zhihuacc

Why get_special_tokens_mask appending a [1] at the end while build_inputs_with_special_tokens does not append a [SEP] at the end for a single input sequence ?

3

Hi, I found in this line[[build_inputs_with_special_tokens](https://github.com/salesforce/ALBEF/blob/75376bee33df87af9c206b4afb53c876927e7b2b/models/tokenization_bert.py#L294)](url) the returned list is appended a [1] at the end for a single input sequence, while the returned list [[here](https://github.com/salesforce/ALBEF/blob/75376bee33df87af9c206b4afb53c876927e7b2b/models/tokenization_bert.py#L262)](url) is NOT appended a...

I'm not sure but I think including test set to build vocabulary is spilling information from test set to training set. If using test set was OK, then why using...

Hi, I'm wondering if you used x_extractor = x_detector intending to create a new model. But I think this simple equal assignment does not create a new model. The two...

bgp_unnumbered_test and route_server_ipv4_v6_test failed due to IPv6 is not enabled in test container un_g1

3

I run test locally following [here](https://github.com/osrg/gobgp/tree/master/test/scenario_test) but got cases failed as below. Seems it because ipv6 is not enabled in some docker containers like un_g1. I login the container un_g1...

Lambdarank does not work on MSLR-WEB10K

1

Hi, I run the code on MSLR-WEB10K and found that ranknet/default and ranknet/factor can converge on MSLR-WEB10K. E.g., NDCG@5 can increase from 0.3 to higher than 0.4. However the lambdarank/default...