in creating the test file.
Here is the code that I have written

from mmdet.datasets import build_dataloader
cfg.data.test.test_mode = True
distributed = False
val_dataset = build_dataset(cfg.data.val)
data_loader = build_dataloader(
    val_dataset,
    samples_per_gpu=1,
    workers_per_gpu=1,
    dist=distributed,
    shuffle=False)
from mmcv.runner import load_checkpoint
from mmcv.parallel import MMDataParallel, MMDistributedDataParallel
from mmdet.apis import single_gpu_test
device = torch.device('cuda') if torch.cuda.is_available() else torch.device('cpu')

model = build_detector(cfg.model, train_cfg=None, test_cfg=cfg.test_cfg)
checkpoint = load_checkpoint(model, WEIGHTS_FILE, map_location='cpu')

model.CLASSES = dataset.CLASSES

model = MMDataParallel(model, device_ids=[0])
outputs = single_gpu_test(model, data_loader, False, None, 0.5)
val_dataset.format_results(outputs)

However I get the following output

---------------------------------------------------------------------------
IndexError                                Traceback (most recent call last)
<ipython-input-122-4158962a1af4> in <module>()
----> 1 val_dataset.format_results(outputs)

2 frames
/content/mmdetection/mmdet/datasets/coco.py in format_results(self, results, jsonfile_prefix, **kwargs)
    359         else:
    360             tmp_dir = None
--> 361         result_files = self.results2json(results, jsonfile_prefix)
    362         return result_files, tmp_dir
    363 

/content/mmdetection/mmdet/datasets/coco.py in results2json(self, results, outfile_prefix)
    291         result_files = dict()
    292         if isinstance(results[0], list):
--> 293             json_results = self._det2json(results)
    294             result_files['bbox'] = f'{outfile_prefix}.bbox.json'
    295             result_files['proposal'] = f'{outfile_prefix}.bbox.json'

/content/mmdetection/mmdet/datasets/coco.py in _det2json(self, results)
    228                     data['bbox'] = self.xyxy2xywh(bboxes[i])
    229                     data['score'] = float(bboxes[i][4])
--> 230                     data['category_id'] = self.cat_ids[label]
    231                     json_results.append(data)
    232         return json_results

IndexError: list index out of range

I guess I am unable to get the category_id but I cant find how to fix that.
Please help

NeurIPS 2020: Procgen Competition

Team Up for the challenge

About 4 years ago

Hi
I am a newcomer and I have done some small projects in Deep RL before. If anyone is interested in teaming up DM at testandplayalltime@gmail.com.

MINILEAVES

Train and testSet Size is Different

About 4 years ago

Hi bhavesh thanks for pointing that out, changes have been made

Starter Notebook

About 4 years ago

Hi thanks for sharing. we more than welcome community contributions

FOODC

Need to download datasets is necessary? or else any other way

About 4 years ago

hey you can mount your drive and save it there. you can then load from there at your convinience

DIBRD

Baseline - DIBRD

About 4 years ago

Hi thanks for pointing out. we have made the changes

Food Recognition Challenge

Approach to solving

About 4 years ago

Hi
Currently im going over the model zoo provided by mmdetection and wanted to ask what are some metrics to decide a model. Also what could be some hyper parameters(pre and post) which could be used to improve the score?

SPCRT

A way to improve Decision Tree

About 4 years ago

Hi
Yes secondary score for the challenge is Root Mean Square error.

Computer Graphics Researcher at CVIT,IIIT Hyderabad