Upload folder using huggingface_hub
Browse files- audio_3000.mlmodelc/analytics/coremldata.bin +3 -0
- audio_3000.mlmodelc/coremldata.bin +3 -0
- audio_3000.mlmodelc/metadata.json +81 -0
- audio_3000.mlmodelc/model.mil +0 -0
- audio_3000.mlmodelc/weights/weight.bin +3 -0
- audio_6000.mlmodelc/analytics/coremldata.bin +3 -0
- audio_6000.mlmodelc/coremldata.bin +3 -0
- audio_6000.mlmodelc/metadata.json +81 -0
- audio_6000.mlmodelc/model.mil +0 -0
- audio_6000.mlmodelc/weights/weight.bin +3 -0
- decoder.mlmodelc/analytics/coremldata.bin +3 -0
- decoder.mlmodelc/coremldata.bin +3 -0
- decoder.mlmodelc/metadata.json +310 -0
- decoder.mlmodelc/model.mil +0 -0
- decoder.mlmodelc/weights/weight.bin +3 -0
- embeddings.mlmodelc/analytics/coremldata.bin +3 -0
- embeddings.mlmodelc/coremldata.bin +3 -0
- embeddings.mlmodelc/metadata.json +67 -0
- embeddings.mlmodelc/model.mil +21 -0
- embeddings.mlmodelc/weights/weight.bin +3 -0
- lm_head.mlmodelc/analytics/coremldata.bin +3 -0
- lm_head.mlmodelc/coremldata.bin +3 -0
- lm_head.mlmodelc/metadata.json +79 -0
- lm_head.mlmodelc/model.mil +348 -0
- lm_head.mlmodelc/weights/weight.bin +3 -0
- manifest.json +49 -0
audio_3000.mlmodelc/analytics/coremldata.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d39ce3cf8f0afc6ebdbaf28b432473d6495c371f13270d265eebb194f109d81
|
| 3 |
+
size 243
|
audio_3000.mlmodelc/coremldata.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a2292e34e32a1f5a144d787ec13e8e5004ed293022400e7dbb6ea24e2a63d63f
|
| 3 |
+
size 425
|
audio_3000.mlmodelc/metadata.json
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"metadataOutputVersion" : "3.0",
|
| 4 |
+
"storagePrecision" : "Float16",
|
| 5 |
+
"outputSchema" : [
|
| 6 |
+
{
|
| 7 |
+
"hasShapeFlexibility" : "0",
|
| 8 |
+
"isOptional" : "0",
|
| 9 |
+
"dataType" : "Float32",
|
| 10 |
+
"formattedType" : "MultiArray (Float32 1 × 390 × 1024)",
|
| 11 |
+
"shortDescription" : "",
|
| 12 |
+
"shape" : "[1, 390, 1024]",
|
| 13 |
+
"name" : "audio_embeddings",
|
| 14 |
+
"type" : "MultiArray"
|
| 15 |
+
}
|
| 16 |
+
],
|
| 17 |
+
"modelParameters" : [
|
| 18 |
+
|
| 19 |
+
],
|
| 20 |
+
"specificationVersion" : 9,
|
| 21 |
+
"mlProgramOperationTypeHistogram" : {
|
| 22 |
+
"Ios18.mul" : 18,
|
| 23 |
+
"Ios18.softmax" : 18,
|
| 24 |
+
"Ios18.linear" : 111,
|
| 25 |
+
"Ios18.gelu" : 22,
|
| 26 |
+
"Ios18.layerNorm" : 37,
|
| 27 |
+
"Ios18.transpose" : 74,
|
| 28 |
+
"Ios18.matmul" : 36,
|
| 29 |
+
"Ios18.cast" : 3,
|
| 30 |
+
"Ios18.conv" : 3,
|
| 31 |
+
"Ios18.reshape" : 76,
|
| 32 |
+
"Ios18.add" : 55
|
| 33 |
+
},
|
| 34 |
+
"computePrecision" : "Mixed (Float16, Float32, Int32)",
|
| 35 |
+
"isUpdatable" : "0",
|
| 36 |
+
"stateSchema" : [
|
| 37 |
+
|
| 38 |
+
],
|
| 39 |
+
"availability" : {
|
| 40 |
+
"macOS" : "15.0",
|
| 41 |
+
"tvOS" : "18.0",
|
| 42 |
+
"visionOS" : "2.0",
|
| 43 |
+
"watchOS" : "11.0",
|
| 44 |
+
"iOS" : "18.0",
|
| 45 |
+
"macCatalyst" : "18.0"
|
| 46 |
+
},
|
| 47 |
+
"modelType" : {
|
| 48 |
+
"name" : "MLModelType_mlProgram"
|
| 49 |
+
},
|
| 50 |
+
"userDefinedMetadata" : {
|
| 51 |
+
"com.github.apple.coremltools.conversion_date" : "2026-07-15",
|
| 52 |
+
"com.github.apple.coremltools.source" : "torch==2.7.0",
|
| 53 |
+
"com.github.apple.coremltools.version" : "9.0",
|
| 54 |
+
"com.github.apple.coremltools.source_dialect" : "TorchScript"
|
| 55 |
+
},
|
| 56 |
+
"inputSchema" : [
|
| 57 |
+
{
|
| 58 |
+
"hasShapeFlexibility" : "0",
|
| 59 |
+
"isOptional" : "0",
|
| 60 |
+
"dataType" : "Float32",
|
| 61 |
+
"formattedType" : "MultiArray (Float32 1 × 128 × 3000)",
|
| 62 |
+
"shortDescription" : "",
|
| 63 |
+
"shape" : "[1, 128, 3000]",
|
| 64 |
+
"name" : "mel_features",
|
| 65 |
+
"type" : "MultiArray"
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"hasShapeFlexibility" : "0",
|
| 69 |
+
"isOptional" : "0",
|
| 70 |
+
"dataType" : "Float32",
|
| 71 |
+
"formattedType" : "MultiArray (Float32 1 × 1 × 390 × 390)",
|
| 72 |
+
"shortDescription" : "",
|
| 73 |
+
"shape" : "[1, 1, 390, 390]",
|
| 74 |
+
"name" : "attention_mask",
|
| 75 |
+
"type" : "MultiArray"
|
| 76 |
+
}
|
| 77 |
+
],
|
| 78 |
+
"generatedClassName" : "audio_3000",
|
| 79 |
+
"method" : "predict"
|
| 80 |
+
}
|
| 81 |
+
]
|
audio_3000.mlmodelc/model.mil
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
audio_3000.mlmodelc/weights/weight.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e48b1c016174b3fc76c085585cd8c739cf94eb72587447e38e78e58bf88d5469
|
| 3 |
+
size 372796928
|
audio_6000.mlmodelc/analytics/coremldata.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51e03083424e2d2b775e5d7427132b5ea48c2bddb173af27cf93416605059da8
|
| 3 |
+
size 243
|
audio_6000.mlmodelc/coremldata.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:534dd47984a058d34ee122f30899b9a1cc8e0e3d0ce50e4c75d55ab6f3bdd805
|
| 3 |
+
size 425
|
audio_6000.mlmodelc/metadata.json
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"metadataOutputVersion" : "3.0",
|
| 4 |
+
"storagePrecision" : "Float16",
|
| 5 |
+
"outputSchema" : [
|
| 6 |
+
{
|
| 7 |
+
"hasShapeFlexibility" : "0",
|
| 8 |
+
"isOptional" : "0",
|
| 9 |
+
"dataType" : "Float32",
|
| 10 |
+
"formattedType" : "MultiArray (Float32 1 × 780 × 1024)",
|
| 11 |
+
"shortDescription" : "",
|
| 12 |
+
"shape" : "[1, 780, 1024]",
|
| 13 |
+
"name" : "audio_embeddings",
|
| 14 |
+
"type" : "MultiArray"
|
| 15 |
+
}
|
| 16 |
+
],
|
| 17 |
+
"modelParameters" : [
|
| 18 |
+
|
| 19 |
+
],
|
| 20 |
+
"specificationVersion" : 9,
|
| 21 |
+
"mlProgramOperationTypeHistogram" : {
|
| 22 |
+
"Ios18.mul" : 18,
|
| 23 |
+
"Ios18.softmax" : 18,
|
| 24 |
+
"Ios18.linear" : 111,
|
| 25 |
+
"Ios18.gelu" : 22,
|
| 26 |
+
"Ios18.layerNorm" : 37,
|
| 27 |
+
"Ios18.transpose" : 74,
|
| 28 |
+
"Ios18.matmul" : 36,
|
| 29 |
+
"Ios18.cast" : 3,
|
| 30 |
+
"Ios18.conv" : 3,
|
| 31 |
+
"Ios18.reshape" : 76,
|
| 32 |
+
"Ios18.add" : 55
|
| 33 |
+
},
|
| 34 |
+
"computePrecision" : "Mixed (Float16, Float32, Int32)",
|
| 35 |
+
"isUpdatable" : "0",
|
| 36 |
+
"stateSchema" : [
|
| 37 |
+
|
| 38 |
+
],
|
| 39 |
+
"availability" : {
|
| 40 |
+
"macOS" : "15.0",
|
| 41 |
+
"tvOS" : "18.0",
|
| 42 |
+
"visionOS" : "2.0",
|
| 43 |
+
"watchOS" : "11.0",
|
| 44 |
+
"iOS" : "18.0",
|
| 45 |
+
"macCatalyst" : "18.0"
|
| 46 |
+
},
|
| 47 |
+
"modelType" : {
|
| 48 |
+
"name" : "MLModelType_mlProgram"
|
| 49 |
+
},
|
| 50 |
+
"userDefinedMetadata" : {
|
| 51 |
+
"com.github.apple.coremltools.conversion_date" : "2026-07-15",
|
| 52 |
+
"com.github.apple.coremltools.source" : "torch==2.7.0",
|
| 53 |
+
"com.github.apple.coremltools.version" : "9.0",
|
| 54 |
+
"com.github.apple.coremltools.source_dialect" : "TorchScript"
|
| 55 |
+
},
|
| 56 |
+
"inputSchema" : [
|
| 57 |
+
{
|
| 58 |
+
"hasShapeFlexibility" : "0",
|
| 59 |
+
"isOptional" : "0",
|
| 60 |
+
"dataType" : "Float32",
|
| 61 |
+
"formattedType" : "MultiArray (Float32 1 × 128 × 6000)",
|
| 62 |
+
"shortDescription" : "",
|
| 63 |
+
"shape" : "[1, 128, 6000]",
|
| 64 |
+
"name" : "mel_features",
|
| 65 |
+
"type" : "MultiArray"
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"hasShapeFlexibility" : "0",
|
| 69 |
+
"isOptional" : "0",
|
| 70 |
+
"dataType" : "Float32",
|
| 71 |
+
"formattedType" : "MultiArray (Float32 1 × 1 × 780 × 780)",
|
| 72 |
+
"shortDescription" : "",
|
| 73 |
+
"shape" : "[1, 1, 780, 780]",
|
| 74 |
+
"name" : "attention_mask",
|
| 75 |
+
"type" : "MultiArray"
|
| 76 |
+
}
|
| 77 |
+
],
|
| 78 |
+
"generatedClassName" : "audio_6000",
|
| 79 |
+
"method" : "predict"
|
| 80 |
+
}
|
| 81 |
+
]
|
audio_6000.mlmodelc/model.mil
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
audio_6000.mlmodelc/weights/weight.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e48b1c016174b3fc76c085585cd8c739cf94eb72587447e38e78e58bf88d5469
|
| 3 |
+
size 372796928
|
decoder.mlmodelc/analytics/coremldata.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5e5f46f066b2748b71b8f4b1a2e454ef1ebcfcc32e5d1d23af15379d53bd9e1b
|
| 3 |
+
size 243
|
decoder.mlmodelc/coremldata.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4449840be7138b0c38f1b319e0439e24bb741ff57ec574be37f9843758e58901
|
| 3 |
+
size 542
|
decoder.mlmodelc/metadata.json
ADDED
|
@@ -0,0 +1,310 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"metadataOutputVersion" : "3.0",
|
| 4 |
+
"storagePrecision" : "Mixed (Float16, Palettized (13 bits), Palettized (14 bits), Palettized (15 bits), UInt6)",
|
| 5 |
+
"outputSchema" : [
|
| 6 |
+
{
|
| 7 |
+
"hasShapeFlexibility" : "0",
|
| 8 |
+
"isOptional" : "0",
|
| 9 |
+
"dataType" : "Float16",
|
| 10 |
+
"formattedType" : "MultiArray (Float16 1 × 1 × 1024)",
|
| 11 |
+
"shortDescription" : "",
|
| 12 |
+
"shape" : "[1, 1, 1024]",
|
| 13 |
+
"name" : "output_hidden_states",
|
| 14 |
+
"type" : "MultiArray"
|
| 15 |
+
}
|
| 16 |
+
],
|
| 17 |
+
"modelParameters" : [
|
| 18 |
+
|
| 19 |
+
],
|
| 20 |
+
"specificationVersion" : 9,
|
| 21 |
+
"functions" : [
|
| 22 |
+
{
|
| 23 |
+
"inputSchema" : [
|
| 24 |
+
{
|
| 25 |
+
"hasShapeFlexibility" : "0",
|
| 26 |
+
"isOptional" : "0",
|
| 27 |
+
"dataType" : "Float16",
|
| 28 |
+
"formattedType" : "MultiArray (Float16 1 × 1 × 1024)",
|
| 29 |
+
"shortDescription" : "",
|
| 30 |
+
"shape" : "[1, 1, 1024]",
|
| 31 |
+
"name" : "hidden_states",
|
| 32 |
+
"type" : "MultiArray"
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"hasShapeFlexibility" : "0",
|
| 36 |
+
"isOptional" : "0",
|
| 37 |
+
"dataType" : "Int32",
|
| 38 |
+
"formattedType" : "MultiArray (Int32 1)",
|
| 39 |
+
"shortDescription" : "",
|
| 40 |
+
"shape" : "[1]",
|
| 41 |
+
"name" : "position_ids",
|
| 42 |
+
"type" : "MultiArray"
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"hasShapeFlexibility" : "0",
|
| 46 |
+
"isOptional" : "0",
|
| 47 |
+
"dataType" : "Float16",
|
| 48 |
+
"formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1536)",
|
| 49 |
+
"shortDescription" : "",
|
| 50 |
+
"shape" : "[1, 1, 1, 1536]",
|
| 51 |
+
"name" : "causal_mask",
|
| 52 |
+
"type" : "MultiArray"
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"hasShapeFlexibility" : "0",
|
| 56 |
+
"isOptional" : "0",
|
| 57 |
+
"dataType" : "Int32",
|
| 58 |
+
"formattedType" : "MultiArray (Int32 1)",
|
| 59 |
+
"shortDescription" : "",
|
| 60 |
+
"shape" : "[1]",
|
| 61 |
+
"name" : "current_pos",
|
| 62 |
+
"type" : "MultiArray"
|
| 63 |
+
}
|
| 64 |
+
],
|
| 65 |
+
"computePrecision" : "Mixed (Float16, Int16, Int32, UInt16)",
|
| 66 |
+
"storagePrecision" : "Mixed (Float16, Palettized (13 bits), Palettized (14 bits), Palettized (15 bits), UInt6)",
|
| 67 |
+
"stateSchema" : [
|
| 68 |
+
{
|
| 69 |
+
"dataType" : "Float16",
|
| 70 |
+
"isOptional" : "0",
|
| 71 |
+
"formattedType" : "State (Float16 56 × 8 × 1536 × 128)",
