Hugging Face's logo

Intel
/
distilbart-cnn-12-6-int8-dynamic-inc

Text2Text Generation
Transformers
PyTorch
bart
int8
Intel® Neural Compressor
neural-compressor
PostTrainingDynamic
Inference Endpoints
Model card Files Community
xinhe commited on
Commit
9418e53
•
1 Parent(s): afc3f72

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -0
README.md CHANGED
@@ -32,6 +32,7 @@ Below linear modules (21/133) are fallbacked to fp32 for less than 1% relative a
32
  ### Load with optimum:
33
 
34
  ```python
 
35
  from optimum.intel.neural_compressor.quantization import IncQuantizedModelForSeq2SeqLM
36
  int8_model = IncQuantizedModelForSeq2SeqLM.from_pretrained(
37
  'Intel/distilbart-cnn-12-6-int8-dynamic',
 
32
  ### Load with optimum:
33
 
34
  ```python
35
+ # transformers <= 4.23.0
36
  from optimum.intel.neural_compressor.quantization import IncQuantizedModelForSeq2SeqLM
37
  int8_model = IncQuantizedModelForSeq2SeqLM.from_pretrained(
38
  'Intel/distilbart-cnn-12-6-int8-dynamic',