Upload folder using huggingface_hub
Browse files- config.json +1 -1
 - generation_config.json +2 -2
 - onnx/decoder_model.onnx +2 -2
 - onnx/decoder_model_merged.onnx +2 -2
 - onnx/decoder_model_merged_quantized.onnx +2 -2
 - onnx/decoder_model_quantized.onnx +2 -2
 - onnx/decoder_with_past_model.onnx +2 -2
 - onnx/decoder_with_past_model_quantized.onnx +2 -2
 - quantize_config.json +75 -75
 - tokenizer.json +0 -0
 
    	
        config.json
    CHANGED
    
    | 
         @@ -54,7 +54,7 @@ 
     | 
|
| 54 | 
         
             
              "scale_embedding": true,
         
     | 
| 55 | 
         
             
              "share_encoder_decoder_embeddings": true,
         
     | 
| 56 | 
         
             
              "static_position_embeddings": true,
         
     | 
| 57 | 
         
            -
              "transformers_version": "4. 
     | 
| 58 | 
         
             
              "use_cache": true,
         
     | 
| 59 | 
         
             
              "vocab_size": 59659
         
     | 
| 60 | 
         
             
            }
         
     | 
| 
         | 
|
| 54 | 
         
             
              "scale_embedding": true,
         
     | 
| 55 | 
         
             
              "share_encoder_decoder_embeddings": true,
         
     | 
| 56 | 
         
             
              "static_position_embeddings": true,
         
     | 
| 57 | 
         
            +
              "transformers_version": "4.34.0.dev0",
         
     | 
| 58 | 
         
             
              "use_cache": true,
         
     | 
| 59 | 
         
             
              "vocab_size": 59659
         
     | 
| 60 | 
         
             
            }
         
     | 
    	
        generation_config.json
    CHANGED
    
    | 
         @@ -1,5 +1,4 @@ 
     | 
|
| 1 | 
         
             
            {
         
     | 
| 2 | 
         
            -
              "_from_model_config": true,
         
     | 
| 3 | 
         
             
              "bad_words_ids": [
         
     | 
| 4 | 
         
             
                [
         
     | 
| 5 | 
         
             
                  59658
         
     | 
| 
         @@ -12,5 +11,6 @@ 
     | 
|
| 12 | 
         
             
              "max_length": 512,
         
     | 
| 13 | 
         
             
              "num_beams": 4,
         
     | 
| 14 | 
         
             
              "pad_token_id": 59658,
         
     | 
| 15 | 
         
            -
              " 
     | 
| 
         | 
|
| 16 | 
         
             
            }
         
     | 
| 
         | 
|
| 1 | 
         
             
            {
         
     | 
| 
         | 
|
| 2 | 
         
             
              "bad_words_ids": [
         
     | 
| 3 | 
         
             
                [
         
     | 
| 4 | 
         
             
                  59658
         
     | 
| 
         | 
|
| 11 | 
         
             
              "max_length": 512,
         
     | 
| 12 | 
         
             
              "num_beams": 4,
         
     | 
| 13 | 
         
             
              "pad_token_id": 59658,
         
     | 
| 14 | 
         
            +
              "renormalize_logits": true,
         
     | 
| 15 | 
         
            +
              "transformers_version": "4.34.0.dev0"
         
     | 
| 16 | 
         
             
            }
         
     | 
    	
        onnx/decoder_model.onnx
    CHANGED
    
    | 
         @@ -1,3 +1,3 @@ 
     | 
|
| 1 | 
         
             
            version https://git-lfs.github.com/spec/v1
         
     | 
| 2 | 
         
            -
            oid sha256: 
     | 
| 3 | 
         
            -
            size  
     | 
| 
         | 
|
| 1 | 
         
             
            version https://git-lfs.github.com/spec/v1
         
     | 
| 2 | 
         
            +
            oid sha256:b29c232c224536e38fd892466c998748bfe217ea516d67a7f2ed7165a359caa7
         