|
| 72 |
+
"shortDescription" : "",
|
| 73 |
+
"shape" : "[56, 8, 1536, 128]",
|
| 74 |
+
"name" : "model_model_kv_cache_0",
|
| 75 |
+
"type" : "State"
|
| 76 |
+
}
|
| 77 |
+
],
|
| 78 |
+
"outputSchema" : [
|
| 79 |
+
{
|
| 80 |
+
"hasShapeFlexibility" : "0",
|
| 81 |
+
"isOptional" : "0",
|
| 82 |
+
"dataType" : "Float16",
|
| 83 |
+
"formattedType" : "MultiArray (Float16 1 × 1 × 1024)",
|
| 84 |
+
"shortDescription" : "",
|
| 85 |
+
"shape" : "[1, 1, 1024]",
|
| 86 |
+
"name" : "output_hidden_states",
|
| 87 |
+
"type" : "MultiArray"
|
| 88 |
+
}
|
| 89 |
+
],
|
| 90 |
+
"name" : "infer",
|
| 91 |
+
"mlProgramOperationTypeHistogram" : {
|
| 92 |
+
"Ios18.expandDims" : 112,
|
| 93 |
+
"Ios18.mul" : 450,
|
| 94 |
+
"Ios18.softmax" : 28,
|
| 95 |
+
"Ios18.matmul" : 56,
|
| 96 |
+
"Identity" : 1,
|
| 97 |
+
"Ios18.greaterEqual" : 2,
|
| 98 |
+
"Select" : 2,
|
| 99 |
+
"Ios18.readState" : 57,
|
| 100 |
+
"Tile" : 56,
|
| 101 |
+
"Ios18.gather" : 2,
|
| 102 |
+
"Ios18.add" : 143,
|
| 103 |
+
"Ios18.layerNorm" : 113,
|
| 104 |
+
"Ios18.sliceUpdate" : 56,
|
| 105 |
+
"Ios18.writeState" : 56,
|
| 106 |
+
"Ios18.reshape" : 170,
|
| 107 |
+
"Ios18.constexprLutToDense" : 196,
|
| 108 |
+
"Ios18.conv" : 196,
|
| 109 |
+
"Ios18.concat" : 281,
|
| 110 |
+
"Ios18.transpose" : 168,
|
| 111 |
+
"Ios18.cast" : 5,
|
| 112 |
+
"Ios18.silu" : 28,
|
| 113 |
+
"Ios18.sliceByIndex" : 281,
|
| 114 |
+
"Ios18.squeeze" : 84
|
| 115 |
+
}
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"inputSchema" : [
|
| 119 |
+
{
|
| 120 |
+
"hasShapeFlexibility" : "0",
|
| 121 |
+
"isOptional" : "0",
|
| 122 |
+
"dataType" : "Float16",
|
| 123 |
+
"formattedType" : "MultiArray (Float16 1 × 64 × 1024)",
|
| 124 |
+
"shortDescription" : "",
|
| 125 |
+
"shape" : "[1, 64, 1024]",
|
| 126 |
+
"name" : "hidden_states",
|
| 127 |
+
"type" : "MultiArray"
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"hasShapeFlexibility" : "0",
|
| 131 |
+
"isOptional" : "0",
|
| 132 |
+
"dataType" : "Int32",
|
| 133 |
+
"formattedType" : "MultiArray (Int32 64)",
|
| 134 |
+
"shortDescription" : "",
|
| 135 |
+
"shape" : "[64]",
|
| 136 |
+
"name" : "position_ids",
|
| 137 |
+
"type" : "MultiArray"
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"hasShapeFlexibility" : "0",
|
| 141 |
+
"isOptional" : "0",
|
| 142 |
+
"dataType" : "Float16",
|
| 143 |
+
"formattedType" : "MultiArray (Float16 1 × 1 × 64 × 1536)",
|
| 144 |
+
"shortDescription" : "",
|
| 145 |
+
"shape" : "[1, 1, 64, 1536]",
|
| 146 |
+
"name" : "causal_mask",
|
| 147 |
+
"type" : "MultiArray"
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"hasShapeFlexibility" : "0",
|
| 151 |
+
"isOptional" : "0",
|
| 152 |
+
"dataType" : "Int32",
|
| 153 |
+
"formattedType" : "MultiArray (Int32 1)",
|
| 154 |
+
"shortDescription" : "",
|
| 155 |
+
"shape" : "[1]",
|
| 156 |
+
"name" : "current_pos",
|
| 157 |
+
"type" : "MultiArray"
|
| 158 |
+
}
|
| 159 |
+
],
|
| 160 |
+
"computePrecision" : "Mixed (Float16, Int16, Int32, UInt16)",
|
| 161 |
+
"storagePrecision" : "Mixed (Float16, Palettized (13 bits), Palettized (14 bits), Palettized (15 bits), UInt6)",
|
| 162 |
+
"stateSchema" : [
|
| 163 |
+
{
|
| 164 |
+
"dataType" : "Float16",
|
| 165 |
+
"isOptional" : "0",
|
| 166 |
+
"formattedType" : "State (Float16 56 × 8 × 1536 × 128)",
|
| 167 |
+
"shortDescription" : "",
|
| 168 |
+
"shape" : "[56, 8, 1536, 128]",
|
| 169 |
+
"name" : "model_model_kv_cache_0",
|
| 170 |
+
"type" : "State"
|
| 171 |
+
}
|
| 172 |
+
],
|
| 173 |
+
"outputSchema" : [
|
| 174 |
+
{
|
| 175 |
+
"hasShapeFlexibility" : "0",
|
| 176 |
+
"isOptional" : "0",
|
| 177 |
+
"dataType" : "Float16",
|
| 178 |
+
"formattedType" : "MultiArray (Float16 1 × 1 × 1024)",
|
| 179 |
+
"shortDescription" : "",
|
| 180 |
+
"shape" : "[1, 1, 1024]",
|
| 181 |
+
"name" : "output_hidden_states",
|
| 182 |
+
"type" : "MultiArray"
|
| 183 |
+
}
|
| 184 |
+
],
|
| 185 |
+
"name" : "prefill",
|
| 186 |
+
"mlProgramOperationTypeHistogram" : {
|
| 187 |
+
"Ios18.expandDims" : 112,
|
| 188 |
+
"Ios18.mul" : 448,
|
| 189 |
+
"Ios18.softmax" : 28,
|
| 190 |
+
"Ios18.matmul" : 56,
|
| 191 |
+
"Ios18.greaterEqual" : 2,
|
| 192 |
+
"Select" : 2,
|
| 193 |
+
"Ios18.readState" : 57,
|
| 194 |
+
"Tile" : 56,
|
| 195 |
+
"Ios18.gather" : 2,
|
| 196 |
+
"Ios18.add" : 143,
|
| 197 |
+
"Ios18.layerNorm" : 112,
|
| 198 |
+
"Ios18.sliceUpdate" : 56,
|
| 199 |
+
"Ios18.writeState" : 56,
|
| 200 |
+
"Ios18.reshape" : 226,
|
| 201 |
+
"Ios18.constexprLutToDense" : 196,
|
| 202 |
+
"Ios18.conv" : 196,
|
| 203 |
+
"Ios18.concat" : 280,
|
| 204 |
+
"Ios18.transpose" : 254,
|
| 205 |
+
"Ios18.cast" : 5,
|
| 206 |
+
"Ios18.silu" : 28,
|
| 207 |
+
"Ios18.sliceByIndex" : 281,
|
| 208 |
+
"Ios18.squeeze" : 84
|
| 209 |
+
}
|
| 210 |
+
}
|
| 211 |
+
],
|
| 212 |
+
"mlProgramOperationTypeHistogram" : {
|
| 213 |
+
"Ios18.expandDims" : 112,
|
| 214 |
+
"Ios18.mul" : 450,
|
| 215 |
+
"Ios18.softmax" : 28,
|
| 216 |
+
"Ios18.matmul" : 56,
|
| 217 |
+
"Identity" : 1,
|
| 218 |
+
"Ios18.greaterEqual" : 2,
|
| 219 |
+
"Select" : 2,
|
| 220 |
+
"Ios18.readState" : 57,
|
| 221 |
+
"Tile" : 56,
|
| 222 |
+
"Ios18.gather" : 2,
|
| 223 |
+
"Ios18.add" : 143,
|
| 224 |
+
"Ios18.layerNorm" : 113,
|
| 225 |
+
"Ios18.sliceUpdate" : 56,
|
| 226 |
+
"Ios18.writeState" : 56,
|
| 227 |
+
"Ios18.reshape" : 170,
|
| 228 |
+
"Ios18.constexprLutToDense" : 196,
|
| 229 |
+
"Ios18.conv" : 196,
|
| 230 |
+
"Ios18.concat" : 281,
|
| 231 |
+
"Ios18.transpose" : 168,
|
| 232 |
+
"Ios18.cast" : 5,
|
| 233 |
+
"Ios18.silu" : 28,
|
| 234 |
+
"Ios18.sliceByIndex" : 281,
|
| 235 |
+
"Ios18.squeeze" : 84
|
| 236 |
+
},
|
| 237 |
+
"isUpdatable" : "0",
|
| 238 |
+
"stateSchema" : [
|
| 239 |
+
{
|
| 240 |
+
"dataType" : "Float16",
|
| 241 |
+
"isOptional" : "0",
|
| 242 |
+
"formattedType" : "State (Float16 56 × 8 × 1536 × 128)",
|
| 243 |
+
"shortDescription" : "",
|
| 244 |
+
"shape" : "[56, 8, 1536, 128]",
|
| 245 |
+
"name" : "model_model_kv_cache_0",
|
| 246 |
+
"type" : "State"
|
| 247 |
+
}
|
| 248 |
+
],
|
| 249 |
+
"availability" : {
|
| 250 |
+
"macOS" : "15.0",
|
| 251 |
+
"tvOS" : "18.0",
|
| 252 |
+
"visionOS" : "2.0",
|
| 253 |
+
"watchOS" : "11.0",
|
| 254 |
+
"iOS" : "18.0",
|
| 255 |
+
"macCatalyst" : "18.0"
|
| 256 |
+
},
|
| 257 |
+
"computePrecision" : "Mixed (Float16, Int16, Int32, UInt16)",
|
| 258 |
+
"modelType" : {
|
| 259 |
+
"name" : "MLModelType_mlProgram"
|
| 260 |
+
},
|
| 261 |
+
"inputSchema" : [
|
| 262 |
+
{
|
| 263 |
+
"hasShapeFlexibility" : "0",
|
| 264 |
+
"isOptional" : "0",
|
| 265 |
+
"dataType" : "Float16",
|
| 266 |
+
"formattedType" : "MultiArray (Float16 1 × 1 × 1024)",
|
| 267 |
+
"shortDescription" : "",
|
| 268 |
+
"shape" : "[1, 1, 1024]",
|
| 269 |
+
"name" : "hidden_states",
|
| 270 |
+
"type" : "MultiArray"
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"hasShapeFlexibility" : "0",
|
| 274 |
+
"isOptional" : "0",
|
| 275 |
+
"dataType" : "Int32",
|
| 276 |
+
"formattedType" : "MultiArray (Int32 1)",
|
| 277 |
+
"shortDescription" : "",
|
| 278 |
+
"shape" : "[1]",
|
| 279 |
+
"name" : "position_ids",
|
| 280 |
+
"type" : "MultiArray"
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"hasShapeFlexibility" : "0",
|
| 284 |
+
"isOptional" : "0",
|
| 285 |
+
"dataType" : "Float16",
|
| 286 |
+
"formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1536)",
|
| 287 |
+
"shortDescription" : "",
|
| 288 |
+
"shape" : "[1, 1, 1, 1536]",
|
| 289 |
+
"name" : "causal_mask",
|
| 290 |
+
"type" : "MultiArray"
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"hasShapeFlexibility" : "0",
|
| 294 |
+
"isOptional" : "0",
|
| 295 |
+
"dataType" : "Int32",
|
| 296 |
+
"formattedType" : "MultiArray (Int32 1)",
|
| 297 |
+
"shortDescription" : "",
|
| 298 |
+
"shape" : "[1]",
|
| 299 |
+
"name" : "current_pos",
|
| 300 |
+
"type" : "MultiArray"
|
| 301 |
+
}
|
| 302 |
+
],
|
| 303 |
+
"defaultFunctionName" : "infer",
|
| 304 |
+
"generatedClassName" : "qwen_asr_decoder_lut6",
|
| 305 |
+
"userDefinedMetadata" : {
|
| 306 |
+
|
| 307 |
+
},
|
| 308 |
+
"method" : "predict"
|
| 309 |
+
}
|
| 310 |
+
]
|
decoder.mlmodelc/model.mil
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
decoder.mlmodelc/weights/weight.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d591bead53ede8964d8f8dd6923ef516321a2356e2971423e855a9448f5079ca
|
| 3 |
+
size 337542912
|
embeddings.mlmodelc/analytics/coremldata.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd5f2df05c2650f83c6dd7b1a302c3a02732f84d039bcea42bdcb596061b65e3
|
| 3 |
+
size 243
|
embeddings.mlmodelc/coremldata.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d99b5a0f4160499fc1905f4b6fa544a14099915f260d123980a109c48d291b2c
|
| 3 |
+
size 390
|
embeddings.mlmodelc/metadata.json
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"metadataOutputVersion" : "3.0",
|
| 4 |
+
"storagePrecision" : "Mixed (Float16, Palettized (21 bits), UInt6)",
|
| 5 |
+
"outputSchema" : [
|
| 6 |
+
{
|
| 7 |
+
"hasShapeFlexibility" : "0",
|
| 8 |
+
"isOptional" : "0",
|
| 9 |
+
"dataType" : "Float16",
|
| 10 |
+
"formattedType" : "MultiArray (Float16)",
|
| 11 |
+
"shortDescription" : "",
|
| 12 |
+
"shape" : "[]",
|
| 13 |
+
"name" : "hidden_states",
|
| 14 |
+
"type" : "MultiArray"
|
| 15 |
+
}
|
| 16 |
+
],
|
| 17 |
+
"modelParameters" : [
|
| 18 |
+
|
| 19 |
+
],
|
| 20 |
+
"specificationVersion" : 9,
|
| 21 |
+
"mlProgramOperationTypeHistogram" : {
|
| 22 |
+
"Ios18.greaterEqual" : 2,
|
| 23 |
+
"Ios18.constexprLutToDense" : 1,
|
| 24 |
+
"Ios18.add" : 2,
|
| 25 |
+
"Select" : 2,
|
| 26 |
+
"Ios18.gather" : 1
|
| 27 |
+
},
|
| 28 |
+
"computePrecision" : "Mixed (Float16, Int32)",
|
| 29 |
+
"isUpdatable" : "0",
|
| 30 |
+
"stateSchema" : [
|
| 31 |
+
|
| 32 |
+
],
|
| 33 |
+
"availability" : {
|
| 34 |
+
"macOS" : "15.0",
|
| 35 |
+
"tvOS" : "18.0",
|
| 36 |
+
"visionOS" : "2.0",
|
| 37 |
+
"watchOS" : "11.0",
|
| 38 |
+
"iOS" : "18.0",
|
| 39 |
+
"macCatalyst" : "18.0"
|
| 40 |
+
},
|
| 41 |
+
"modelType" : {
|
| 42 |
+
"name" : "MLModelType_mlProgram"
|
| 43 |
+
},
|
| 44 |
+
"userDefinedMetadata" : {
|
| 45 |
+
"com.github.apple.coremltools.conversion_date" : "2026-07-15",
|
| 46 |
+
"com.github.apple.coremltools.source" : "torch==2.5.0",
|
| 47 |
+
"com.github.apple.coremltools.version" : "9.0",
|
| 48 |
+
"com.github.apple.coremltools.source_dialect" : "TorchScript"
|
| 49 |
+
},
|
| 50 |
+
"inputSchema" : [
|
| 51 |
+
{
|
| 52 |
+
"shortDescription" : "",
|
| 53 |
+
"dataType" : "Int32",
|
| 54 |
+
"hasShapeFlexibility" : "1",
|
| 55 |
+
"isOptional" : "0",
|
| 56 |
+
"shapeFlexibility" : "1 × 1 | 1 × 64",
|
| 57 |
+
"formattedType" : "MultiArray (Int32 1 × 1)",
|
| 58 |
+
