We lifted 8 functions out of this paper's own repositories and ran 6 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| google/seqio | canonical | 6 of 8 |
| Function | Status | Where it lives |
|---|---|---|
| globally_sort_model_output | Ran | google/seqio/seqio/metrics.py code served (permissive licence) · get_code("ff070b6501ae20c1") |
| non_padding_position | Ran | google/seqio/seqio/feature_converters.py code served (permissive licence) · get_code("8dc0bb06136c4f18") |
| remove_padding_examples | Ran | google/seqio/seqio/metrics.py code served (permissive licence) · get_code("fe01b1b2c2661706") |
| skip_none_value_dict_factory | Ran | google/seqio/seqio/loggers.py code served (permissive licence) · get_code("306b1c630154c8e4") |
| tfexample_ragged_length_key | Ran | google/seqio/seqio/utils.py code served (permissive licence) · get_code("8a2972f5d6059698") |
| tfexample_sparse_indices_key | Ran | google/seqio/seqio/utils.py code served (permissive licence) · get_code("ca80aa652b23a246") |
| dict_to_tfexample | Not yet run | google/seqio/seqio/utils.py code served (permissive licence) · get_code("192f110808c15f37") |
| informative_upstream_callsites_from_frame | Not yet run | google/seqio/seqio/task_registry_provenance_tracking.py code served (permissive licence) · get_code("c1aa5409a61c75ba") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Recent neural network-based language models have benefited greatly from scaling up the size of training datasets and the number of parameters in the models themselves. Scaling can be complicated due to various factors including the need to distribute computation on supercomputer clusters (e.g., TPUs), prevent bottlenecks when infeeding data, and ensure reproducible results. In this work, we present two software libraries that ease these issues: $\texttt{t5x}$ simplifies the process of building and training large language models at scale while maintaining ease of use, and $\texttt{seqio}$ provides a task-based API for simple creation of fast and reproducible training data and evaluation pipelines. These open-source libraries have been used to train models with hundreds of billions of parameters on datasets with multiple terabytes of training data. Along with the libraries, we release configurations and instructions for T5-like encoder-decoder models as well as GPT-like decoder-only architectures. $\texttt{t5x}$ and $\texttt{seqio}$ are open source and available at https://github.com/google-research/t5x and https://github.com/google/seqio, respectively.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2203.17189")
get_code_for_paper("2203.17189")
have("2203.17189")
Connect an agent — have() is free.