open-solution-mapping-challenge

open-solution-mapping-challenge copied to clipboard

File b'data/meta/metadata.csv' does not exist: b'data/meta/metadata.csv'

1

running python main.py prepare_masks and python main.py train --pipeline_name unet_weighted d， it threw out an error: FileNotFoundError: [Errno 2] File b'data/meta/metadata.csv' does not exist: b'data/meta/metadata.csv'，please tell me how to solve...

Angel0003

KeyError: "['file_path_mask_eroded_3'] not in index"

39

When running local pure python with `python main.py -- train_evaluate_predict --pipeline_name unet --chunk_size 5000` , the following error occurs, any help? > neptune: Executing in Offline Mode. neptune: Executing in...

XYAskWhy

Model overfitted or mapping challenge dataset not diverse? Poor performance with custom images

4

Excellent results can be achieved using the Developers' fully-trained model (found here: https://app.neptune.ml/neptune-ml/Mapping-Challenge/files) to predict on the test_images set provided as part of the mapping challenge. However, over the past...

carbonox-infernox

Modify input image size for training/inference

2

Good day, I would like to train on my own map images. However I noticed that the input image size is 300 * 300, following the crowdAI (AIcrowd) Mapping Challenge...

cvKDean

weighted loss

1

Hello, I am interested in loss function combined with size and distance but I have no idea how it works. Can you give me more information（related paper, reference and so...

SonwYang

Pretrained weights

6

Has anyone uploaded pretrained models of the same? I dont have mutiple GPUs to train so its taking a lot of time to train!

nishant-neo

How would you go about training this data on a larger/more diverse data set? I've looked into some open source data from SpaceNet, and was wondering if I could use...

DanielHyunilKim

when training finishes it hits the memory error

2

Running train crashes when the pipelines are collating results from running transform on the entire dataset.

jakubczakon

sorry , i can not find the 50000 examples subset of the dataset, which should be downloaded online or generate through all the dataset?

qintaohu

UnicodeDecodeError: 'gbk' codec can't decode byte 0xa5 in position 404: illegal multibyte sequence

8

Hi, I'm sorry to bother you. There are some problems in the replication experiment, and the description of the problem is >reader.py，line 178, in update_raw data = self.stream.read(size) UnicodeDecodeError: 'gbk'...

pnzsl