"type" : "MultiArray",
|
| 59 |
+
"shape" : "[1, 1]",
|
| 60 |
+
"name" : "input_ids",
|
| 61 |
+
"enumeratedShapes" : "[[1, 1], [1, 64]]"
|
| 62 |
+
}
|
| 63 |
+
],
|
| 64 |
+
"generatedClassName" : "qwen_asr_embeddings_lut6",
|
| 65 |
+
"method" : "predict"
|
| 66 |
+
}
|
| 67 |
+
]
|
embeddings.mlmodelc/model.mil
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
program(1.3)
|
| 2 |
+
[buildInfo = dict<string, string>({{"coremlc-component-MIL", "3510.2.1"}, {"coremlc-version", "3500.32.1"}})]
|
| 3 |
+
{
|
| 4 |
+
func main<ios18>(tensor<int32, [1, ?]> input_ids) [FlexibleShapeInformation = tuple<tuple<string, dict<string, tensor<int32, [?]>>>, tuple<string, dict<string, dict<string, tensor<int32, [?]>>>>>((("DefaultShapes", {{"input_ids", [1, 1]}}), ("EnumeratedShapes", {{"79ae981e", {{"input_ids", [1, 1]}}}, {"ed9b58c8", {{"input_ids", [1, 64]}}}})))] {
|
| 5 |
+
int32 hidden_states_batch_dims_0 = const()[name = string("hidden_states_batch_dims_0"), val = int32(0)];
|
| 6 |
+
bool hidden_states_validate_indices_0 = const()[name = string("hidden_states_validate_indices_0"), val = bool(false)];
|
| 7 |
+
tensor<fp16, [151936, 1024]> embed_tokens_weight_to_fp16_palettized = constexpr_lut_to_dense(indices = tensor<uint6, [151936, 1024]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(64))), lut = tensor<fp16, [18992, 1, 64, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(116686976))))[name = string("embed_tokens_weight_to_fp16_palettized")];
|
| 8 |
+
int32 greater_equal_0_y_0 = const()[name = string("greater_equal_0_y_0"), val = int32(0)];
|
| 9 |
+
tensor<bool, [1, ?]> greater_equal_0 = greater_equal(x = input_ids, y = greater_equal_0_y_0)[name = string("greater_equal_0")];
|
| 10 |
+
int32 slice_by_index_0 = const()[name = string("slice_by_index_0"), val = int32(151936)];
|
| 11 |
+
tensor<int32, [1, ?]> add_0 = add(x = input_ids, y = slice_by_index_0)[name = string("add_0")];
|
| 12 |
+
tensor<int32, [1, ?]> select_0 = select(a = input_ids, b = add_0, cond = greater_equal_0)[name = string("select_0")];
|
| 13 |
+
int32 greater_equal_0_y_0_1 = const()[name = string("greater_equal_0_y_0_1"), val = int32(0)];
|
| 14 |
+
tensor<bool, [1, ?]> greater_equal_0_1 = greater_equal(x = select_0, y = greater_equal_0_y_0_1)[name = string("greater_equal_0_1")];
|
| 15 |
+
int32 slice_by_index_0_1 = const()[name = string("slice_by_index_0_1"), val = int32(151936)];
|
| 16 |
+
tensor<int32, [1, ?]> add_0_1 = add(x = select_0, y = slice_by_index_0_1)[name = string("add_0_1")];
|
| 17 |
+
tensor<int32, [1, ?]> select_0_1 = select(a = select_0, b = add_0_1, cond = greater_equal_0_1)[name = string("select_0_1")];
|
| 18 |
+
int32 hidden_states_cast_fp16_axis_0 = const()[name = string("hidden_states_cast_fp16_axis_0"), val = int32(0)];
|
| 19 |
+
tensor<fp16, [1, ?, 1024]> hidden_states = gather(axis = hidden_states_cast_fp16_axis_0, batch_dims = hidden_states_batch_dims_0, indices = select_0_1, validate_indices = hidden_states_validate_indices_0, x = embed_tokens_weight_to_fp16_palettized)[name = string("hidden_states_cast_fp16")];
|
| 20 |
+
} -> (hidden_states);
|
| 21 |
+
}
|
embeddings.mlmodelc/weights/weight.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3daddee29a4aa85c4b78a96f3cb8415443a3b4939df90f2bac7544ed7f2e0389
|
| 3 |
+
size 119118016
|
lm_head.mlmodelc/analytics/coremldata.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b8386bff3c6fb854bc52629e16c9e4892ec1da3ef472a41099a1662a8b020787
|
| 3 |
+
size 243
|
lm_head.mlmodelc/coremldata.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:586993c435e7bb374f9ba4d24e55fa7440d62bc8fd549fd2919af4106fd4b208
|
| 3 |
+
size 406
|
lm_head.mlmodelc/metadata.json
ADDED
|
@@ -0,0 +1,79 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"metadataOutputVersion" : "3.0",
|
| 4 |
+
"storagePrecision" : "Mixed (Float16, Palettized (17 bits), UInt6)",
|
| 5 |
+
"outputSchema" : [
|
| 6 |
+
{
|
| 7 |
+
"hasShapeFlexibility" : "0",
|
| 8 |
+
"isOptional" : "0",
|
| 9 |
+
"dataType" : "Int32",
|
| 10 |
+
"formattedType" : "MultiArray (Int32 16)",
|
| 11 |
+
"shortDescription" : "",
|
| 12 |
+
"shape" : "[16]",
|
| 13 |
+
"name" : "argmax_idx",
|
| 14 |
+
"type" : "MultiArray"
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"hasShapeFlexibility" : "0",
|
| 18 |
+
"isOptional" : "0",
|
| 19 |
+
"dataType" : "Float16",
|
| 20 |
+
"formattedType" : "MultiArray (Float16 16)",
|
| 21 |
+
"shortDescription" : "",
|
| 22 |
+
"shape" : "[16]",
|
| 23 |
+
"name" : "argmax_val",
|
| 24 |
+
"type" : "MultiArray"
|
| 25 |
+
}
|
| 26 |
+
],
|
| 27 |
+
"modelParameters" : [
|
| 28 |
+
|
| 29 |
+
],
|
| 30 |
+
"specificationVersion" : 9,
|
| 31 |
+
"mlProgramOperationTypeHistogram" : {
|
| 32 |
+
"Ios18.squeeze" : 20,
|
| 33 |
+
"Ios18.gatherAlongAxis" : 16,
|
| 34 |
+
"Ios18.reduceArgmax" : 16,
|
| 35 |
+
"Ios18.concat" : 2,
|
| 36 |
+
"Ios18.transpose" : 17,
|
| 37 |
+
"Ios18.constexprLutToDense" : 16,
|
| 38 |
+
"Ios18.expandDims" : 1,
|
| 39 |
+
"Ios18.conv" : 16,
|
| 40 |
+
"Ios18.cast" : 18
|
| 41 |
+
},
|
| 42 |
+
"computePrecision" : "Mixed (Float16, Int16, Int32, UInt16)",
|
| 43 |
+
"isUpdatable" : "0",
|
| 44 |
+
"stateSchema" : [
|
| 45 |
+
|
| 46 |
+
],
|
| 47 |
+
"availability" : {
|
| 48 |
+
"macOS" : "15.0",
|
| 49 |
+
"tvOS" : "18.0",
|
| 50 |
+
"visionOS" : "2.0",
|
| 51 |
+
"watchOS" : "11.0",
|
| 52 |
+
"iOS" : "18.0",
|
| 53 |
+
"macCatalyst" : "18.0"
|
| 54 |
+
},
|
| 55 |
+
"modelType" : {
|
| 56 |
+
"name" : "MLModelType_mlProgram"
|
| 57 |
+
},
|
| 58 |
+
"userDefinedMetadata" : {
|
| 59 |
+
"com.github.apple.coremltools.conversion_date" : "2026-07-15",
|
| 60 |
+
"com.github.apple.coremltools.source" : "torch==2.5.0",
|
| 61 |
+
"com.github.apple.coremltools.version" : "9.0",
|
| 62 |
+
"com.github.apple.coremltools.source_dialect" : "TorchScript"
|
| 63 |
+
},
|
| 64 |
+
"inputSchema" : [
|
| 65 |
+
{
|
| 66 |
+
"hasShapeFlexibility" : "0",
|
| 67 |
+
"isOptional" : "0",
|
| 68 |
+
"dataType" : "Float16",
|
| 69 |
+
"formattedType" : "MultiArray (Float16 1 × 1 × 1024)",
|
| 70 |
+
"shortDescription" : "",
|
| 71 |
+
"shape" : "[1, 1, 1024]",
|
| 72 |
+
"name" : "hidden_states",
|
| 73 |
+
"type" : "MultiArray"
|
| 74 |
+
}
|
| 75 |
+
],
|
| 76 |
+
"generatedClassName" : "qwen_asr_lm_head_lut6",
|
| 77 |
+
"method" : "predict"
|
| 78 |
+
}
|
| 79 |
+
]
|
lm_head.mlmodelc/model.mil
ADDED
|
@@ -0,0 +1,348 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
program(1.3)
|
| 2 |
+
[buildInfo = dict<string, string>({{"coremlc-component-MIL", "3510.2.1"}, {"coremlc-version", "3500.32.1"}})]
|
| 3 |
+
{
|
| 4 |
+
func main<ios18>(tensor<fp16, [1, 1, 1024]> hidden_states) {
|
| 5 |
+
tensor<int32, [3]> var_5 = const()[name = string("op_5"), val = tensor<int32, [3]>([0, 2, 1])];
|
| 6 |
+
tensor<int32, [1]> input_axes_0 = const()[name = string("input_axes_0"), val = tensor<int32, [1]>([2])];
|
| 7 |
+
tensor<fp16, [1, 1024, 1]> var_6_cast_fp16 = transpose(perm = var_5, x = hidden_states)[name = string("transpose_16")];
|
| 8 |
+
tensor<fp16, [1, 1024, 1, 1]> input_cast_fp16 = expand_dims(axes = input_axes_0, x = var_6_cast_fp16)[name = string("input_cast_fp16")];
|
| 9 |
+
string var_29_pad_type_0 = const()[name = string("op_29_pad_type_0"), val = string("valid")];
|
| 10 |
+
tensor<int32, [2]> var_29_strides_0 = const()[name = string("op_29_strides_0"), val = tensor<int32, [2]>([1, 1])];
|
| 11 |
+
tensor<int32, [4]> var_29_pad_0 = const()[name = string("op_29_pad_0"), val = tensor<int32, [4]>([0, 0, 0, 0])];
|
| 12 |
+
tensor<int32, [2]> var_29_dilations_0 = const()[name = string("op_29_dilations_0"), val = tensor<int32, [2]>([1, 1])];
|
| 13 |
+
int32 var_29_groups_0 = const()[name = string("op_29_groups_0"), val = int32(1)];
|
| 14 |
+
tensor<fp16, [9496, 1024, 1, 1]> op_9_promoted_to_fp16_palettized = constexpr_lut_to_dense(indices = tensor<uint6, [9496, 1024, 1, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(64))), lut = tensor<fp16, [1187, 1, 1, 1, 64, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(7293056))))[name = string("op_9_promoted_to_fp16_palettized")];
|
| 15 |
+
tensor<fp16, [1, 9496, 1, 1]> var_29_cast_fp16 = conv(dilations = var_29_dilations_0, groups = var_29_groups_0, pad = var_29_pad_0, pad_type = var_29_pad_type_0, strides = var_29_strides_0, weight = op_9_promoted_to_fp16_palettized, x = input_cast_fp16)[name = string("op_29_cast_fp16")];
|
| 16 |
+
tensor<int32, [1]> var_31_axes_0 = const()[name = string("op_31_axes_0"), val = tensor<int32, [1]>([2])];
|
| 17 |
+
tensor<fp16, [1, 9496, 1]> var_31_cast_fp16 = squeeze(axes = var_31_axes_0, x = var_29_cast_fp16)[name = string("op_31_cast_fp16")];
|
| 18 |
+
tensor<int32, [3]> logits_1_perm_0 = const()[name = string("logits_1_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
|
| 19 |
+
string var_55_pad_type_0 = const()[name = string("op_55_pad_type_0"), val = string("valid")];
|
| 20 |
+
tensor<int32, [2]> var_55_strides_0 = const()[name = string("op_55_strides_0"), val = tensor<int32, [2]>([1, 1])];
|
| 21 |
+
tensor<int32, [4]> var_55_pad_0 = const()[name = string("op_55_pad_0"), val = tensor<int32, [4]>([0, 0, 0, 0])];
|
| 22 |
+
tensor<int32, [2]> var_55_dilations_0 = const()[name = string("op_55_dilations_0"), val = tensor<int32, [2]>([1, 1])];
|
| 23 |
+
int32 var_55_groups_0 = const()[name = string("op_55_groups_0"), val = int32(1)];
|
| 24 |
+
tensor<fp16, [9496, 1024, 1, 1]> op_35_promoted_to_fp16_palettized = constexpr_lut_to_dense(indices = tensor<uint6, [9496, 1024, 1, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(7445056))), lut = tensor<fp16, [1187, 1, 1, 1, 64, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(14738048))))[name = string("op_35_promoted_to_fp16_palettized")];
|
| 25 |
+
tensor<fp16, [1, 9496, 1, 1]> var_55_cast_fp16 = conv(dilations = var_55_dilations_0, groups = var_55_groups_0, pad = var_55_pad_0, pad_type = var_55_pad_type_0, strides = var_55_strides_0, weight = op_35_promoted_to_fp16_palettized, x = input_cast_fp16)[name = string("op_55_cast_fp16")];
|
| 26 |
+
tensor<int32, [1]> var_57_axes_0 = const()[name = string("op_57_axes_0"), val = tensor<int32, [1]>([2])];
|
| 27 |
+
tensor<fp16, [1, 9496, 1]> var_57_cast_fp16 = squeeze(axes = var_57_axes_0, x = var_55_cast_fp16)[name = string("op_57_cast_fp16")];
|
| 28 |
+
tensor<int32, [3]> logits_3_perm_0 = const()[name = string("logits_3_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
|
| 29 |
+
string var_81_pad_type_0 = const()[name = string("op_81_pad_type_0"), val = string("valid")];
|
| 30 |
+
tensor<int32, [2]> var_81_strides_0 = const()[name = string("op_81_strides_0"), val = tensor<int32, [2]>([1, 1])];
|
| 31 |
+
