fmy7834

Hello, I'm interested in your excellent work. I want to know **if there is difference between downloading openImages from 2017_11 folder or 2018_4 folder**? The download_images.py you provided only support...

4M version checkpoint

1

Is there any download link for 4M version checkpoint of Tag2Text, RAM and RAM++?

Evaluation Benchmarks

3

It's a great work. Could you release ImageNet-Multi and HICO?

How do you get thresholds for clip model in results of Table 3 ?

4

Hi, I want to know how do you get thresholds for clip model in results of Table 3 ? Is it the same way like you said in another issue?...

error: failed to get `numpy` as a dependency of package `tokenizers-python v0.10.3

1

When I installled package transformers, I met this error. Does anyone met this problem before? error: failed to get `numpy` as a dependency of package `tokenizers-python v0.10.3 (/tmp/pip-install-ulgbj6di/tokenizers_2c687e1980e640e483b11bc6ace87ebc)` Caused by:...

Hello~, I'm interested in your work and reproducing experiment results with your code. I find that **the mAP in ZSL setting on NUS-WIDE dataset after first-stage training** is **42.2** using...

Hello, I 'm interested in your excellent work. And I have two questions when re-produce your work. 1. In line 210 of lmpt/train.py: **class_weights = sfm(torch.from_numpy(np.asarray(mmcv.load(freq_file)['class_freq'])).to(torch.float32).cuda())**, Why apply softmax on...

[Questio why 'mm_vision_select_layer' == -2 in config ? n]

2

### Question In training scripts, 'mm_vision_select_layer' is set to be -2, which means the penultimate layer's output of CLIP vision encoder is used as image features. I wonder why not...