Hugging Face's logo

tarekziade
/
vit-base-patch16-224-in21k-distilgpt2

Image-to-Text
Transformers
ONNX
Safetensors
vision-encoder-decoder
image-text-to-text
image-captioning
Inference Endpoints
Model card Files Community
tarekziade commited on
Commit
a12acc3
•
1 Parent(s): 4d0f6bc

Upload 16 files

Browse files
Files changed (12) hide show
  1. config.json +1 -1
  2. generation_config.json +1 -1
  3. onnx/decoder_model.onnx +2 -2
  4. onnx/decoder_model_merged.onnx +2 -2
  5. onnx/decoder_model_merged_quantized.onnx +2 -2
  6. onnx/decoder_model_quantized.onnx +2 -2
  7. onnx/decoder_with_past_model.onnx +2 -2
  8. onnx/decoder_with_past_model_quantized.onnx +2 -2
  9. onnx/encoder_model.onnx +2 -2
  10. onnx/encoder_model_quantized.onnx +2 -2
  11. quantize_config.json +71 -71
  12. tokenizer.json +0 -1
config.json CHANGED
@@ -179,6 +179,6 @@
179
  "model_type": "vision-encoder-decoder",
180
  "pad_token_id": 50256,
181
  "tie_word_embeddings": false,
182
- "transformers_version": "4.40.2",
183
  "use_cache": true
184
  }
 
179
  "model_type": "vision-encoder-decoder",
180
  "pad_token_id": 50256,
181
  "tie_word_embeddings": false,
182
+ "transformers_version": "4.33.2",
183
  "use_cache": true
184
  }
generation_config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
  "bos_token_id": 50256,
3
  "eos_token_id": 50256,
4
- "transformers_version": "4.40.2"
5
  }
 
1
  {
2
  "bos_token_id": 50256,
3
  "eos_token_id": 50256,
4
+ "transformers_version": "4.33.2"
5
  }
onnx/decoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a4d053ed1152fb8b4e3ac40e8edf80bc24645ed32e8fc1a7531cb2c92b0d1630
3
- size 385864191
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7351a7eaeeec455e70e68007a1dcf7307e245473f53da6b13312016ad5d992d3
3
+ size 385864797
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dc96e82c222400e7271a94e88c53e87e005a5b607c65595a84dad16700b5843f
3
- size 387341374
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f951eec84e5b043b03ac6a94556e7b0f16846282bc577adec061c848119aadac
3
+ size 387342586
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:534ffa60a4a26b59f017e1f770e5a8e9e23196ddf5bd764a2bf325cfbbd56b7b
3
- size 99765277
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ff8ee2b032955fd71b217c61751a92a6386e7b0e31228ef806a89a923bca34d
3
+ size 99759578
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:96d69b1df1679cdd10e6c7c820552abc141f4626371d9d8156a56ad9fcc640ce
3
- size 98068691
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:737759dec1b90e1e29eb61e271951db84e3e677d6f8332db64bd4a15fe931711
3
+ size 98065762
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c0365309a279acb426e47f38c2cd1df4a76fd0cc82f151a3ff3329e193c2eb6d
3
- size 385863771
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b99196af93603df1ef7eeb9676177d99c0a4a11b01e266ae5cb9327abe968c54
3
+ size 385864377
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:613a7a0d281df4932f9257896a9361d19d4ded28d39a7dc04ab9abbe6403431c
3
- size 98066098
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:12e5aad5d80e36cd4f9b2d45297732b01fd9e6503e6393aa393b4c13533c53e7
3
+ size 98063169
onnx/encoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:50b99dc721aed553c9d05984c787386745eb874707463d935b11ff5d298b49ed
3
- size 343440610
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aacd44f8cc26c1a0539f5db1fcb98cf3baad2c5e62583b913ed501daf1419e96
3
+ size 343440632
onnx/encoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0a6a0bbf9f22df59f2dfe55509a972365c99e274a295048d5cdd605c1022d398
3
- size 87038401
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:13062fd1801c38f1039ed439e792b9e08dc8eb0b006d4dd64146575e641c4286
3
+ size 87038173
quantize_config.json CHANGED
@@ -4,120 +4,120 @@
4
  "per_model_config": {
5
  "decoder_with_past_model": {
6
  "op_types": [
7
- "Reshape",
8
- "ConstantOfShape",
9
  "Sub",
10
- "Add",
11
- "Range",
12
- "Mul",
13
  "Softmax",
 
 
14
  "Gemm",
15
- "Sqrt",
16
- "Tanh",
17
- "Transpose",
18
  "Concat",
19
- "MatMul",
20
- "Constant",
21
- "Slice",
22
- "Squeeze",
23
- "Gather",
24
  "Where",
25
- "Split",
26
- "ReduceMean",
27
- "Cast",
28
  "Unsqueeze",
 
 
 
 
 
29
  "Shape",
30
- "Pow",
31
- "Div"
 
 
 
 
 
 
32
  ],
33
  "weight_type": "QInt8"
34
  },
35
  "decoder_model": {
36
  "op_types": [
37
- "Reshape",
38
- "ConstantOfShape",
39
  "Sub",
40
- "Add",
41
- "Range",
42
- "Mul",
43
  "Softmax",
 
 
44
  "Gemm",
45
- "Sqrt",
46
- "Tanh",
47
- "Transpose",
48
  "Concat",
49
- "MatMul",
50
- "Constant",
51
- "Slice",
52
- "Squeeze",
53
- "Gather",
54
  "Where",
55
- "Split",
56
- "ReduceMean",
57
- "Cast",
58
  "Unsqueeze",
 