tensor<int32, [4]> var_81_pad_0 = const()[name = string("op_81_pad_0"), val = tensor<int32, [4]>([0, 0, 0, 0])];
|
| 32 |
+
tensor<int32, [2]> var_81_dilations_0 = const()[name = string("op_81_dilations_0"), val = tensor<int32, [2]>([1, 1])];
|
| 33 |
+
int32 var_81_groups_0 = const()[name = string("op_81_groups_0"), val = int32(1)];
|
| 34 |
+
tensor<fp16, [9496, 1024, 1, 1]> op_61_promoted_to_fp16_palettized = constexpr_lut_to_dense(indices = tensor<uint6, [9496, 1024, 1, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(14890048))), lut = tensor<fp16, [1187, 1, 1, 1, 64, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(22183040))))[name = string("op_61_promoted_to_fp16_palettized")];
|
| 35 |
+
tensor<fp16, [1, 9496, 1, 1]> var_81_cast_fp16 = conv(dilations = var_81_dilations_0, groups = var_81_groups_0, pad = var_81_pad_0, pad_type = var_81_pad_type_0, strides = var_81_strides_0, weight = op_61_promoted_to_fp16_palettized, x = input_cast_fp16)[name = string("op_81_cast_fp16")];
|
| 36 |
+
tensor<int32, [1]> var_83_axes_0 = const()[name = string("op_83_axes_0"), val = tensor<int32, [1]>([2])];
|
| 37 |
+
tensor<fp16, [1, 9496, 1]> var_83_cast_fp16 = squeeze(axes = var_83_axes_0, x = var_81_cast_fp16)[name = string("op_83_cast_fp16")];
|
| 38 |
+
tensor<int32, [3]> logits_5_perm_0 = const()[name = string("logits_5_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
|
| 39 |
+
string var_107_pad_type_0 = const()[name = string("op_107_pad_type_0"), val = string("valid")];
|
| 40 |
+
tensor<int32, [2]> var_107_strides_0 = const()[name = string("op_107_strides_0"), val = tensor<int32, [2]>([1, 1])];
|
| 41 |
+
tensor<int32, [4]> var_107_pad_0 = const()[name = string("op_107_pad_0"), val = tensor<int32, [4]>([0, 0, 0, 0])];
|
| 42 |
+
tensor<int32, [2]> var_107_dilations_0 = const()[name = string("op_107_dilations_0"), val = tensor<int32, [2]>([1, 1])];
|
| 43 |
+
int32 var_107_groups_0 = const()[name = string("op_107_groups_0"), val = int32(1)];
|
| 44 |
+
tensor<fp16, [9496, 1024, 1, 1]> op_87_promoted_to_fp16_palettized = constexpr_lut_to_dense(indices = tensor<uint6, [9496, 1024, 1, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(22335040))), lut = tensor<fp16, [1187, 1, 1, 1, 64, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(29628032))))[name = string("op_87_promoted_to_fp16_palettized")];
|
| 45 |
+
tensor<fp16, [1, 9496, 1, 1]> var_107_cast_fp16 = conv(dilations = var_107_dilations_0, groups = var_107_groups_0, pad = var_107_pad_0, pad_type = var_107_pad_type_0, strides = var_107_strides_0, weight = op_87_promoted_to_fp16_palettized, x = input_cast_fp16)[name = string("op_107_cast_fp16")];
|
| 46 |
+
tensor<int32, [1]> var_109_axes_0 = const()[name = string("op_109_axes_0"), val = tensor<int32, [1]>([2])];
|
| 47 |
+
tensor<fp16, [1, 9496, 1]> var_109_cast_fp16 = squeeze(axes = var_109_axes_0, x = var_107_cast_fp16)[name = string("op_109_cast_fp16")];
|
| 48 |
+
tensor<int32, [3]> logits_7_perm_0 = const()[name = string("logits_7_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
|
| 49 |
+
string var_133_pad_type_0 = const()[name = string("op_133_pad_type_0"), val = string("valid")];
|
| 50 |
+
tensor<int32, [2]> var_133_strides_0 = const()[name = string("op_133_strides_0"), val = tensor<int32, [2]>([1, 1])];
|
| 51 |
+
tensor<int32, [4]> var_133_pad_0 = const()[name = string("op_133_pad_0"), val = tensor<int32, [4]>([0, 0, 0, 0])];
|
| 52 |
+
tensor<int32, [2]> var_133_dilations_0 = const()[name = string("op_133_dilations_0"), val = tensor<int32, [2]>([1, 1])];
|
| 53 |
+
int32 var_133_groups_0 = const()[name = string("op_133_groups_0"), val = int32(1)];
|
| 54 |
+
tensor<fp16, [9496, 1024, 1, 1]> op_113_promoted_to_fp16_palettized = constexpr_lut_to_dense(indices = tensor<uint6, [9496, 1024, 1, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(29780032))), lut = tensor<fp16, [1187, 1, 1, 1, 64, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(37073024))))[name = string("op_113_promoted_to_fp16_palettized")];
|
| 55 |
+
tensor<fp16, [1, 9496, 1, 1]> var_133_cast_fp16 = conv(dilations = var_133_dilations_0, groups = var_133_groups_0, pad = var_133_pad_0, pad_type = var_133_pad_type_0, strides = var_133_strides_0, weight = op_113_promoted_to_fp16_palettized, x = input_cast_fp16)[name = string("op_133_cast_fp16")];
|
| 56 |
+
tensor<int32, [1]> var_135_axes_0 = const()[name = string("op_135_axes_0"), val = tensor<int32, [1]>([2])];
|
| 57 |
+
tensor<fp16, [1, 9496, 1]> var_135_cast_fp16 = squeeze(axes = var_135_axes_0, x = var_133_cast_fp16)[name = string("op_135_cast_fp16")];
|
| 58 |
+
tensor<int32, [3]> logits_9_perm_0 = const()[name = string("logits_9_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
|
| 59 |
+
string var_159_pad_type_0 = const()[name = string("op_159_pad_type_0"), val = string("valid")];
|
| 60 |
+
tensor<int32, [2]> var_159_strides_0 = const()[name = string("op_159_strides_0"), val = tensor<int32, [2]>([1, 1])];
|
| 61 |
+
tensor<int32, [4]> var_159_pad_0 = const()[name = string("op_159_pad_0"), val = tensor<int32, [4]>([0, 0, 0, 0])];
|
| 62 |
+
tensor<int32, [2]> var_159_dilations_0 = const()[name = string("op_159_dilations_0"), val = tensor<int32, [2]>([1, 1])];
|
| 63 |
+
int32 var_159_groups_0 = const()[name = string("op_159_groups_0"), val = int32(1)];
|
| 64 |
+
tensor<fp16, [9496, 1024, 1, 1]> op_139_promoted_to_fp16_palettized = constexpr_lut_to_dense(indices = tensor<uint6, [9496, 1024, 1, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(37225024))), lut = tensor<fp16, [1187, 1, 1, 1, 64, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(44518016))))[name = string("op_139_promoted_to_fp16_palettized")];
|
| 65 |
+
tensor<fp16, [1, 9496, 1, 1]> var_159_cast_fp16 = conv(dilations = var_159_dilations_0, groups = var_159_groups_0, pad = var_159_pad_0, pad_type = var_159_pad_type_0, strides = var_159_strides_0, weight = op_139_promoted_to_fp16_palettized, x = input_cast_fp16)[name = string("op_159_cast_fp16")];
|
| 66 |
+
tensor<int32, [1]> var_161_axes_0 = const()[name = string("op_161_axes_0"), val = tensor<int32, [1]>([2])];
|
| 67 |
+
tensor<fp16, [1, 9496, 1]> var_161_cast_fp16 = squeeze(axes = var_161_axes_0, x = var_159_cast_fp16)[name = string("op_161_cast_fp16")];
|
| 68 |
+
tensor<int32, [3]> logits_11_perm_0 = const()[name = string("logits_11_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
|
| 69 |
+
string var_185_pad_type_0 = const()[name = string("op_185_pad_type_0"), val = string("valid")];
|
| 70 |
+
tensor<int32, [2]> var_185_strides_0 = const()[name = string("op_185_strides_0"), val = tensor<int32, [2]>([1, 1])];
|
| 71 |
+
tensor<int32, [4]> var_185_pad_0 = const()[name = string("op_185_pad_0"), val = tensor<int32, [4]>([0, 0, 0, 0])];
|
| 72 |
+
tensor<int32, [2]> var_185_dilations_0 = const()[name = string("op_185_dilations_0"), val = tensor<int32, [2]>([1, 1])];
|
| 73 |
+
int32 var_185_groups_0 = const()[name = string("op_185_groups_0"), val = int32(1)];
|
| 74 |
+
tensor<fp16, [9496, 1024, 1, 1]> op_165_promoted_to_fp16_palettized = constexpr_lut_to_dense(indices = tensor<uint6, [9496, 1024, 1, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(44670016))), lut = tensor<fp16, [1187, 1, 1, 1, 64, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(51963008))))[name = string("op_165_promoted_to_fp16_palettized")];
|
| 75 |
+
tensor<fp16, [1, 9496, 1, 1]> var_185_cast_fp16 = conv(dilations = var_185_dilations_0, groups = var_185_groups_0, pad = var_185_pad_0, pad_type = var_185_pad_type_0, strides = var_185_strides_0, weight = op_165_promoted_to_fp16_palettized, x = input_cast_fp16)[name = string("op_185_cast_fp16")];
|
| 76 |
+
tensor<int32, [1]> var_187_axes_0 = const()[name = string("op_187_axes_0"), val = tensor<int32, [1]>([2])];
|
| 77 |
+
tensor<fp16, [1, 9496, 1]> var_187_cast_fp16 = squeeze(axes = var_187_axes_0, x = var_185_cast_fp16)[name = string("op_187_cast_fp16")];
|
| 78 |
+
tensor<int32, [3]> logits_13_perm_0 = const()[name = string("logits_13_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
|
| 79 |
+
string var_211_pad_type_0 = const()[name = string("op_211_pad_type_0"), val = string("valid")];
|
| 80 |
+
tensor<int32, [2]> var_211_strides_0 = const()[name = string("op_211_strides_0"), val = tensor<int32, [2]>([1, 1])];
|
| 81 |
+
tensor<int32, [4]> var_211_pad_0 = const()[name = string("op_211_pad_0"), val = tensor<int32, [4]>([0, 0, 0, 0])];
|
| 82 |
+
tensor<int32, [2]> var_211_dilations_0 = const()[name = string("op_211_dilations_0"), val = tensor<int32, [2]>([1, 1])];
|
| 83 |
+
int32 var_211_groups_0 = const()[name = string("op_211_groups_0"), val = int32(1)];
|
| 84 |
+
tensor<fp16, [9496, 1024, 1, 1]> op_191_promoted_to_fp16_palettized = constexpr_lut_to_dense(indices = tensor<uint6, [9496, 1024, 1, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(52115008))), lut = tensor<fp16, [1187, 1, 1, 1, 64, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(59408000))))[name = string("op_191_promoted_to_fp16_palettized")];
|
| 85 |
+
tensor<fp16, [1, 9496, 1, 1]> var_211_cast_fp16 = conv(dilations = var_211_dilations_0, groups = var_211_groups_0, pad = var_211_pad_0, pad_type = var_211_pad_type_0, strides = var_211_strides_0, weight = op_191_promoted_to_fp16_palettized, x = input_cast_fp16)[name = string("op_211_cast_fp16")];
|
| 86 |
+
tensor<int32, [1]> var_213_axes_0 = const()[name = string("op_213_axes_0"), val = tensor<int32, [1]>([2])];
|
| 87 |
+
tensor<fp16, [1, 9496, 1]> var_213_cast_fp16 = squeeze(axes = var_213_axes_0, x = var_211_cast_fp16)[name = string("op_213_cast_fp16")];
|
| 88 |
+
tensor<int32, [3]> logits_15_perm_0 = const()[name = string("logits_15_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
|
| 89 |
+
string var_237_pad_type_0 = const()[name = string("op_237_pad_type_0"), val = string("valid")];
|
| 90 |
+
tensor<int32, [2]> var_237_strides_0 = const()[name = string("op_237_strides_0"), val = tensor<int32, [2]>([1, 1])];
|
| 91 |
+
tensor<int32, [4]> var_237_pad_0 = const()[name = string("op_237_pad_0"), val = tensor<int32, [4]>([0, 0, 0, 0])];
|
| 92 |
+
tensor<int32, [2]> var_237_dilations_0 = const()[name = string("op_237_dilations_0"), val = tensor<int32, [2]>([1, 1])];
|
| 93 |
+
int32 var_237_groups_0 = const()[name = string("op_237_groups_0"), val = int32(1)];
|
| 94 |
+
tensor<fp16, [9496, 1024, 1, 1]> op_217_promoted_to_fp16_palettized = constexpr_lut_to_dense(indices = tensor<uint6, [9496, 1024, 1, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(59560000))), lut = tensor<fp16, [1187, 1, 1, 1, 64, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(66852992))))[name = string("op_217_promoted_to_fp16_palettized")];
|
| 95 |
+
