dataset not found - Githubissues

mjag7682 / NLP-of-StockTwits-data-for-predicting-stocks

32 stars 10 forks source link

dataset not found #1

Open lloscarll opened 3 years ago

lloscarll commented 3 years ago

hi mjag7682,

can i have complete dataset/data file. coz dataset on https://drive.google.com/file/d/1RtUEwOBWLTs1JaV6onuahpNPFLJX3lsb/view?usp=sharing not complete, there are some dataset/data not in the file, after i try running on google colab.

reuters_docs.txt,
StockTwits FAANG
combineV2_book10per.txt

regards, oscar

pclankswert commented 3 years ago

Oscar,

I wrote some python to create reuters_docs.txt that I am happy to share. May be missing more, but right now I am blocked on the following: --input_file="gs://capstonecs9-1/combinedPre" --output_dir="gs://capstonecs9-1/models_recc2" --albert_config_file="/content/drive/My Drive/Colab Notebooks/Capstone/albert_config.json"

Do you see a source for these three?

Pat

m1996 commented 2 years ago

Hi thank you so much for your research and sharing it, I am interested in the subject and I want to do some experiments on this topic. I need to download the StockTwits dataset for some specific stocks and specific period of time. Would you please share the code for data gathering with me? regards