     | 
| 3 | 
         
            +
            size 224641885
         
     | 
    	
        onnx/decoder_model_merged.onnx
    CHANGED
    
    | 
         @@ -1,3 +1,3 @@ 
     | 
|
| 1 | 
         
             
            version https://git-lfs.github.com/spec/v1
         
     | 
| 2 | 
         
            -
            oid sha256: 
     | 
| 3 | 
         
            -
            size  
     | 
| 
         | 
|
| 1 | 
         
             
            version https://git-lfs.github.com/spec/v1
         
     | 
| 2 | 
         
            +
            oid sha256:84c27d0c6f373ac705ecb227a88b034b56f811019f785cdd144bfda5696bb049
         
     | 
| 3 | 
         
            +
            size 224877452
         
     | 
    	
        onnx/decoder_model_merged_quantized.onnx
    CHANGED
    
    | 
         @@ -1,3 +1,3 @@ 
     | 
|
| 1 | 
         
             
            version https://git-lfs.github.com/spec/v1
         
     | 
| 2 | 
         
            -
            oid sha256: 
     | 
| 3 | 
         
            -
            size  
     | 
| 
         | 
|
| 1 | 
         
             
            version https://git-lfs.github.com/spec/v1
         
     | 
| 2 | 
         
            +
            oid sha256:320fb97fed0f0989a3a1a4777f77a8b236f41c2b59c4503f659f5eea5b5f01c6
         
     | 
| 3 | 
         
            +
            size 57456332
         
     | 
    	
        onnx/decoder_model_quantized.onnx
    CHANGED
    
    | 
         @@ -1,3 +1,3 @@ 
     | 
|
| 1 | 
         
             
            version https://git-lfs.github.com/spec/v1
         
     | 
| 2 | 
         
            -
            oid sha256: 
     | 
| 3 | 
         
            -
            size  
     | 
| 
         | 
|
| 1 | 
         
             
            version https://git-lfs.github.com/spec/v1
         
     | 
| 2 | 
         
            +
            oid sha256:d1001aff511a27a09f7f4d1406f7996c1b255fe9c5a40cc0deeb220bd675b615
         
     | 
| 3 | 
         
            +
            size 57085630
         
     | 
    	
        onnx/decoder_with_past_model.onnx
    CHANGED
    
    | 
         @@ -1,3 +1,3 @@ 
     | 
|
| 1 | 
         
             
            version https://git-lfs.github.com/spec/v1
         
     | 
| 2 | 
         
            -
            oid sha256: 
     | 
| 3 | 
         
            -
            size  
     | 
| 
         | 
|
| 1 | 
         
             
            version https://git-lfs.github.com/spec/v1
         
     | 
| 2 | 
         
            +
            oid sha256:13556762b7353b3e02c82229d77091e573b70ea2b11667175e6caa9687d29def
         
     | 
| 3 | 
         
            +
            size 211991376
         
     | 
    	
        onnx/decoder_with_past_model_quantized.onnx
    CHANGED
    
    | 
         @@ -1,3 +1,3 @@ 
     | 
|
| 1 | 
         
             
            version https://git-lfs.github.com/spec/v1
         
     | 
| 2 | 
         
            -
            oid sha256: 
     | 
| 3 | 
         
            -
            size  
     | 
| 
         | 
|
| 1 | 
         
             
            version https://git-lfs.github.com/spec/v1
         
     | 
| 2 | 
         
            +
            oid sha256:907c208852baca172c0dab218aa0cc9611b71d561e69a1be5eb007cedb8bb648
         