tensor<fp16, [1, 9496, 1, 1]> var_237_cast_fp16 = conv(dilations = var_237_dilations_0, groups = var_237_groups_0, pad = var_237_pad_0, pad_type = var_237_pad_type_0, strides = var_237_strides_0, weight = op_217_promoted_to_fp16_palettized, x = input_cast_fp16)[name = string("op_237_cast_fp16")];
|
| 96 |
+
tensor<int32, [1]> var_239_axes_0 = const()[name = string("op_239_axes_0"), val = tensor<int32, [1]>([2])];
|
| 97 |
+
tensor<fp16, [1, 9496, 1]> var_239_cast_fp16 = squeeze(axes = var_239_axes_0, x = var_237_cast_fp16)[name = string("op_239_cast_fp16")];
|
| 98 |
+
tensor<int32, [3]> logits_17_perm_0 = const()[name = string("logits_17_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
|
| 99 |
+
string var_263_pad_type_0 = const()[name = string("op_263_pad_type_0"), val = string("valid")];
|
| 100 |
+
tensor<int32, [2]> var_263_strides_0 = const()[name = string("op_263_strides_0"), val = tensor<int32, [2]>([1, 1])];
|
| 101 |
+
tensor<int32, [4]> var_263_pad_0 = const()[name = string("op_263_pad_0"), val = tensor<int32, [4]>([0, 0, 0, 0])];
|
| 102 |
+
tensor<int32, [2]> var_263_dilations_0 = const()[name = string("op_263_dilations_0"), val = tensor<int32, [2]>([1, 1])];
|
| 103 |
+
int32 var_263_groups_0 = const()[name = string("op_263_groups_0"), val = int32(1)];
|
| 104 |
+
tensor<fp16, [9496, 1024, 1, 1]> op_243_promoted_to_fp16_palettized = constexpr_lut_to_dense(indices = tensor<uint6, [9496, 1024, 1, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(67004992))), lut = tensor<fp16, [1187, 1, 1, 1, 64, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(74297984))))[name = string("op_243_promoted_to_fp16_palettized")];
|
| 105 |
+
tensor<fp16, [1, 9496, 1, 1]> var_263_cast_fp16 = conv(dilations = var_263_dilations_0, groups = var_263_groups_0, pad = var_263_pad_0, pad_type = var_263_pad_type_0, strides = var_263_strides_0, weight = op_243_promoted_to_fp16_palettized, x = input_cast_fp16)[name = string("op_263_cast_fp16")];
|
| 106 |
+
tensor<int32, [1]> var_265_axes_0 = const()[name = string("op_265_axes_0"), val = tensor<int32, [1]>([2])];
|
| 107 |
+
tensor<fp16, [1, 9496, 1]> var_265_cast_fp16 = squeeze(axes = var_265_axes_0, x = var_263_cast_fp16)[name = string("op_265_cast_fp16")];
|
| 108 |
+
tensor<int32, [3]> logits_19_perm_0 = const()[name = string("logits_19_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
|
| 109 |
+
string var_289_pad_type_0 = const()[name = string("op_289_pad_type_0"), val = string("valid")];
|
| 110 |
+
tensor<int32, [2]> var_289_strides_0 = const()[name = string("op_289_strides_0"), val = tensor<int32, [2]>([1, 1])];
|
| 111 |
+
tensor<int32, [4]> var_289_pad_0 = const()[name = string("op_289_pad_0"), val = tensor<int32, [4]>([0, 0, 0, 0])];
|
| 112 |
+
tensor<int32, [2]> var_289_dilations_0 = const()[name = string("op_289_dilations_0"), val = tensor<int32, [2]>([1, 1])];
|
| 113 |
+
int32 var_289_groups_0 = const()[name = string("op_289_groups_0"), val = int32(1)];
|
| 114 |
+
tensor<fp16, [9496, 1024, 1, 1]> op_269_promoted_to_fp16_palettized = constexpr_lut_to_dense(indices = tensor<uint6, [9496, 1024, 1, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(74449984))), lut = tensor<fp16, [1187, 1, 1, 1, 64, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(81742976))))[name = string("op_269_promoted_to_fp16_palettized")];
|
| 115 |
+
tensor<fp16, [1, 9496, 1, 1]> var_289_cast_fp16 = conv(dilations = var_289_dilations_0, groups = var_289_groups_0, pad = var_289_pad_0, pad_type = var_289_pad_type_0, strides = var_289_strides_0, weight = op_269_promoted_to_fp16_palettized, x = input_cast_fp16)[name = string("op_289_cast_fp16")];
|
| 116 |
+
tensor<int32, [1]> var_291_axes_0 = const()[name = string("op_291_axes_0"), val = tensor<int32, [1]>([2])];
|
| 117 |
+
tensor<fp16, [1, 9496, 1]> var_291_cast_fp16 = squeeze(axes = var_291_axes_0, x = var_289_cast_fp16)[name = string("op_291_cast_fp16")];
|
| 118 |
+
tensor<int32, [3]> logits_21_perm_0 = const()[name = string("logits_21_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
|
| 119 |
+
string var_315_pad_type_0 = const()[name = string("op_315_pad_type_0"), val = string("valid")];
|
| 120 |
+
tensor<int32, [2]> var_315_strides_0 = const()[name = string("op_315_strides_0"), val = tensor<int32, [2]>([1, 1])];
|
| 121 |
+
tensor<int32, [4]> var_315_pad_0 = const()[name = string("op_315_pad_0"), val = tensor<int32, [4]>([0, 0, 0, 0])];
|
| 122 |
+
tensor<int32, [2]> var_315_dilations_0 = const()[name = string("op_315_dilations_0"), val = tensor<int32, [2]>([1, 1])];
|
| 123 |
+
int32 var_315_groups_0 = const()[name = string("op_315_groups_0"), val = int32(1)];
|
| 124 |
+
tensor<fp16, [9496, 1024, 1, 1]> op_295_promoted_to_fp16_palettized = constexpr_lut_to_dense(indices = tensor<uint6, [9496, 1024, 1, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(81894976))), lut = tensor<fp16, [1187, 1, 1, 1, 64, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(89187968))))[name = string("op_295_promoted_to_fp16_palettized")];
|
| 125 |
+
tensor<fp16, [1, 9496, 1, 1]> var_315_cast_fp16 = conv(dilations = var_315_dilations_0, groups = var_315_groups_0, pad = var_315_pad_0, pad_type = var_315_pad_type_0, strides = var_315_strides_0, weight = op_295_promoted_to_fp16_palettized, x = input_cast_fp16)[name = string("op_315_cast_fp16")];
|
| 126 |
+
tensor<int32, [1]> var_317_axes_0 = const()[name = string("op_317_axes_0"), val = tensor<int32, [1]>([2])];
|
| 127 |
+
tensor<fp16, [1, 9496, 1]> var_317_cast_fp16 = squeeze(axes = var_317_axes_0, x = var_315_cast_fp16)[name = string("op_317_cast_fp16")];
|
| 128 |
+
tensor<int32, [3]> logits_23_perm_0 = const()[name = string("logits_23_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
|
| 129 |
+
string var_341_pad_type_0 = const()[name = string("op_341_pad_type_0"), val = string("valid")];
|
| 130 |
+
tensor<int32, [2]> var_341_strides_0 = const()[name = string("op_341_strides_0"), val = tensor<int32, [2]>([1, 1])];
|
| 131 |
+
tensor<int32, [4]> var_341_pad_0 = const()[name = string("op_341_pad_0"), val = tensor<int32, [4]>([0, 0, 0, 0])];
|
| 132 |
+
tensor<int32, [2]> var_341_dilations_0 = const()[name = string("op_341_dilations_0"), val = tensor<int32, [2]>([1, 1])];
|
| 133 |
+
int32 var_341_groups_0 = const()[name = string("op_341_groups_0"), val = int32(1)];
|
| 134 |
+
tensor<fp16, [9496, 1024, 1, 1]> op_321_promoted_to_fp16_palettized = constexpr_lut_to_dense(indices = tensor<uint6, [9496, 1024, 1, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(89339968))), lut = tensor<fp16, [1187, 1, 1, 1, 64, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(96632960))))[name = string("op_321_promoted_to_fp16_palettized")];
|
| 135 |
+
tensor<fp16, [1, 9496, 1, 1]> var_341_cast_fp16 = conv(dilations = var_341_dilations_0, groups = var_341_groups_0, pad = var_341_pad_0, pad_type = var_341_pad_type_0, strides = var_341_strides_0, weight = op_321_promoted_to_fp16_palettized, x = input_cast_fp16)[name = string("op_341_cast_fp16")];
|
| 136 |
+
tensor<int32, [1]> var_343_axes_0 = const()[name = string("op_343_axes_0"), val = tensor<int32, [1]>([2])];
|
| 137 |
+
tensor<fp16, [1, 9496, 1]> var_343_cast_fp16 = squeeze(axes = var_343_axes_0, x = var_341_cast_fp16)[name = string("op_343_cast_fp16")];
|
| 138 |
+
tensor<int32, [3]> logits_25_perm_0 = const()[name = string("logits_25_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
|
| 139 |
+
string var_367_pad_type_0 = const()[name = string("op_367_pad_type_0"), val = string("valid")];
|
| 140 |
+
tensor<int32, [2]> var_367_strides_0 = const()[name = string("op_367_strides_0"), val = tensor<int32, [2]>([1, 1])];
|
| 141 |
+
tensor<int32, [4]> var_367_pad_0 = const()[name = string("op_367_pad_0"), val = tensor<int32, [4]>([0, 0, 0, 0])];
|
| 142 |
+
tensor<int32, [2]> var_367_dilations_0 = const()[name = string("op_367_dilations_0"), val = tensor<int32, [2]>([1, 1])];
|
| 143 |
+
int32 var_367_groups_0 = const()[name = string("op_367_groups_0"), val = int32(1)];
|
| 144 |
+
tensor<fp16, [9496, 1024, 1, 1]> op_347_promoted_to_fp16_palettized = constexpr_lut_to_dense(indices = tensor<uint6, [9496, 1024, 1, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(96784960))), lut = tensor<fp16, [1187, 1, 1, 1, 64, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(104077952))))[name = string("op_347_promoted_to_fp16_palettized")];
|
| 145 |
+
tensor<fp16, [1, 9496, 1, 1]> var_367_cast_fp16 = conv(dilations = var_367_dilations_0, groups = var_367_groups_0, pad = var_367_pad_0, pad_type = var_367_pad_type_0, strides = var_367_strides_0, weight = op_347_promoted_to_fp16_palettized, x = input_cast_fp16)[name = string("op_367_cast_fp16")];
|
| 146 |
+
tensor<int32, [1]> var_369_axes_0 = const()[name = string("op_369_axes_0"), val = tensor<int32, [1]>([2])];
|
| 147 |
+
tensor<fp16, [1, 9496, 1]> var_369_cast_fp16 = squeeze(axes = var_369_axes_0, x = var_367_cast_fp16)[name = string("op_369_cast_fp16")];
|
| 148 |
+
tensor<int32, [3]> logits_27_perm_0 = const()[name = string("logits_27_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
|
| 149 |
+
string var_393_pad_type_0 = const()[name = string("op_393_pad_type_0"), val = string("valid")];
|
| 150 |
+
tensor<int32, [2]> var_393_strides_0 = const()[name = string("op_393_strides_0"), val = tensor<int32, [2]>([1, 1])];
|
| 151 |
+
tensor<int32, [4]> var_393_pad_0 = const()[name = string("op_393_pad_0"), val = tensor<int32, [4]>([0, 0, 0, 0])];
|
| 152 |
+
tensor<int32, [2]> var_393_dilations_0 = const()[name = string("op_393_dilations_0"), val = tensor<int32, [2]>([1, 1])];
|
| 153 |
+
int32 var_393_groups_0 = const()[name = string("op_393_groups_0"), val = int32(1)];
|
| 154 |
+
tensor<fp16, [9496, 1024, 1, 1]> op_373_promoted_to_fp16_palettized = constexpr_lut_to_dense(indices = tensor<uint6, [9496, 1024, 1, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(104229952))), lut = tensor<fp16, [1187, 1, 1, 1, 64, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(111522944))))[name = string("op_373_promoted_to_fp16_palettized")];
|
| 155 |
+
tensor<fp16, [1, 9496, 1, 1]> var_393_cast_fp16 = conv(dilations = var_393_dilations_0, groups = var_393_groups_0, pad = var_393_pad_0, pad_type = var_393_pad_type_0, strides = var_393_strides_0, weight = op_373_promoted_to_fp16_palettized, x = input_cast_fp16)[name = string("op_393_cast_fp16")];
|
| 156 |
+
tensor<int32, [1]> var_395_axes_0 = const()[name = string("op_395_axes_0"), val = tensor<int32, [1]>([2])];
|
| 157 |
+
tensor<fp16, [1, 9496, 1]> var_395_cast_fp16 = squeeze(axes = var_395_axes_0, x = var_393_cast_fp16)[name = string("op_395_cast_fp16")];
|
| 158 |
+
tensor<int32, [3]> logits_29_perm_0 = const()[name = string("logits_29_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
