Closed vincentlux closed 4 years ago
Hi @vincentlux, thanks for reporting this! We had an issue on AWS that we've just fixed. The config is available now.
This bug is happening again OSError: Couldn't reach server at 'https://nlp-architect-data.s3-us-west-2.amazonaws.com/models/transformers/bert-base-uncased.json' to download pretrained model configuration file.
Please use the recent release from the master branch. We had to update all our links. Thanks
From: Pankesh Bamotra notifications@github.com Sent: Friday, October 16, 2020 2:59:34 AM To: NervanaSystems/nlp-architect nlp-architect@noreply.github.com Cc: Izsak, Peter peter.izsak@intel.com; State change state_change@noreply.github.com Subject: Re: [NervanaSystems/nlp-architect] bug: [quant bert config json not available] (#163)
This bug is happening again OSError: Couldn't reach server at 'https://nlp-architect-data.s3-us-west-2.amazonaws.com/models/transformers/bert-base-uncased.json' to download pretrained model configuration file.
Intel Israel (74) Limited
This e-mail and any attachments may contain confidential material for the sole use of the intended recipient(s). Any review or distribution by others is strictly prohibited. If you are not the intended recipient, please contact the sender and delete all copies.
Describe the bug It appears that https://nlp-architect-data.s3-us-west-2.amazonaws.com/models/transformers/bert-base-uncased.json link from https://github.com/NervanaSystems/nlp-architect/blob/70cb5f3fc1923efeccf98263dd80896bb058f94b/nlp_architect/models/transformers/quantized_bert.py#L59 is dead and a config.json file cannot be loaded properly. May I ask if the link can be updated? Thanks!