Shuyue Jia (Bruce)

Is there a method to output the used FLOPs of the DVQA model? Thanks in advance.

How to implement batch training?

24

Hi, Thanks a lot for your great work! I really appreciate it! Is there any way to implement batch training? Thanks in advance. Best, Shuyue

LIVE Database - InvalidArgumentError: Incompatible shapes: [1,512,640,1] vs. [1,512,768,1]

1

**Describe the bug** Incompatible shapes of distorted image and reference image e.g., [1,512,640,1] vs. [1,512,768,1] **Additional context** ```python Generating splits...: 0%| | 0/1 [00:00

Inquire about the database split

2

Dear Wei, Thanks a lot for your excellent summary! It helps a lot in the field. Wei, I can see that most SOTA works utilized a strategy of 80% of...

In the codes, an adequacy model is applied to score and rank the generated sentences. (1) `def _formal_to_casual()`: https://github.com/PrithivirajDamodaran/Styleformer/blob/main/styleformer/styleformer.py#L69-L95 (2) `def _casual_to_formal()`: https://github.com/PrithivirajDamodaran/Styleformer/blob/main/styleformer/styleformer.py#L97-L123 May I know why it is not...

Add `torch_dtype=torch.float32` because "[RuntimeError: "addmm_impl_cpu_" not implemented for 'Half'](https://stackoverflow.com/questions/73530569/pytorch-matmul-runtimeerror-addmm-impl-cpu-not-implemented-for-half)" ```python self.ctf_model = AutoModelForSeq2SeqLM.from_pretrained(ctf_model_tag, use_auth_token=False, torch_dtype=torch.float32) self.ftc_model = AutoModelForSeq2SeqLM.from_pretrained(ftc_model_tag, use_auth_token=False, torch_dtype=torch.float32) self.atp_model = AutoModelForSeq2SeqLM.from_pretrained(atp_model_tag, use_auth_token=False, torch_dtype=torch.float32) self.pta_model = AutoModelForSeq2SeqLM.from_pretrained(pta_model_tag, use_auth_token=False, torch_dtype=torch.float32)...

Error triggered!

1

Thank you very much! Shuyue Feb. 21st, 2023

Update README.md

1

Add double quotes to the `huggingface_hub[hf_transfer]` and `ckpt-0/*`. ```shell git clone https://github.com/xai-org/grok-1.git && cd grok-1 pip install "huggingface_hub[hf_transfer]" huggingface-cli download xai-org/grok-1 --repo-type model --include "ckpt-0/*" --local-dir checkpoints --local-dir-use-symlinks False ```

Correct a minor numerical mistake in the `train.jsonl`

2

In the `train.jsonl` file: **The final numerical answer of one example is wrong**: **Original**: ```json {"question": "In a yard, the number of tanks is five times the number of trucks....