|
| 159 |
+
string var_419_pad_type_0 = const()[name = string("op_419_pad_type_0"), val = string("valid")];
|
| 160 |
+
tensor<int32, [2]> var_419_strides_0 = const()[name = string("op_419_strides_0"), val = tensor<int32, [2]>([1, 1])];
|
| 161 |
+
tensor<int32, [4]> var_419_pad_0 = const()[name = string("op_419_pad_0"), val = tensor<int32, [4]>([0, 0, 0, 0])];
|
| 162 |
+
tensor<int32, [2]> var_419_dilations_0 = const()[name = string("op_419_dilations_0"), val = tensor<int32, [2]>([1, 1])];
|
| 163 |
+
int32 var_419_groups_0 = const()[name = string("op_419_groups_0"), val = int32(1)];
|
| 164 |
+
tensor<fp16, [9496, 1024, 1, 1]> op_399_promoted_to_fp16_palettized = constexpr_lut_to_dense(indices = tensor<uint6, [9496, 1024, 1, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(111674944))), lut = tensor<fp16, [1187, 1, 1, 1, 64, 1]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(118967936))))[name = string("op_399_promoted_to_fp16_palettized")];
|
| 165 |
+
tensor<fp16, [1, 9496, 1, 1]> var_419_cast_fp16 = conv(dilations = var_419_dilations_0, groups = var_419_groups_0, pad = var_419_pad_0, pad_type = var_419_pad_type_0, strides = var_419_strides_0, weight = op_399_promoted_to_fp16_palettized, x = input_cast_fp16)[name = string("op_419_cast_fp16")];
|
| 166 |
+
tensor<int32, [1]> var_421_axes_0 = const()[name = string("op_421_axes_0"), val = tensor<int32, [1]>([2])];
|
| 167 |
+
tensor<fp16, [1, 9496, 1]> var_421_cast_fp16 = squeeze(axes = var_421_axes_0, x = var_419_cast_fp16)[name = string("op_421_cast_fp16")];
|
| 168 |
+
tensor<int32, [3]> logits_perm_0 = const()[name = string("logits_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
|
| 169 |
+
int32 chunk_argmax_1_axis_0 = const()[name = string("chunk_argmax_1_axis_0"), val = int32(-1)];
|
| 170 |
+
bool chunk_argmax_1_keep_dims_0 = const()[name = string("chunk_argmax_1_keep_dims_0"), val = bool(true)];
|
| 171 |
+
string chunk_argmax_1_output_dtype_0 = const()[name = string("chunk_argmax_1_output_dtype_0"), val = string("int32")];
|
| 172 |
+
tensor<fp16, [1, 1, 9496]> logits_1_cast_fp16 = transpose(perm = logits_1_perm_0, x = var_31_cast_fp16)[name = string("transpose_15")];
|
| 173 |
+
tensor<int32, [1, 1, 1]> chunk_argmax_1_cast_fp16 = reduce_argmax(axis = chunk_argmax_1_axis_0, keep_dims = chunk_argmax_1_keep_dims_0, output_dtype = chunk_argmax_1_output_dtype_0, x = logits_1_cast_fp16)[name = string("chunk_argmax_1_cast_fp16")];
|
| 174 |
+
int32 var_428 = const()[name = string("op_428"), val = int32(-1)];
|
| 175 |
+
bool var_430_validate_indices_0 = const()[name = string("op_430_validate_indices_0"), val = bool(false)];
|
| 176 |
+
string chunk_argmax_1_cast_fp16_to_uint16_dtype_0 = const()[name = string("chunk_argmax_1_cast_fp16_to_uint16_dtype_0"), val = string("uint16")];
|
| 177 |
+
tensor<uint16, [1, 1, 1]> chunk_argmax_1_cast_fp16_to_uint16 = cast(dtype = chunk_argmax_1_cast_fp16_to_uint16_dtype_0, x = chunk_argmax_1_cast_fp16)[name = string("cast_17")];
|
| 178 |
+
tensor<fp16, [1, 1, 1]> var_430_cast_fp16_cast_int16 = gather_along_axis(axis = var_428, indices = chunk_argmax_1_cast_fp16_to_uint16, validate_indices = var_430_validate_indices_0, x = logits_1_cast_fp16)[name = string("op_430_cast_fp16_cast_int16")];
|
| 179 |
+
int32 chunk_argmax_3_axis_0 = const()[name = string("chunk_argmax_3_axis_0"), val = int32(-1)];
|
| 180 |
+
bool chunk_argmax_3_keep_dims_0 = const()[name = string("chunk_argmax_3_keep_dims_0"), val = bool(true)];
|
| 181 |
+
string chunk_argmax_3_output_dtype_0 = const()[name = string("chunk_argmax_3_output_dtype_0"), val = string("int32")];
|
| 182 |
+
tensor<fp16, [1, 1, 9496]> logits_3_cast_fp16 = transpose(perm = logits_3_perm_0, x = var_57_cast_fp16)[name = string("transpose_14")];
|
| 183 |
+
tensor<int32, [1, 1, 1]> chunk_argmax_3_cast_fp16 = reduce_argmax(axis = chunk_argmax_3_axis_0, keep_dims = chunk_argmax_3_keep_dims_0, output_dtype = chunk_argmax_3_output_dtype_0, x = logits_3_cast_fp16)[name = string("chunk_argmax_3_cast_fp16")];
|
| 184 |
+
int32 var_439 = const()[name = string("op_439"), val = int32(-1)];
|
| 185 |
+
bool var_441_validate_indices_0 = const()[name = string("op_441_validate_indices_0"), val = bool(false)];
|
| 186 |
+
string chunk_argmax_3_cast_fp16_to_int16_dtype_0 = const()[name = string("chunk_argmax_3_cast_fp16_to_int16_dtype_0"), val = string("int16")];
|
| 187 |
+
tensor<int16, [1, 1, 1]> chunk_argmax_3_cast_fp16_to_int16 = cast(dtype = chunk_argmax_3_cast_fp16_to_int16_dtype_0, x = chunk_argmax_3_cast_fp16)[name = string("cast_16")];
|
| 188 |
+
tensor<fp16, [1, 1, 1]> var_441_cast_fp16_cast_int16 = gather_along_axis(axis = var_439, indices = chunk_argmax_3_cast_fp16_to_int16, validate_indices = var_441_validate_indices_0, x = logits_3_cast_fp16)[name = string("op_441_cast_fp16_cast_int16")];
|
| 189 |
+
int32 chunk_argmax_5_axis_0 = const()[name = string("chunk_argmax_5_axis_0"), val = int32(-1)];
|
| 190 |
+
bool chunk_argmax_5_keep_dims_0 = const()[name = string("chunk_argmax_5_keep_dims_0"), val = bool(true)];
|
| 191 |
+
string chunk_argmax_5_output_dtype_0 = const()[name = string("chunk_argmax_5_output_dtype_0"), val = string("int32")];
|
| 192 |
+
tensor<fp16, [1, 1, 9496]> logits_5_cast_fp16 = transpose(perm = logits_5_perm_0, x = var_83_cast_fp16)[name = string("transpose_13")];
|
| 193 |
+
tensor<int32, [1, 1, 1]> chunk_argmax_5_cast_fp16 = reduce_argmax(axis = chunk_argmax_5_axis_0, keep_dims = chunk_argmax_5_keep_dims_0, output_dtype = chunk_argmax_5_output_dtype_0, x = logits_5_cast_fp16)[name = string("chunk_argmax_5_cast_fp16")];
|
| 194 |
+
int32 var_450 = const()[name = string("op_450"), val = int32(-1)];
|
| 195 |
+
bool var_452_validate_indices_0 = const()[name = string("op_452_validate_indices_0"), val = bool(false)];
|
| 196 |
+
string chunk_argmax_5_cast_fp16_to_int16_dtype_0 = const()[name = string("chunk_argmax_5_cast_fp16_to_int16_dtype_0"), val = string("int16")];
|
| 197 |
+
tensor<int16, [1, 1, 1]> chunk_argmax_5_cast_fp16_to_int16 = cast(dtype = chunk_argmax_5_cast_fp16_to_int16_dtype_0, x = chunk_argmax_5_cast_fp16)[name = string("cast_15")];
|
| 198 |
+
tensor<fp16, [1, 1, 1]> var_452_cast_fp16_cast_int16 = gather_along_axis(axis = var_450, indices = chunk_argmax_5_cast_fp16_to_int16, validate_indices = var_452_validate_indices_0, x = logits_5_cast_fp16)[name = string("op_452_cast_fp16_cast_int16")];
|
| 199 |
+
int32 chunk_argmax_7_axis_0 = const()[name = string("chunk_argmax_7_axis_0"), val = int32(-1)];
|
| 200 |
+
bool chunk_argmax_7_keep_dims_0 = const()[name = string("chunk_argmax_7_keep_dims_0"), val = bool(true)];
|
| 201 |
+
string chunk_argmax_7_output_dtype_0 = const()[name = string("chunk_argmax_7_output_dtype_0"), val = string("int32")];
|
| 202 |
+
tensor<fp16, [1, 1, 9496]> logits_7_cast_fp16 = transpose(perm = logits_7_perm_0, x = var_109_cast_fp16)[name = string("transpose_12")];
|
| 203 |
+
tensor<int32, [1, 1, 1]> chunk_argmax_7_cast_fp16 = reduce_argmax(axis = chunk_argmax_7_axis_0, keep_dims = chunk_argmax_7_keep_dims_0, output_dtype = chunk_argmax_7_output_dtype_0, x = logits_7_cast_fp16)[name = string("chunk_argmax_7_cast_fp16")];
|
| 204 |
+
int32 var_461 = const()[name = string("op_461"), val = int32(-1)];
|
| 205 |
+
bool var_463_validate_indices_0 = const()[name = string("op_463_validate_indices_0"), val = bool(false)];
|
| 206 |
+
string chunk_argmax_7_cast_fp16_to_int16_dtype_0 = const()[name = string("chunk_argmax_7_cast_fp16_to_int16_dtype_0"), val = string("int16")];
|
| 207 |
+
tensor<int16, [1, 1, 1]> chunk_argmax_7_cast_fp16_to_int16 = cast(dtype = chunk_argmax_7_cast_fp16_to_int16_dtype_0, x = chunk_argmax_7_cast_fp16)[name = string("cast_14")];
|
| 208 |
+
tensor<fp16, [1, 1, 1]> var_463_cast_fp16_cast_int16 = gather_along_axis(axis = var_461, indices = chunk_argmax_7_cast_fp16_to_int16, validate_indices = var_463_validate_indices_0, x = logits_7_cast_fp16)[name = string("op_463_cast_fp16_cast_int16")];
|
| 209 |
+
int32 chunk_argmax_9_axis_0 = const()[name = string("chunk_argmax_9_axis_0"), val = int32(-1)];
|
| 210 |
+
bool chunk_argmax_9_keep_dims_0 = const()[name = string("chunk_argmax_9_keep_dims_0"), val = bool(true)];
|
| 211 |
+
string chunk_argmax_9_output_dtype_0 = const()[name = string("chunk_argmax_9_output_dtype_0"), val = string("int32")];
|
| 212 |
+
tensor<fp16, [1, 1, 9496]> logits_9_cast_fp16 = transpose(perm = logits_9_perm_0, x = var_135_cast_fp16)[name = string("transpose_11")];
|
| 213 |
+
tensor<int32, [1, 1, 1]> chunk_argmax_9_cast_fp16 = reduce_argmax(axis = chunk_argmax_9_axis_0, keep_dims = chunk_argmax_9_keep_dims_0, output_dtype = chunk_argmax_9_output_dtype_0, x = logits_9_cast_fp16)[name = string("chunk_argmax_9_cast_fp16")];
|
| 214 |
+
int32 var_472 = const()[name = string("op_472"), val = int32(-1)];
|
| 215 |
+
bool var_474_validate_indices_0 = const()[name = string("op_474_validate_indices_0"), val = bool(false)];
|
| 216 |
+
string chunk_argmax_9_cast_fp16_to_int16_dtype_0 = const()[name = string("chunk_argmax_9_cast_fp16_to_int16_dtype_0"), val = string("int16")];
|
| 217 |
+
tensor<int16, [1, 1, 1]> chunk_argmax_9_cast_fp16_to_int16 = cast(dtype = chunk_argmax_9_cast_fp16_to_int16_dtype_0, x = chunk_argmax_9_cast_fp16)[name = string("cast_13")];
|
| 218 |
+
tensor<fp16, [1, 1, 1]> var_474_cast_fp16_cast_int16 = gather_along_axis(axis = var_472, indices = chunk_argmax_9_cast_fp16_to_int16, validate_indices = var_474_validate_indices_0, x = logits_9_cast_fp16)[name = string("op_474_cast_fp16_cast_int16")];
|
| 219 |
+
int32 chunk_argmax_11_axis_0 = const()[name = string("chunk_argmax_11_axis_0"), val = int32(-1)];
|
| 220 |
+
bool chunk_argmax_11_keep_dims_0 = const()[name = string("chunk_argmax_11_keep_dims_0"), val = bool(true)];
|
| 221 |
+
string chunk_argmax_11_output_dtype_0 = const()[name = string("chunk_argmax_11_output_dtype_0"), val = string("int32")];
|
| 222 |
+
tensor<fp16, [1, 1, 9496]> logits_11_cast_fp16 = transpose(perm = logits_11_perm_0, x = var_161_cast_fp16)[name = string("transpose_10")];
|
| 223 |
+
tensor<int32, [1, 1, 1]> chunk_argmax_11_cast_fp16 = reduce_argmax(axis = chunk_argmax_11_axis_0, keep_dims = chunk_argmax_11_keep_dims_0, output_dtype = chunk_argmax_11_output_dtype_0, x = logits_11_cast_fp16)[name = string("chunk_argmax_11_cast_fp16")];
|
| 224 |
+
int32 var_483 = const()[name = string("op_483"), val = int32(-1)];
|
| 225 |
+
bool var_485_validate_indices_0 = const()[name = string("op_485_validate_indices_0"), val = bool(false)];
|
| 226 |
+
string chunk_argmax_11_cast_fp16_to_int16_dtype_0 = const()[name = string("chunk_argmax_11_cast_fp16_to_int16_dtype_0"), val = string("int16")];
