seanZhuh

Hi, thanks for your interest. A naive way would be setting batch_size to 1. If you want a inference script for demo show, you can first build the dataset, dataloader,...

``` import argparse import os.path as osp from typing import Sequence from mmcv import Config, DictAction from mmcv.utils import mkdir_or_exist from seqtr.apis import inference_model def parse_args(): parser = argparse.ArgumentParser(description="macvg-inference") parser.add_argument('config',...

``` EPS = 1e-2 def color_val_matplotlib(color): """Convert various input in BGR order to normalized RGB matplotlib color tuples, Args: color (:obj:`Color`/str/tuple/int/ndarray): Color inputs Returns: tuple[float]: A tuple of 3 normalized...

Yeah, but I'm preparing the job interview right now, maybe two months later I'll write the script. Please stay tuned, I'll try my best to make this repository as easy...

Hi, please upload the full traceback. Did it show the lan_enc.embedding.weight does not match the size, pre-training uses a larger word vocabulary, while fine-tuning only need a subset of this...

Did you use DDP during fine-tuning, if that's the case, the keys in pre-trained state_dict need to prepend "module." since we move it in line 58-59. By default we fine-tune...

During pretraining, we disable EMA and LSJ, so there is no ema_state_dict of the model. Just comment this line and loading the state_dict would be fine. ________________________________ 发件人: 陈崇彦 ***@***.***>...

Hi, you can import the function as below: import matplotlib as mpl import matplotlib.pyplot as plt import matplotlib.colors as mplc import matplotlib.figure as mplfigure import pycocotools.mask as maskUtils from matplotlib.patches...

aha, I didn't notice this, does it make a difference ?

1) Yes 2) No, the bbox format is in [x_tl, y_tl, x_br, y_br], tl for top-left, br for bottom-right