 
 
 
 
59
  "Shape",
60
- "Pow",
61
- "Div"
 
 
 
 
 
 
62
  ],
63
  "weight_type": "QInt8"
64
  },
65
  "encoder_model": {
66
  "op_types": [
67
- "Reshape",
68
- "ConstantOfShape",
69
  "Sub",
70
- "Add",
71
- "Mul",
72
  "Softmax",
73
- "Sqrt",
74
- "Transpose",
75
- "Concat",
76
- "MatMul",
77
- "Constant",
78
- "Equal",
79
  "Slice",
80
- "Gather",
 
81
  "Where",
82
- "Erf",
83
- "ReduceMean",
84
- "Unsqueeze",
85
- "Shape",
86
  "Pow",
87
- "Expand",
 
88
  "Conv",
89
- "Div"
 
 
 
 
 
 
 
 
 
 
90
  ],
91
  "weight_type": "QUInt8"
92
  },
93
  "decoder_model_merged": {
94
  "op_types": [
95
- "Reshape",
96
- "ConstantOfShape",
97
  "Sub",
98
- "Add",
99
- "Range",
100
- "Mul",
101
  "Softmax",
 
 
102
  "Gemm",
103
- "Sqrt",
 
 
 
 
 
 
 
104
  "Tanh",
 
 
 
105
  "Transpose",
106
  "If",
107
- "Concat",
108
- "MatMul",
109
  "Constant",
110
- "Slice",
111
  "Squeeze",
112
- "Gather",
113
- "Where",
114
- "Split",
115
- "ReduceMean",
116
- "Cast",
117
- "Unsqueeze",
118
- "Shape",
119
- "Pow",
120
- "Div"
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
4
  "per_model_config": {
5
  "decoder_with_past_model": {
6
  "op_types": [
 
 
7
  "Sub",
8
+ "Cast",
 
 
9
  "Softmax",
10
+ "Add",
11
+ "Slice",
12
  "Gemm",
 
 
 
13
  "Concat",
14
+ "ConstantOfShape",
 
 
 
 
15
  "Where",
16
+ "Pow",
 
 
17
  "Unsqueeze",
18
+ "Reshape",
19
+ "MatMul",
20
+ "Div",
21
+ "Tanh",
22
+ "Gather",
23
  "Shape",
24
+ "Range",
25
+ "Transpose",
26
+ "ReduceMean",
27
+ "Mul",
28
+ "Constant",
29
+ "Sqrt",
30
+ "Squeeze",
31
+ "Split"
32
  ],
33
  "weight_type": "QInt8"
34
  },
35
  "decoder_model": {
36
  "op_types": [
 
 
37
  "Sub",
38
+ "Cast",
 
 
39
  "Softmax",
40
+ "Add",
41
+ "Slice",
42
  "Gemm",
 
 
 
43
  "Concat",
44
+ "ConstantOfShape",
 
 
 
 
45
  "Where",
46
+ "Pow",
 
 
47
  "Unsqueeze",
48
+ "Reshape",
49
+ "MatMul",
50
+ "Div",
51
+ "Tanh",
52
+ "Gather",
53
  "Shape",
54
+ "Range",
55
+ "Transpose",
56
+ "ReduceMean",
57
+ "Mul",
58
+ "Constant",
59
+ "Sqrt",
60
+ "Squeeze",
61
+ "Split"
62
  ],
63
  "weight_type": "QInt8"
64
  },
65
  "encoder_model": {
66
  "op_types": [
 
 
67
  "Sub",
 
 
68
  "Softmax",
69
+ "Expand",
70
+ "Add",
 
 
 
 
71
  "Slice",
72
+ "Concat",
73
+ "ConstantOfShape",
74
  "Where",
 
 
 
 
75
  "Pow",
76
+ "Unsqueeze",
77
+ "Equal",
78
  "Conv",
79
+ "Reshape",
80
+ "MatMul",
81
+ "Erf",
82
+ "Div",
83
+ "Gather",
84
+ "Shape",
85
+ "Transpose",
86
+ "ReduceMean",
87
+ "Mul",
88
+ "Constant",
89
+ "Sqrt"
90
  ],
91
  "weight_type": "QUInt8"
92
  },
93
  "decoder_model_merged": {
94
  "op_types": [
 
 
95
  "Sub",
96
+ "Cast",
 
 
97
  "Softmax",
98
+ "Add",
99
+ "Slice",
100
  "Gemm",
101
+ "Concat",
102
+ "ConstantOfShape",
103
+ "Where",
104
+ "Pow",
105
+ "Unsqueeze",
106
+ "Reshape",
107
+ "MatMul",
108
+ "Div",
109
  "Tanh",
110
+ "Gather",
111
+ "Shape",
112
+ "Range",
113
  "Transpose",
114
  "If",
115
+ "ReduceMean",
116
+ "Mul",
117
  "Constant",
118
+ "Sqrt",
119
  "Squeeze",
120
+ "Split"
 
 
 
 
 
 
 
 
121
  ],
122
  "weight_type": "QInt8"
123
  }
tokenizer.json CHANGED
@@ -40,7 +40,6 @@
40
  "end_of_word_suffix": "",
41
  "fuse_unk": false,
42
  "byte_fallback": false,
43
- "ignore_merges": false,
44
  "vocab": {
45
  "!": 0,
46
  "\"": 1,
 
40
  "end_of_word_suffix": "",
41
  "fuse_unk": false,
42
  "byte_fallback": false,
 
43
  "vocab": {
44
  "!": 0,
45
  "\"": 1,