|
| 227 |
+
tensor<int16, [1, 1, 1]> chunk_argmax_11_cast_fp16_to_int16 = cast(dtype = chunk_argmax_11_cast_fp16_to_int16_dtype_0, x = chunk_argmax_11_cast_fp16)[name = string("cast_12")];
|
| 228 |
+
tensor<fp16, [1, 1, 1]> var_485_cast_fp16_cast_int16 = gather_along_axis(axis = var_483, indices = chunk_argmax_11_cast_fp16_to_int16, validate_indices = var_485_validate_indices_0, x = logits_11_cast_fp16)[name = string("op_485_cast_fp16_cast_int16")];
|
| 229 |
+
int32 chunk_argmax_13_axis_0 = const()[name = string("chunk_argmax_13_axis_0"), val = int32(-1)];
|
| 230 |
+
bool chunk_argmax_13_keep_dims_0 = const()[name = string("chunk_argmax_13_keep_dims_0"), val = bool(true)];
|
| 231 |
+
string chunk_argmax_13_output_dtype_0 = const()[name = string("chunk_argmax_13_output_dtype_0"), val = string("int32")];
|
| 232 |
+
tensor<fp16, [1, 1, 9496]> logits_13_cast_fp16 = transpose(perm = logits_13_perm_0, x = var_187_cast_fp16)[name = string("transpose_9")];
|
| 233 |
+
tensor<int32, [1, 1, 1]> chunk_argmax_13_cast_fp16 = reduce_argmax(axis = chunk_argmax_13_axis_0, keep_dims = chunk_argmax_13_keep_dims_0, output_dtype = chunk_argmax_13_output_dtype_0, x = logits_13_cast_fp16)[name = string("chunk_argmax_13_cast_fp16")];
|
| 234 |
+
int32 var_494 = const()[name = string("op_494"), val = int32(-1)];
|
| 235 |
+
bool var_496_validate_indices_0 = const()[name = string("op_496_validate_indices_0"), val = bool(false)];
|
| 236 |
+
string chunk_argmax_13_cast_fp16_to_int16_dtype_0 = const()[name = string("chunk_argmax_13_cast_fp16_to_int16_dtype_0"), val = string("int16")];
|
| 237 |
+
tensor<int16, [1, 1, 1]> chunk_argmax_13_cast_fp16_to_int16 = cast(dtype = chunk_argmax_13_cast_fp16_to_int16_dtype_0, x = chunk_argmax_13_cast_fp16)[name = string("cast_11")];
|
| 238 |
+
tensor<fp16, [1, 1, 1]> var_496_cast_fp16_cast_int16 = gather_along_axis(axis = var_494, indices = chunk_argmax_13_cast_fp16_to_int16, validate_indices = var_496_validate_indices_0, x = logits_13_cast_fp16)[name = string("op_496_cast_fp16_cast_int16")];
|
| 239 |
+
int32 chunk_argmax_15_axis_0 = const()[name = string("chunk_argmax_15_axis_0"), val = int32(-1)];
|
| 240 |
+
bool chunk_argmax_15_keep_dims_0 = const()[name = string("chunk_argmax_15_keep_dims_0"), val = bool(true)];
|
| 241 |
+
string chunk_argmax_15_output_dtype_0 = const()[name = string("chunk_argmax_15_output_dtype_0"), val = string("int32")];
|
| 242 |
+
tensor<fp16, [1, 1, 9496]> logits_15_cast_fp16 = transpose(perm = logits_15_perm_0, x = var_213_cast_fp16)[name = string("transpose_8")];
|
| 243 |
+
tensor<int32, [1, 1, 1]> chunk_argmax_15_cast_fp16 = reduce_argmax(axis = chunk_argmax_15_axis_0, keep_dims = chunk_argmax_15_keep_dims_0, output_dtype = chunk_argmax_15_output_dtype_0, x = logits_15_cast_fp16)[name = string("chunk_argmax_15_cast_fp16")];
|
| 244 |
+
int32 var_505 = const()[name = string("op_505"), val = int32(-1)];
|
| 245 |
+
bool var_507_validate_indices_0 = const()[name = string("op_507_validate_indices_0"), val = bool(false)];
|
| 246 |
+
string chunk_argmax_15_cast_fp16_to_int16_dtype_0 = const()[name = string("chunk_argmax_15_cast_fp16_to_int16_dtype_0"), val = string("int16")];
|
| 247 |
+
tensor<int16, [1, 1, 1]> chunk_argmax_15_cast_fp16_to_int16 = cast(dtype = chunk_argmax_15_cast_fp16_to_int16_dtype_0, x = chunk_argmax_15_cast_fp16)[name = string("cast_10")];
|
| 248 |
+
tensor<fp16, [1, 1, 1]> var_507_cast_fp16_cast_int16 = gather_along_axis(axis = var_505, indices = chunk_argmax_15_cast_fp16_to_int16, validate_indices = var_507_validate_indices_0, x = logits_15_cast_fp16)[name = string("op_507_cast_fp16_cast_int16")];
|
| 249 |
+
int32 chunk_argmax_17_axis_0 = const()[name = string("chunk_argmax_17_axis_0"), val = int32(-1)];
|
| 250 |
+
bool chunk_argmax_17_keep_dims_0 = const()[name = string("chunk_argmax_17_keep_dims_0"), val = bool(true)];
|
| 251 |
+
string chunk_argmax_17_output_dtype_0 = const()[name = string("chunk_argmax_17_output_dtype_0"), val = string("int32")];
|
| 252 |
+
tensor<fp16, [1, 1, 9496]> logits_17_cast_fp16 = transpose(perm = logits_17_perm_0, x = var_239_cast_fp16)[name = string("transpose_7")];
|
| 253 |
+
tensor<int32, [1, 1, 1]> chunk_argmax_17_cast_fp16 = reduce_argmax(axis = chunk_argmax_17_axis_0, keep_dims = chunk_argmax_17_keep_dims_0, output_dtype = chunk_argmax_17_output_dtype_0, x = logits_17_cast_fp16)[name = string("chunk_argmax_17_cast_fp16")];
|
| 254 |
+
int32 var_516 = const()[name = string("op_516"), val = int32(-1)];
|
| 255 |
+
bool var_518_validate_indices_0 = const()[name = string("op_518_validate_indices_0"), val = bool(false)];
|
| 256 |
+
string chunk_argmax_17_cast_fp16_to_int16_dtype_0 = const()[name = string("chunk_argmax_17_cast_fp16_to_int16_dtype_0"), val = string("int16")];
|
| 257 |
+
tensor<int16, [1, 1, 1]> chunk_argmax_17_cast_fp16_to_int16 = cast(dtype = chunk_argmax_17_cast_fp16_to_int16_dtype_0, x = chunk_argmax_17_cast_fp16)[name = string("cast_9")];
|
| 258 |
+
tensor<fp16, [1, 1, 1]> var_518_cast_fp16_cast_int16 = gather_along_axis(axis = var_516, indices = chunk_argmax_17_cast_fp16_to_int16, validate_indices = var_518_validate_indices_0, x = logits_17_cast_fp16)[name = string("op_518_cast_fp16_cast_int16")];
|
| 259 |
+
int32 chunk_argmax_19_axis_0 = const()[name = string("chunk_argmax_19_axis_0"), val = int32(-1)];
|
| 260 |
+
bool chunk_argmax_19_keep_dims_0 = const()[name = string("chunk_argmax_19_keep_dims_0"), val = bool(true)];
|
| 261 |
+
string chunk_argmax_19_output_dtype_0 = const()[name = string("chunk_argmax_19_output_dtype_0"), val = string("int32")];
|
| 262 |
+
tensor<fp16, [1, 1, 9496]> logits_19_cast_fp16 = transpose(perm = logits_19_perm_0, x = var_265_cast_fp16)[name = string("transpose_6")];
|
| 263 |
+
tensor<int32, [1, 1, 1]> chunk_argmax_19_cast_fp16 = reduce_argmax(axis = chunk_argmax_19_axis_0, keep_dims = chunk_argmax_19_keep_dims_0, output_dtype = chunk_argmax_19_output_dtype_0, x = logits_19_cast_fp16)[name = string("chunk_argmax_19_cast_fp16")];
|
| 264 |
+
int32 var_527 = const()[name = string("op_527"), val = int32(-1)];
|
| 265 |
+
bool var_529_validate_indices_0 = const()[name = string("op_529_validate_indices_0"), val = bool(false)];
|
| 266 |
+
string chunk_argmax_19_cast_fp16_to_int16_dtype_0 = const()[name = string("chunk_argmax_19_cast_fp16_to_int16_dtype_0"), val = string("int16")];
|
| 267 |
+
tensor<int16, [1, 1, 1]> chunk_argmax_19_cast_fp16_to_int16 = cast(dtype = chunk_argmax_19_cast_fp16_to_int16_dtype_0, x = chunk_argmax_19_cast_fp16)[name = string("cast_8")];
|
| 268 |
+
tensor<fp16, [1, 1, 1]> var_529_cast_fp16_cast_int16 = gather_along_axis(axis = var_527, indices = chunk_argmax_19_cast_fp16_to_int16, validate_indices = var_529_validate_indices_0, x = logits_19_cast_fp16)[name = string("op_529_cast_fp16_cast_int16")];
|
| 269 |
+
int32 chunk_argmax_21_axis_0 = const()[name = string("chunk_argmax_21_axis_0"), val = int32(-1)];
|
| 270 |
+
bool chunk_argmax_21_keep_dims_0 = const()[name = string("chunk_argmax_21_keep_dims_0"), val = bool(true)];
|
| 271 |
+
string chunk_argmax_21_output_dtype_0 = const()[name = string("chunk_argmax_21_output_dtype_0"), val = string("int32")];
|
| 272 |
+
tensor<fp16, [1, 1, 9496]> logits_21_cast_fp16 = transpose(perm = logits_21_perm_0, x = var_291_cast_fp16)[name = string("transpose_5")];
|
| 273 |
+
tensor<int32, [1, 1, 1]> chunk_argmax_21_cast_fp16 = reduce_argmax(axis = chunk_argmax_21_axis_0, keep_dims = chunk_argmax_21_keep_dims_0, output_dtype = chunk_argmax_21_output_dtype_0, x = logits_21_cast_fp16)[name = string("chunk_argmax_21_cast_fp16")];
|
| 274 |
+
int32 var_538 = const()[name = string("op_538"), val = int32(-1)];
|
| 275 |
+
bool var_540_validate_indices_0 = const()[name = string("op_540_validate_indices_0"), val = bool(false)];
|
| 276 |
+
string chunk_argmax_21_cast_fp16_to_int16_dtype_0 = const()[name = string("chunk_argmax_21_cast_fp16_to_int16_dtype_0"), val = string("int16")];
|
| 277 |
+
tensor<int16, [1, 1, 1]> chunk_argmax_21_cast_fp16_to_int16 = cast(dtype = chunk_argmax_21_cast_fp16_to_int16_dtype_0, x = chunk_argmax_21_cast_fp16)[name = string("cast_7")];
|
| 278 |
+
tensor<fp16, [1, 1, 1]> var_540_cast_fp16_cast_int16 = gather_along_axis(axis = var_538, indices = chunk_argmax_21_cast_fp16_to_int16, validate_indices = var_540_validate_indices_0, x = logits_21_cast_fp16)[name = string("op_540_cast_fp16_cast_int16")];
|
| 279 |
+
int32 chunk_argmax_23_axis_0 = const()[name = string("chunk_argmax_23_axis_0"), val = int32(-1)];
|
| 280 |
+
bool chunk_argmax_23_keep_dims_0 = const()[name = string("chunk_argmax_23_keep_dims_0"), val = bool(true)];
|
| 281 |
+
string chunk_argmax_23_output_dtype_0 = const()[name = string("chunk_argmax_23_output_dtype_0"), val = string("int32")];
|
| 282 |
+
tensor<fp16, [1, 1, 9496]> logits_23_cast_fp16 = transpose(perm = logits_23_perm_0, x = var_317_cast_fp16)[name = string("transpose_4")];
|
| 283 |
+
tensor<int32, [1, 1, 1]> chunk_argmax_23_cast_fp16 = reduce_argmax(axis = chunk_argmax_23_axis_0, keep_dims = chunk_argmax_23_keep_dims_0, output_dtype = chunk_argmax_23_output_dtype_0, x = logits_23_cast_fp16)[name = string("chunk_argmax_23_cast_fp16")];
|
| 284 |
+
int32 var_549 = const()[name = string("op_549"), val = int32(-1)];
|
| 285 |
+
bool var_551_validate_indices_0 = const()[name = string("op_551_validate_indices_0"), val = bool(false)];
|
| 286 |
+
string chunk_argmax_23_cast_fp16_to_int16_dtype_0 = const()[name = string("chunk_argmax_23_cast_fp16_to_int16_dtype_0"), val = string("int16")];
|
| 287 |
+
tensor<int16, [1, 1, 1]> chunk_argmax_23_cast_fp16_to_int16 = cast(dtype = chunk_argmax_23_cast_fp16_to_int16_dtype_0, x = chunk_argmax_23_cast_fp16)[name = string("cast_6")];
|
| 288 |
+
tensor<fp16, [1, 1, 1]> var_551_cast_fp16_cast_int16 = gather_along_axis(axis = var_549, indices = chunk_argmax_23_cast_fp16_to_int16, validate_indices = var_551_validate_indices_0, x = logits_23_cast_fp16)[name = string("op_551_cast_fp16_cast_int16")];
|
| 289 |
+
int32 chunk_argmax_25_axis_0 = const()[name = string("chunk_argmax_25_axis_0"), val = int32(-1)];
|
| 290 |
+
bool chunk_argmax_25_keep_dims_0 = const()[name = string("chunk_argmax_25_keep_dims_0"), val = bool(true)];
|
| 291 |
+
string chunk_argmax_25_output_dtype_0 = const()[name = string("chunk_argmax_25_output_dtype_0"), val = string("int32")];
|
| 292 |
+
tensor<fp16, [1, 1, 9496]> logits_25_cast_fp16 = transpose(perm = logits_25_perm_0, x = var_343_cast_fp16)[name = string("transpose_3")];
|
| 293 |
+
tensor<int32, [1, 1, 1]> chunk_argmax_25_cast_fp16 = reduce_argmax(axis = chunk_argmax_25_axis_0, keep_dims = chunk_argmax_25_keep_dims_0, output_dtype = chunk_argmax_25_output_dtype_0, x = logits_25_cast_fp16)[name = string("chunk_argmax_25_cast_fp16")];
