Update README.md
Browse files
README.md
CHANGED
@@ -74,7 +74,7 @@ for model in model_list:
|
|
74 |
"group_size": 32}}},
|
75 |
op_name_dict={'/proj_out/MatMul': FP32},) # fallback last matmul in decoder to FP32
|
76 |
q_model = quantization.fit(
|
77 |
-
os.path.join("/path/to/whisper-tiny", model), # FP32 model path
|
78 |
config,
|
79 |
calib_dataloader=dataloader)
|
80 |
q_model.save(os.path.join("/path/to/whisper-tiny-onnx-int4", model)) # INT4 model path
|
|
|
74 |
"group_size": 32}}},
|
75 |
op_name_dict={'/proj_out/MatMul': FP32},) # fallback last matmul in decoder to FP32
|
76 |
q_model = quantization.fit(
|
77 |
+
os.path.join("/path/to/whisper-tiny-with-past", model), # FP32 model path
|
78 |
config,
|
79 |
calib_dataloader=dataloader)
|
80 |
q_model.save(os.path.join("/path/to/whisper-tiny-onnx-int4", model)) # INT4 model path
|