     | 
| 3 | 
         
            +
            size 53814166
         
     | 
    	
        quantize_config.json
    CHANGED
    
    | 
         @@ -4,120 +4,120 @@ 
     | 
|
| 4 | 
         
             
                "per_model_config": {
         
     | 
| 5 | 
         
             
                    "encoder_model": {
         
     | 
| 6 | 
         
             
                        "op_types": [
         
     | 
| 7 | 
         
            -
                            " 
     | 
| 8 | 
         
            -
                            " 
     | 
| 9 | 
         
            -
                            "Softmax",
         
     | 
| 10 | 
         
            -
                            "Sigmoid",
         
     | 
| 11 | 
         
            -
                            "Sqrt",
         
     | 
| 12 | 
         
            -
                            "MatMul",
         
     | 
| 13 | 
         
            -
                            "Where",
         
     | 
| 14 | 
         
             
                            "Cast",
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 15 | 
         
             
                            "Transpose",
         
     | 
| 
         | 
|
| 16 | 
         
             
                            "Mul",
         
     | 
| 17 | 
         
            -
                            " 
     | 
| 18 | 
         
            -
                            " 
     | 
| 19 | 
         
            -
                            "Constant",
         
     | 
| 20 | 
         
             
                            "Gather",
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 21 | 
         
             
                            "Div",
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 22 | 
         
             
                            "ConstantOfShape",
         
     | 
| 23 | 
         
            -
                            " 
     | 
| 24 | 
         
            -
                            " 
     | 
| 25 | 
         
            -
                            "ReduceMean",
         
     | 
| 26 | 
         
            -
                            "Reshape",
         
     | 
| 27 | 
         
            -
                            "Sub",
         
     | 
| 28 | 
         
            -
                            "Expand",
         
     | 
| 29 | 
         
            -
                            "Concat"
         
     | 
| 30 | 
         
             
                        ],
         
     | 
| 31 | 
         
             
                        "weight_type": "QInt8"
         
     | 
| 32 | 
         
             
                    },
         
     | 
| 33 | 
         
             
                    "decoder_with_past_model": {
         
     | 
| 34 | 
         
             
                        "op_types": [
         
     | 
| 35 | 
         
            -
                            " 
     | 
| 36 | 
         
            -
                            " 
     | 
| 37 | 
         
            -
                            "Softmax",
         
     | 
| 38 | 
         
            -
                            "Sigmoid",
         
     | 
| 39 | 
         
            -
                            "Sqrt",
         
     | 
| 40 | 
         
            -
                            "MatMul",
         
     | 
| 41 | 
         
            -
                            "Where",
         
     | 
| 42 | 
         
             
                            "Cast",
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 43 | 
         
             
                            "Transpose",
         
     | 
| 
         | 
|
| 44 | 
         
             
                            "Mul",
         
     | 
| 45 | 
         
            -
                            " 
     | 
| 46 | 
         
            -
                            " 
     | 
| 47 | 
         
            -
                            "Constant",
         
     | 
| 48 | 
         
             
                            "Gather",
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 49 | 
         
             
                            "Div",
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 50 | 
         
             
                            "ConstantOfShape",
         
     | 
| 51 | 
         
            -
                            " 
     | 
| 52 | 
         
            -
                            " 
     | 
| 53 | 
         
            -
                            "ReduceMean",
         
     | 
| 54 | 
         
            -
                            "Reshape",
         
     | 
| 55 | 
         
            -
                            "Sub",
         
     | 
| 56 | 
         
            -
                            "Expand",
         
     | 
| 57 | 
         
            -
                            "Concat"
         
     | 
| 58 | 
         
             
                        ],
         
     | 
| 59 | 
         
             
                        "weight_type": "QInt8"
         