|
| 294 |
+
int32 var_560 = const()[name = string("op_560"), val = int32(-1)];
|
| 295 |
+
bool var_562_validate_indices_0 = const()[name = string("op_562_validate_indices_0"), val = bool(false)];
|
| 296 |
+
string chunk_argmax_25_cast_fp16_to_int16_dtype_0 = const()[name = string("chunk_argmax_25_cast_fp16_to_int16_dtype_0"), val = string("int16")];
|
| 297 |
+
tensor<int16, [1, 1, 1]> chunk_argmax_25_cast_fp16_to_int16 = cast(dtype = chunk_argmax_25_cast_fp16_to_int16_dtype_0, x = chunk_argmax_25_cast_fp16)[name = string("cast_5")];
|
| 298 |
+
tensor<fp16, [1, 1, 1]> var_562_cast_fp16_cast_int16 = gather_along_axis(axis = var_560, indices = chunk_argmax_25_cast_fp16_to_int16, validate_indices = var_562_validate_indices_0, x = logits_25_cast_fp16)[name = string("op_562_cast_fp16_cast_int16")];
|
| 299 |
+
int32 chunk_argmax_27_axis_0 = const()[name = string("chunk_argmax_27_axis_0"), val = int32(-1)];
|
| 300 |
+
bool chunk_argmax_27_keep_dims_0 = const()[name = string("chunk_argmax_27_keep_dims_0"), val = bool(true)];
|
| 301 |
+
string chunk_argmax_27_output_dtype_0 = const()[name = string("chunk_argmax_27_output_dtype_0"), val = string("int32")];
|
| 302 |
+
tensor<fp16, [1, 1, 9496]> logits_27_cast_fp16 = transpose(perm = logits_27_perm_0, x = var_369_cast_fp16)[name = string("transpose_2")];
|
| 303 |
+
tensor<int32, [1, 1, 1]> chunk_argmax_27_cast_fp16 = reduce_argmax(axis = chunk_argmax_27_axis_0, keep_dims = chunk_argmax_27_keep_dims_0, output_dtype = chunk_argmax_27_output_dtype_0, x = logits_27_cast_fp16)[name = string("chunk_argmax_27_cast_fp16")];
|
| 304 |
+
int32 var_571 = const()[name = string("op_571"), val = int32(-1)];
|
| 305 |
+
bool var_573_validate_indices_0 = const()[name = string("op_573_validate_indices_0"), val = bool(false)];
|
| 306 |
+
string chunk_argmax_27_cast_fp16_to_int16_dtype_0 = const()[name = string("chunk_argmax_27_cast_fp16_to_int16_dtype_0"), val = string("int16")];
|
| 307 |
+
tensor<int16, [1, 1, 1]> chunk_argmax_27_cast_fp16_to_int16 = cast(dtype = chunk_argmax_27_cast_fp16_to_int16_dtype_0, x = chunk_argmax_27_cast_fp16)[name = string("cast_4")];
|
| 308 |
+
tensor<fp16, [1, 1, 1]> var_573_cast_fp16_cast_int16 = gather_along_axis(axis = var_571, indices = chunk_argmax_27_cast_fp16_to_int16, validate_indices = var_573_validate_indices_0, x = logits_27_cast_fp16)[name = string("op_573_cast_fp16_cast_int16")];
|
| 309 |
+
int32 chunk_argmax_29_axis_0 = const()[name = string("chunk_argmax_29_axis_0"), val = int32(-1)];
|
| 310 |
+
bool chunk_argmax_29_keep_dims_0 = const()[name = string("chunk_argmax_29_keep_dims_0"), val = bool(true)];
|
| 311 |
+
string chunk_argmax_29_output_dtype_0 = const()[name = string("chunk_argmax_29_output_dtype_0"), val = string("int32")];
|
| 312 |
+
tensor<fp16, [1, 1, 9496]> logits_29_cast_fp16 = transpose(perm = logits_29_perm_0, x = var_395_cast_fp16)[name = string("transpose_1")];
|
| 313 |
+
tensor<int32, [1, 1, 1]> chunk_argmax_29_cast_fp16 = reduce_argmax(axis = chunk_argmax_29_axis_0, keep_dims = chunk_argmax_29_keep_dims_0, output_dtype = chunk_argmax_29_output_dtype_0, x = logits_29_cast_fp16)[name = string("chunk_argmax_29_cast_fp16")];
|
| 314 |
+
int32 var_582 = const()[name = string("op_582"), val = int32(-1)];
|
| 315 |
+
bool var_584_validate_indices_0 = const()[name = string("op_584_validate_indices_0"), val = bool(false)];
|
| 316 |
+
string chunk_argmax_29_cast_fp16_to_int16_dtype_0 = const()[name = string("chunk_argmax_29_cast_fp16_to_int16_dtype_0"), val = string("int16")];
|
| 317 |
+
tensor<int16, [1, 1, 1]> chunk_argmax_29_cast_fp16_to_int16 = cast(dtype = chunk_argmax_29_cast_fp16_to_int16_dtype_0, x = chunk_argmax_29_cast_fp16)[name = string("cast_3")];
|
| 318 |
+
tensor<fp16, [1, 1, 1]> var_584_cast_fp16_cast_int16 = gather_along_axis(axis = var_582, indices = chunk_argmax_29_cast_fp16_to_int16, validate_indices = var_584_validate_indices_0, x = logits_29_cast_fp16)[name = string("op_584_cast_fp16_cast_int16")];
|
| 319 |
+
int32 chunk_argmax_axis_0 = const()[name = string("chunk_argmax_axis_0"), val = int32(-1)];
|
| 320 |
+
bool chunk_argmax_keep_dims_0 = const()[name = string("chunk_argmax_keep_dims_0"), val = bool(true)];
|
| 321 |
+
string chunk_argmax_output_dtype_0 = const()[name = string("chunk_argmax_output_dtype_0"), val = string("int32")];
|
| 322 |
+
tensor<fp16, [1, 1, 9496]> logits_cast_fp16 = transpose(perm = logits_perm_0, x = var_421_cast_fp16)[name = string("transpose_0")];
|
| 323 |
+
tensor<int32, [1, 1, 1]> chunk_argmax_cast_fp16 = reduce_argmax(axis = chunk_argmax_axis_0, keep_dims = chunk_argmax_keep_dims_0, output_dtype = chunk_argmax_output_dtype_0, x = logits_cast_fp16)[name = string("chunk_argmax_cast_fp16")];
|
| 324 |
+
int32 var_593 = const()[name = string("op_593"), val = int32(-1)];
|
| 325 |
+
bool chunk_max_val_validate_indices_0 = const()[name = string("chunk_max_val_validate_indices_0"), val = bool(false)];
|
| 326 |
+
string chunk_argmax_cast_fp16_to_int16_dtype_0 = const()[name = string("chunk_argmax_cast_fp16_to_int16_dtype_0"), val = string("int16")];
|
| 327 |
+
tensor<int16, [1, 1, 1]> chunk_argmax_cast_fp16_to_int16 = cast(dtype = chunk_argmax_cast_fp16_to_int16_dtype_0, x = chunk_argmax_cast_fp16)[name = string("cast_2")];
|
| 328 |
+
tensor<fp16, [1, 1, 1]> chunk_max_val_cast_fp16_cast_int16 = gather_along_axis(axis = var_593, indices = chunk_argmax_cast_fp16_to_int16, validate_indices = chunk_max_val_validate_indices_0, x = logits_cast_fp16)[name = string("chunk_max_val_cast_fp16_cast_int16")];
|
| 329 |
+
int32 var_602 = const()[name = string("op_602"), val = int32(-1)];
|
| 330 |
+
bool var_603_interleave_0 = const()[name = string("op_603_interleave_0"), val = bool(false)];
|
| 331 |
+
tensor<int32, [1, 1, 16]> var_603 = concat(axis = var_602, interleave = var_603_interleave_0, values = (chunk_argmax_1_cast_fp16, chunk_argmax_3_cast_fp16, chunk_argmax_5_cast_fp16, chunk_argmax_7_cast_fp16, chunk_argmax_9_cast_fp16, chunk_argmax_11_cast_fp16, chunk_argmax_13_cast_fp16, chunk_argmax_15_cast_fp16, chunk_argmax_17_cast_fp16, chunk_argmax_19_cast_fp16, chunk_argmax_21_cast_fp16, chunk_argmax_23_cast_fp16, chunk_argmax_25_cast_fp16, chunk_argmax_27_cast_fp16, chunk_argmax_29_cast_fp16, chunk_argmax_cast_fp16))[name = string("op_603")];
|
| 332 |
+
tensor<int32, [1]> var_605_axes_0 = const()[name = string("op_605_axes_0"), val = tensor<int32, [1]>([0])];
|
| 333 |
+
string var_603_to_int16_dtype_0 = const()[name = string("op_603_to_int16_dtype_0"), val = string("int16")];
|
| 334 |
+
tensor<int16, [1, 1, 16]> var_603_to_int16 = cast(dtype = var_603_to_int16_dtype_0, x = var_603)[name = string("cast_1")];
|
| 335 |
+
tensor<int16, [1, 16]> var_605_cast_uint16 = squeeze(axes = var_605_axes_0, x = var_603_to_int16)[name = string("op_605_cast_uint16")];
|
| 336 |
+
tensor<int32, [1]> var_607_axes_0 = const()[name = string("op_607_axes_0"), val = tensor<int32, [1]>([0])];
|
| 337 |
+
tensor<int16, [16]> var_607_cast_uint16 = squeeze(axes = var_607_axes_0, x = var_605_cast_uint16)[name = string("op_607_cast_uint16")];
|
| 338 |
+
string var_607_cast_uint16_to_int32_dtype_0 = const()[name = string("op_607_cast_uint16_to_int32_dtype_0"), val = string("int32")];
|
| 339 |
+
int32 var_609 = const()[name = string("op_609"), val = int32(-1)];
|
| 340 |
+
bool var_610_interleave_0 = const()[name = string("op_610_interleave_0"), val = bool(false)];
|
| 341 |
+
tensor<fp16, [1, 1, 16]> var_610_cast_fp16 = concat(axis = var_609, interleave = var_610_interleave_0, values = (var_430_cast_fp16_cast_int16, var_441_cast_fp16_cast_int16, var_452_cast_fp16_cast_int16, var_463_cast_fp16_cast_int16, var_474_cast_fp16_cast_int16, var_485_cast_fp16_cast_int16, var_496_cast_fp16_cast_int16, var_507_cast_fp16_cast_int16, var_518_cast_fp16_cast_int16, var_529_cast_fp16_cast_int16, var_540_cast_fp16_cast_int16, var_551_cast_fp16_cast_int16, var_562_cast_fp16_cast_int16, var_573_cast_fp16_cast_int16, var_584_cast_fp16_cast_int16, chunk_max_val_cast_fp16_cast_int16))[name = string("op_610_cast_fp16")];
|
| 342 |
+
tensor<int32, [1]> var_612_axes_0 = const()[name = string("op_612_axes_0"), val = tensor<int32, [1]>([0])];
|
| 343 |
+
tensor<fp16, [1, 16]> var_612_cast_fp16 = squeeze(axes = var_612_axes_0, x = var_610_cast_fp16)[name = string("op_612_cast_fp16")];
|
| 344 |
+
tensor<int32, [1]> var_614_axes_0 = const()[name = string("op_614_axes_0"), val = tensor<int32, [1]>([0])];
|
| 345 |
+
tensor<fp16, [16]> argmax_val = squeeze(axes = var_614_axes_0, x = var_612_cast_fp16)[name = string("op_614_cast_fp16")];
|
| 346 |
+
tensor<int32, [16]> argmax_idx = cast(dtype = var_607_cast_uint16_to_int32_dtype_0, x = var_607_cast_uint16)[name = string("cast_0")];
|
| 347 |
+
} -> (argmax_idx, argmax_val);
|
| 348 |
+
}
|
lm_head.mlmodelc/weights/weight.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b127cf8561cd11d1f1258b14124989bf59da03baf53cb9fddb10e435dc468c64
|
| 3 |
+
size 119119936
|
manifest.json
ADDED
|
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architecture": "qwen3-asr-0.6b",
|
| 3 |
+
"artifactSha256Trees": {
|
| 4 |
+
"audio_3000.mlmodelc": "a6c28b5845348c4b92c8e5a6bca6e4c9ff9c74b5a650176dafd456ed0ab1ee11",
|
| 5 |
+
"audio_6000.mlmodelc": "eae21c8ccfeae56e0d93859bbf9bcacab7af77e3e89deea2115545587970e30b",
|
| 6 |
+
"decoder.mlmodelc": "0405ad1a402e66164b7bccfac06e050fd3cfc13d86f9d17d1a1d081dc4b7d201",
|
| 7 |
+
"embeddings.mlmodelc": "7c011fa215bd54b365cbfc284f137bba3667361b81dc8c5bdd663bdc7b6e1a53",
|
| 8 |
+
"lm_head.mlmodelc": "9abf62c25e55d52178a9ce9261e59fefe55a3f5c233452e41e9e21a192e5c6ff"
|
| 9 |
+
},
|
| 10 |
+
"audioBlockTokens": 104,
|
| 11 |
+
"audioBuckets": [
|
| 12 |
+
{
|
| 13 |
+
"frames": 3000,
|
| 14 |
+
"model": "audio_3000.mlmodelc",
|
| 15 |
+
"tokens": 390
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"frames": 6000,
|
| 19 |
+
"model": "audio_6000.mlmodelc",
|
| 20 |
+
"tokens": 780
|
| 21 |
+
}
|
| 22 |
+
],
|
| 23 |
+
"contextLength": 1536,
|
| 24 |
+
"decoderModel": "decoder.mlmodelc",
|
| 25 |
+
"embeddingsModel": "embeddings.mlmodelc",
|
| 26 |
+
"hiddenSize": 1024,
|
| 27 |
+
"lmHeadChunkSizes": [
|
| 28 |
+
9496,
|
| 29 |
+
9496,
|
| 30 |
+
9496,
|
| 31 |
+
9496,
|
| 32 |
+
9496,
|
| 33 |
+
9496,
|
| 34 |
+
9496,
|
| 35 |
+
9496,
|
| 36 |
+
9496,
|
| 37 |
+
9496,
|
| 38 |
+
9496,
|
| 39 |
+
9496,
|
| 40 |
+
9496,
|
| 41 |
+
9496,
|
| 42 |
+
9496,
|
| 43 |
+
9496
|
| 44 |
+
],
|
| 45 |
+
"lmHeadModel": "lm_head.mlmodelc",
|
| 46 |
+
"maxOutputTokens": 448,
|
| 47 |
+
"prefillBatchSize": 64,
|
| 48 |
+
"schemaVersion": 1
|
| 49 |
+
}
|