-
Notifications
You must be signed in to change notification settings - Fork 35
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Support loading models by name (#38)
Co-authored-by: Matthew Tamayo-Rios <[email protected]>
- Loading branch information
Showing
4 changed files
with
111 additions
and
1 deletion.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
File renamed without changes.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,93 @@ | ||
package wasi:nn | ||
|
||
world inference { | ||
import tensor | ||
import graph | ||
import execution | ||
import errors | ||
} | ||
|
||
interface tensor { | ||
type tensor-dimensions = list<u32> | ||
type tensor-data = list<u8> | ||
|
||
enum tensor-type { | ||
fp16, | ||
fp32, | ||
bf16, | ||
up8, | ||
ip32 | ||
} | ||
|
||
record tensor { | ||
// Describe the size of the tensor (e.g., 2x2x2x2 -> [2, 2, 2, 2]). To represent a tensor | ||
// containing a single value, use `[1]` for the tensor dimensions. | ||
dimensions: tensor-dimensions, | ||
|
||
// Describe the type of element in the tensor (e.g., f32). | ||
tensor-type: tensor-type, | ||
|
||
// Contains the tensor data. | ||
data: tensor-data, | ||
} | ||
} | ||
|
||
interface graph { | ||
use errors.{error} | ||
type graph-builder = list<u8> | ||
type graph-builder-array = list<graph-builder> | ||
use tensor.{tensor} | ||
|
||
type graph = u32 | ||
|
||
enum graph-encoding { | ||
openvino, | ||
onnx, | ||
tensorflow, | ||
pytorch, | ||
tensorflowlite, | ||
autodetect, | ||
} | ||
|
||
enum execution-target { | ||
cpu, | ||
gpu, | ||
tpu | ||
} | ||
|
||
load: func(builder: graph-builder-array, encoding: graph-encoding, target: execution-target) -> result<graph, error> | ||
load-named-model: func(name: string) -> result<graph, error> | ||
} | ||
|
||
interface execution { | ||
use errors.{error} | ||
use tensor.{tensor, tensor-data} | ||
use graph.{graph} | ||
|
||
type graph-execution-context = u32 | ||
init-execution-context: func(graph: graph) -> result<graph-execution-context, error> | ||
set-input: func(ctx: graph-execution-context, index: u32, tensor: tensor) -> result<_, error> | ||
set-input-by-name: func(ctx: graph-execution-context, name: string, tensor: tensor) -> result<_, error> | ||
compute: func(ctx: graph-execution-context) -> result<_, error> | ||
get-output: func(ctx: graph-execution-context, index: u32) -> result<list<tensor-data>, error> | ||
eval: func(tensors: list<tensor>) -> result<list<tensor>, error> | ||
|
||
} | ||
|
||
interface errors { | ||
enum error { | ||
// Caller module passed an invalid argument. | ||
invalid-argument, | ||
// Invalid encoding. | ||
invalid-encoding, | ||
busy, | ||
// Runtime Error. | ||
runtime-error, | ||
// Unsupported operation | ||
unsupported-operation, | ||
// Model too large | ||
model-too-large, | ||
// Model not found | ||
model-not-found | ||
} | ||
} |