     | 
| 60 | 
         
             
                    },
         
     | 
| 61 | 
         
            -
                    " 
     | 
| 62 | 
         
             
                        "op_types": [
         
     | 
| 63 | 
         
            -
                            " 
     | 
| 64 | 
         
            -
                            " 
     | 
| 65 | 
         
            -
                            "Slice",
         
     | 
| 66 | 
         
            -
                            "Softmax",
         
     | 
| 67 | 
         
            -
                            "Sigmoid",
         
     | 
| 68 | 
         
            -
                            "Sqrt",
         
     | 
| 69 | 
         
            -
                            "MatMul",
         
     | 
| 70 | 
         
            -
                            "Where",
         
     | 
| 71 | 
         
             
                            "Cast",
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 72 | 
         
             
                            "Transpose",
         
     | 
| 
         | 
|
| 73 | 
         
             
                            "Mul",
         
     | 
| 74 | 
         
            -
                            " 
     | 
| 75 | 
         
            -
                            "Equal",
         
     | 
| 76 | 
         
            -
                            "Constant",
         
     | 
| 77 | 
         
             
                            "Less",
         
     | 
| 
         | 
|
| 
         | 
|
| 78 | 
         
             
                            "Gather",
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 79 | 
         
             
                            "Div",
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 80 | 
         
             
                            "ConstantOfShape",
         
     | 
| 81 | 
         
            -
                            " 
     | 
| 82 | 
         
            -
                            " 
     | 
| 83 | 
         
            -
                            " 
     | 
| 84 | 
         
            -
                            "ReduceMean",
         
     | 
| 85 | 
         
            -
                            "Reshape",
         
     | 
| 86 | 
         
            -
                            "Sub",
         
     | 
| 87 | 
         
            -
                            "Expand",
         
     | 
| 88 | 
         
            -
                            "Concat"
         
     | 
| 89 | 
         
             
                        ],
         
     | 
| 90 | 
         
             
                        "weight_type": "QInt8"
         
     | 
| 91 | 
         
             
                    },
         
     | 
| 92 | 
         
            -
                    " 
     | 
| 93 | 
         
             
                        "op_types": [
         
     | 
| 94 | 
         
            -
                            " 
     | 
| 95 | 
         
            -
                            " 
     | 
| 96 | 
         
            -
                            "Slice",
         
     | 
| 97 | 
         
            -
                            "Softmax",
         
     | 
| 98 | 
         
            -
                            "Sigmoid",
         
     | 
| 99 | 
         
            -
                            "Sqrt",
         
     | 
| 100 | 
         
            -
                            "MatMul",
         
     | 
| 101 | 
         
            -
                            "Where",
         
     | 
| 102 | 
         
             
                            "Cast",
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 103 | 
         
             
                            "Transpose",
         
     | 
| 
         | 
|
| 104 | 
         
             
                            "Mul",
         
     | 
| 105 | 
         
            -
                            " 
     | 
| 106 | 
         
            -
                            "Equal",
         
     | 
| 107 | 
         
            -
                            "Constant",
         
     | 
| 108 | 
         
             
                            "Less",
         
     | 
| 
         | 
|
| 
         | 
|
| 109 | 
         
             
                            "Gather",
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 110 | 
         
             
                            "Div",
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 111 | 
         
             
                            "ConstantOfShape",
         
     | 
| 112 | 
         
            -
                            " 
     | 
| 113 | 
         
            -
                            " 
     | 
| 114 | 
         
            -
                            " 
     | 
| 115 | 
         
            -
                            "Add",
         
     | 
| 116 | 
         
            -
                            "ReduceMean",
         
     | 
| 117 | 
         
            -
                            "Reshape",
         
     | 
| 118 | 
         
            -
                            "Sub",
         
     | 
| 119 | 
         
            -
                            "Expand",
         
     | 
| 120 | 
         
            -
                            "Concat"
         
     | 
| 121 | 
         
             
                        ],
         
     | 
| 122 | 
         
             
                        "weight_type": "QInt8"
         
     | 
| 123 | 
         
             
                    }
         
     | 
| 
         | 
|
| 4 | 
         
             
                "per_model_config": {
         
     | 
| 5 | 
         
             
                    "encoder_model": {
         
     | 
| 6 | 
         
             
                        "op_types": [
         
     | 
| 7 | 
         
            +
                            "Expand",
         
     | 
| 8 | 
         
            +
                            "Reshape",
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 9 | 
         
