diff options
author | Jon Bratseth <bratseth@gmail.com> | 2022-08-23 11:12:39 +0200 |
---|---|---|
committer | Jon Bratseth <bratseth@gmail.com> | 2022-08-23 11:12:39 +0200 |
commit | 207f2963125c9094d65f50f5ea41d98cf3ba8524 (patch) | |
tree | 2011e1e3cfddc4eed10411af02eb728526e12e31 /config-model/src/test/cfg | |
parent | 7e290cd7574f69071490dbfb78da9e2773a863e7 (diff) |
Simplify and generalize
Let embedder rewrites simply always change <x path="y" url="z">
to <xPath>y</xPath> <xUrl>z</xUrl>, as well as translating
the id attirebute to the corresponding path if on hosted.
This means that the current built-in embedder accepting
"vocab" and "model" is broken as these names are different
from the names in the config model, but this isn't documented yet
so should be okay.
The effect of this is that our built-in models can be
used in any embedder and config, and the embedder syntax
can be used in application specific embedders.
Diffstat (limited to 'config-model/src/test/cfg')
-rw-r--r-- | config-model/src/test/cfg/application/embed/services.xml | 4 | ||||
-rw-r--r-- | config-model/src/test/cfg/application/embed_generic/configdefinitions/sentence-embedder.def | 3 |
2 files changed, 4 insertions, 3 deletions
diff --git a/config-model/src/test/cfg/application/embed/services.xml b/config-model/src/test/cfg/application/embed/services.xml index f319d875ed9..9a05337f954 100644 --- a/config-model/src/test/cfg/application/embed/services.xml +++ b/config-model/src/test/cfg/application/embed/services.xml @@ -11,8 +11,8 @@ <embedder id="transformer" class="ai.vespa.embedding.BertBaseEmbedder"> <!-- model specifics --> - <model id="test-model-id" url="test-model-url"/> - <vocab path="files/vocab.txt"/> + <transformerModel id="test-model-id" url="test-model-url"/> + <tokenizerVocab path="files/vocab.txt"/> <!-- tunable parameters: number of threads etc --> <onnxIntraOpThreads>4</onnxIntraOpThreads> diff --git a/config-model/src/test/cfg/application/embed_generic/configdefinitions/sentence-embedder.def b/config-model/src/test/cfg/application/embed_generic/configdefinitions/sentence-embedder.def index f62e2019189..ac5c79d2714 100644 --- a/config-model/src/test/cfg/application/embed_generic/configdefinitions/sentence-embedder.def +++ b/config-model/src/test/cfg/application/embed_generic/configdefinitions/sentence-embedder.def @@ -4,7 +4,8 @@ package=ai.vespa.example.paragraph vocab path # Transformer model settings -model path +modelPath path +modelUrl string # Max length of token sequence model can handle transforerMaxTokens int default=128 |