echarlaix HF staff commited on
Commit
291f8c8
1 Parent(s): 1dc7724

Update loading instructions

Browse files
Files changed (1) hide show
  1. README.md +3 -3
README.md CHANGED
@@ -49,10 +49,10 @@ The calibration dataloader is the train dataloader. The default calibration samp
49
  #### Load with Intel® Neural Compressor:
50
 
51
  ```python
52
- from optimum.intel.neural_compressor import IncQuantizedModelForSequenceClassification
53
 
54
  model_id = "Intel/roberta-base-mrpc-int8-static"
55
- int8_model = IncQuantizedModelForSequenceClassification.from_pretrained(model_id)
56
  ```
57
 
58
  ### ONNX
@@ -76,4 +76,4 @@ The calibration dataloader is the eval dataloader. The calibration sampling size
76
  ```python
77
  from optimum.onnxruntime import ORTModelForSequenceClassification
78
  model = ORTModelForSequenceClassification.from_pretrained('Intel/roberta-base-mrpc-int8-static')
79
- ```
 
49
  #### Load with Intel® Neural Compressor:
50
 
51
  ```python
52
+ from optimum.intel import INCModelForSequenceClassification
53
 
54
  model_id = "Intel/roberta-base-mrpc-int8-static"
55
+ int8_model = INCModelForSequenceClassification.from_pretrained(model_id)
56
  ```
57
 
58
  ### ONNX
 
76
  ```python
77
  from optimum.onnxruntime import ORTModelForSequenceClassification
78
  model = ORTModelForSequenceClassification.from_pretrained('Intel/roberta-base-mrpc-int8-static')
79
+ ```