             
                            "Cast",
         
     | 
| 10 | 
         
            +
                            "Range",
         
     | 
| 11 | 
         
            +
                            "ReduceMean",
         
     | 
| 12 | 
         
            +
                            "Constant",
         
     | 
| 13 | 
         
            +
                            "Where",
         
     | 
| 14 | 
         
             
                            "Transpose",
         
     | 
| 15 | 
         
            +
                            "Add",
         
     | 
| 16 | 
         
             
                            "Mul",
         
     | 
| 17 | 
         
            +
                            "Sigmoid",
         
     | 
| 18 | 
         
            +
                            "Sub",
         
     | 
| 
         | 
|
| 19 | 
         
             
                            "Gather",
         
     | 
| 20 | 
         
            +
                            "Concat",
         
     | 
| 21 | 
         
            +
                            "MatMul",
         
     | 
| 22 | 
         
            +
                            "Unsqueeze",
         
     | 
| 23 | 
         
             
                            "Div",
         
     | 
| 24 | 
         
            +
                            "Shape",
         
     | 
| 25 | 
         
            +
                            "Sqrt",
         
     | 
| 26 | 
         
            +
                            "Equal",
         
     | 
| 27 | 
         
             
                            "ConstantOfShape",
         
     | 
| 28 | 
         
            +
                            "Pow",
         
     | 
| 29 | 
         
            +
                            "Softmax"
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 30 | 
         
             
                        ],
         
     | 
| 31 | 
         
             
                        "weight_type": "QInt8"
         
     | 
| 32 | 
         
             
                    },
         
     | 
| 33 | 
         
             
                    "decoder_with_past_model": {
         
     | 
| 34 | 
         
             
                        "op_types": [
         
     | 
| 35 | 
         
            +
                            "Expand",
         
     | 
| 36 | 
         
            +
                            "Reshape",
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 37 | 
         
             
                            "Cast",
         
     | 
| 38 | 
         
            +
                            "Range",
         
     | 
| 39 | 
         
            +
                            "ReduceMean",
         
     | 
| 40 | 
         
            +
                            "Constant",
         
     | 
| 41 | 
         
            +
                            "Where",
         
     | 
| 42 | 
         
             
                            "Transpose",
         
     | 
| 43 | 
         
            +
                            "Add",
         
     | 
| 44 | 
         
             
                            "Mul",
         
     | 
| 45 | 
         
            +
                            "Sigmoid",
         
     | 
| 46 | 
         
            +
                            "Sub",
         
     | 
| 
         | 
|
| 47 | 
         
             
                            "Gather",
         
     | 
| 48 | 
         
            +
                            "Concat",
         
     | 
| 49 | 
         
            +
                            "MatMul",
         
     | 
| 50 | 
         
            +
                            "Unsqueeze",
         
     | 
| 51 | 
         
             
                            "Div",
         
     | 
| 52 | 
         
            +
                            "Shape",
         
     | 
| 53 | 
         
            +
                            "Sqrt",
         
     | 
| 54 | 
         
            +
                            "Equal",
         
     | 
| 55 | 
         
             
                            "ConstantOfShape",
         
     | 
| 56 | 
         
            +
                            "Pow",
         
     | 
| 57 | 
         
            +
                            "Softmax"
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 58 | 
         
             
                        ],
         
     | 
| 59 | 
         
             
                        "weight_type": "QInt8"
         
     | 
| 60 | 
         
             
                    },
         
     | 
| 61 | 
         
            +
                    "decoder_model_merged": {
         
     | 
| 62 | 
         
             
                        "op_types": [
         
     | 
| 63 | 
         
            +
                            "Expand",
         
     | 
| 64 | 
         
            +
                            "Reshape",
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 65 | 
         
