Yijia Shao

> Hi @ink-pad , > > Thanks for open-sourcing the code. I have a question, Do we need to change the line [118 in tabformer_bert.py](https://github.com/IBM/TabFormer/blob/main/models/tabformer_bert.py#L118), from `outputs = (prediction_scores,) +...

@googlebot I signed it!

Currently, we provide a profiling function after processing a topic. You can call [`summary()`](https://github.com/stanford-oval/storm/blob/5d21c393e64d8c877ae6a9db3a0df0a72a926b33/src/interface.py#L417) to see the time, model usage for each step. Run [`reset()`](https://github.com/stanford-oval/storm/blob/main/src/interface.py#L432) before starting a new topic.

I upgrade the flutter stable channel to 2.5.0 and the same problem occurs. Besides, I try some other system call using Process.run() and found the results are different in `flutter...

Yes, I think the code itself doesn't have problem and If you open the app with `flutter run`, there shouldn't be any problem (at least in my case). So is...

I personally agree with you. > I wonder if this has to do with the PATH environment variable or some other environmental change that causes different results. I was hoping...

Hi @xuxiangwork , thanks for reading our codebase carefully! Yes, the current [`StormArticleGenerationModule`](https://github.com/stanford-oval/storm/blob/5d21c393e64d8c877ae6a9db3a0df0a72a926b33/src/storm_wiki/modules/article_generation.py#L14) is implemented in a way that does not require the model to follow the subsection/subsubsection/... when writing...

> I suspect that constraining LM to adhere to predetermined sections may be a challenge that requires some post-processing. I agree. So the default behavior is using the outline to...

Could you try to change `--output-dir` to something like `results`? Based on the log, `../results\\The_promise_and_technical_difficulties_of_SSTO_vehicles\\raw_search_results.json` looks incorrect. We set the default value of `--output-dir` as `../results` in the scripts but...

The error seems to be hitting the free tier limit.