zhouxingguang

icon location denoise, enhancement, superresolution, isp, high dynamic range, quality assessment, recognition, retrieval.

Results 1 comments of zhouxingguang

After I completed the deployment and executed the demo:

change the bert config in modeling_hformer.py bert = 'xxx/bert-base-uncased' print('local bert:', bert) encoder_config = BertConfig.from_pretrained(bert)