             
                            "Cast",
         
     | 
| 66 | 
         
            +
                            "Range",
         
     | 
| 67 | 
         
            +
                            "ReduceMean",
         
     | 
| 68 | 
         
            +
                            "If",
         
     | 
| 69 | 
         
            +
                            "Constant",
         
     | 
| 70 | 
         
            +
                            "Where",
         
     | 
| 71 | 
         
             
                            "Transpose",
         
     | 
| 72 | 
         
            +
                            "Add",
         
     | 
| 73 | 
         
             
                            "Mul",
         
     | 
| 74 | 
         
            +
                            "Sigmoid",
         
     | 
| 
         | 
|
| 
         | 
|
| 75 | 
         
             
                            "Less",
         
     | 
| 76 | 
         
            +
                            "Sub",
         
     | 
| 77 | 
         
            +
                            "Slice",
         
     | 
| 78 | 
         
             
                            "Gather",
         
     | 
| 79 | 
         
            +
                            "Concat",
         
     | 
| 80 | 
         
            +
                            "MatMul",
         
     | 
| 81 | 
         
            +
                            "Unsqueeze",
         
     | 
| 82 | 
         
             
                            "Div",
         
     | 
| 83 | 
         
            +
                            "Shape",
         
     | 
| 84 | 
         
            +
                            "Sqrt",
         
     | 
| 85 | 
         
            +
                            "Equal",
         
     | 
| 86 | 
         
             
                            "ConstantOfShape",
         
     | 
| 87 | 
         
            +
                            "Pow",
         
     | 
| 88 | 
         
            +
                            "Softmax",
         
     | 
| 89 | 
         
            +
                            "Squeeze"
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 90 | 
         
             
                        ],
         
     | 
| 91 | 
         
             
                        "weight_type": "QInt8"
         
     | 
| 92 | 
         
             
                    },
         
     | 
| 93 | 
         
            +
                    "decoder_model": {
         
     | 
| 94 | 
         
             
                        "op_types": [
         
     | 
| 95 | 
         
            +
                            "Expand",
         
     | 
| 96 | 
         
            +
                            "Reshape",
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 97 | 
         
             
                            "Cast",
         
     | 
| 98 | 
         
            +
                            "Range",
         
     | 
| 99 | 
         
            +
                            "ReduceMean",
         
     | 
| 100 | 
         
            +
                            "Constant",
         
     | 
| 101 | 
         
            +
                            "Where",
         
     | 
| 102 | 
         
             
                            "Transpose",
         
     | 
| 103 | 
         
            +
                            "Add",
         
     | 
| 104 | 
         
             
                            "Mul",
         
     | 
| 105 | 
         
            +
                            "Sigmoid",
         
     | 
| 
         | 
|
| 
         | 
|
| 106 | 
         
             
                            "Less",
         
     | 
| 107 | 
         
            +
                            "Sub",
         
     | 
| 108 | 
         
            +
                            "Slice",
         
     | 
| 109 | 
         
             
                            "Gather",
         
     | 
| 110 | 
         
            +
                            "Concat",
         
     | 
| 111 | 
         
            +
                            "MatMul",
         
     | 
| 112 | 
         
            +
                            "Unsqueeze",
         
     | 
| 113 | 
         
             
                            "Div",
         
     | 
| 114 | 
         
            +
                            "Shape",
         
     | 
| 115 | 
         
            +
                            "Sqrt",
         
     | 
| 116 | 
         
            +
                            "Equal",
         
     | 
| 117 | 
         
             
                            "ConstantOfShape",
         
     | 
| 118 | 
         
            +
                            "Pow",
         
     | 
| 119 | 
         
            +
                            "Softmax",
         
     | 
| 120 | 
         
            +
                            "Squeeze"
         
     | 
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 
         | 
|
| 121 | 
         
             
                        ],
         
     | 
| 122 | 
         
             
                        "weight_type": "QInt8"
         
     | 
| 123 | 
         
             
                    }
         
     | 
    	
        tokenizer.json
    CHANGED
    
    | 
         The diff for this file is too large to render. 
		See raw diff 
     | 
| 
         |