terminator123

模型维度问题

5

gpt2模型config里面n_positions=513，会报 size mismatch for transformer.h.0.attn.bias: copying a param with shape torch.Size([1, 1, 512, 512]) from checkpoint, the shape in current model is torch.Size([1, 1, 513, 513]). 改成512后，如果use_gpt2=True，会报 size mismatch for...

预测问题

2

直接用CDial-GPT2_LCCC-base 预测 预测那部分代码有修改，不然跑不通 output = model(input_ids, token_type_ids=token_type_ids) logits = output.logits logits = logits[0, -1, :] / args.temperature 不管输入是什么，结果都如下 [12997, 7635, 12997, 7635, 12997, 12997, 12997, 12997, 7635, 12997, 7635, 12997,...

### Description when i use tst-decoder， i want to get each timestep's probablity, such as 'a' [0.1, 0.1, 0.2.....]。i tried to add log parameter in decode_hparams， it didn't work。 i...

when i run this mode, it throws the error that 'X_tst' 'Y_tst' not defined。 it should be 'x_tst' ,'y_tst' in cnn_train.py

paddle hub 如何 调用gpu

1

同时装了paddlepaddle和paddlepaddle-gpu， model = hub.Module(directory='./baidu_translate')，使用的是cpu，速度比较慢。 如果卸载paddlepaddle，只留paddlepaddle-gpu，hub会报错 File "/home/chenqun/.conda/envs/end2endtrans/lib/python3.9/site-packages/paddlehub/__init__.py", line 20, in _paddle_version = Version(paddle.__version__) AttributeError: module 'paddle' has no attribute '__version__'

[Question] how to merge the middle checkpoint file with lora

6

### Question i want to test the checkpoint-5000 in lora，when i ran python scrips/merge_lora_weights.py --model-path ./checkpoints/llava-v1.5-13b-lora --model-base lmsys/vicuna-13b-v1.5 --save-model-path ./checkpoints/merge it went wrong

wget https://msmarco.blob.core.windows.net/msmarcoranking/collection.tar.gz -O msmarco.tsv the path is not exist，do you know the new path?

多标签训练的时候 beam需要设置吗，我设置成非1，代码报错 candidate += [tgt_vocab.convertToLabels(s.tolist(), utils.EOS) for s in samples] 显示 s 是list对象，无法进行to list操作