stanford-navlab / deep_gnss Goto Github PK

Hi,
Uploading Pixel4_derived.csv…
ground_truth.csv

When I use "2020-06-11-US-MTV-1" (attached) collected by Pixel4 to train your neural network, I get the error "ValueError: No objects to concatenate"

For your reference:
`Processed file not found. Chunking and saving data for 2020-06-11-US-MTV-1/Pixel4
Traceback (most recent call last):
File "train_android.py", line 96, in main
dataset = Android_GNSS_Dataset(data_config)
File "/home/ling/Documents/GitHub/deep_gnss/src/correction_network/android_dataset.py", line 69, in init
self.chunk_n_save(_path, save=True)
File "/home/ling/Documents/GitHub/deep_gnss/src/correction_network/android_dataset.py", line 289, in chunk_n_save
chunk_df = pd.concat(chunk)
File "/home/ling/miniconda3/envs/deep_gnss/lib/python3.8/site-packages/pandas/core/reshape/concat.py", line 285, in concat
op = _Concatenator(
File "/home/ling/miniconda3/envs/deep_gnss/lib/python3.8/site-packages/pandas/core/reshape/concat.py", line 342, in init
raise ValueError("No objects to concatenate")
ValueError: No objects to concatenate

Set the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace.
`

Weights for trained networks can not be found

Hey there,
Thank you for sharing your Deep GNSS codes.
I tried to run it from scratch. According to your guide, I tried to install the conda environment using conda env create -f deep-gnss.yml. But there seems to be no such a file. Should I use the environment.yml instead?

Hello:
Recently, I read your article titled ‘Improving GNSS Positioning using Neural Network-based Corrections' which is a handsome work, so I also learned relevant codes. Meanwhile, I found that this article was quoted in the article 'Designing Deep Neural Networks for Sequential GNSS Positioning'. I was very interested in it and immediately learned it. Hence, I would like to consult you if the code of this article will be released in the near future. I really want to study the code of this paper. Looking forward to your reply, thank you very much!

Yours Sincerely!

Hello:
It is a really cool work to connect the work between GNSS and DNN. When i try to reproduce the results of the paper, I found that the link of weight file is invalid. could you send the file of weight file to me? Thank you!

Yours Sincerely!