add cpu int8 base.en model
Browse files
base.en/whisper_cpu_int8_cpu-cpu_metrics.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"latency-avg": {
|
3 |
+
"value": 2796.97338,
|
4 |
+
"priority": 1,
|
5 |
+
"higher_is_better": false
|
6 |
+
}
|
7 |
+
}
|
base.en/whisper_cpu_int8_cpu-cpu_model.json
ADDED
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"type": "ONNXModel",
|
3 |
+
"config": {
|
4 |
+
"model_path": "/content/Olive/examples/whisper/models/conversion-transformers_optimization-onnx_dynamic_quantization-insert_beam_search-prepost/whisper_cpu_int8_cpu-cpu_model.onnx",
|
5 |
+
"onnx_file_name": null,
|
6 |
+
"inference_settings": null,
|
7 |
+
"use_ort_extensions": true,
|
8 |
+
"hf_config": {
|
9 |
+
"model_name": "openai/whisper-base.en",
|
10 |
+
"task": null,
|
11 |
+
"feature": null,
|
12 |
+
"model_class": "WhisperForConditionalGeneration",
|
13 |
+
"components": [
|
14 |
+
{
|
15 |
+
"name": "encoder_decoder_init",
|
16 |
+
"io_config": "get_encdec_io_config",
|
17 |
+
"component_func": "get_encoder_decoder_init",
|
18 |
+
"dummy_inputs_func": "encoder_decoder_init_dummy_inputs"
|
19 |
+
},
|
20 |
+
{
|
21 |
+
"name": "decoder",
|
22 |
+
"io_config": "get_dec_io_config",
|
23 |
+
"component_func": "get_decoder",
|
24 |
+
"dummy_inputs_func": "decoder_dummy_inputs"
|
25 |
+
}
|
26 |
+
],
|
27 |
+
"dataset": null,
|
28 |
+
"model_loading_args": null
|
29 |
+
}
|
30 |
+
}
|
31 |
+
}
|
base.en/whisper_cpu_int8_cpu-cpu_model.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7c25f7dc48ec75d497a880fac41514b89486d11be07db3d681d50f60a88f16b1
|
3 |
+
size 140061412
|