Hugging Face's logo

tarekziade
/
vit-base-patch16-224-distilgpt2

Image-to-Text
Transformers
PyTorch
ONNX
Safetensors
vision-encoder-decoder
image-captioning
Eval Results
Inference Endpoints
Model card Files Community
tarekziade commited on
Commit
d85acbe
•
1 Parent(s): 6276b49

Upload 19 files

Browse files
Files changed (13) hide show
  1. config.json +1 -1
  2. generation_config.json +5 -6
  3. model.safetensors +2 -2
  4. onnx/decoder_model.onnx +1 -1
  5. onnx/decoder_model_merged.onnx +1 -1
  6. onnx/decoder_model_merged_quantized.onnx +1 -1
  7. onnx/decoder_model_quantized.onnx +1 -1
  8. onnx/decoder_with_past_model.onnx +1 -1
  9. onnx/decoder_with_past_model_quantized.onnx +1 -1
  10. onnx/encoder_model.onnx +1 -1
  11. onnx/encoder_model_quantized.onnx +1 -1
  12. quantize_config.json +72 -72
  13. training_args.bin +1 -1
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "tarekziade/vit-base-patch16-224-distilgpt2",
3
  "architectures": [
4
  "VisionEncoderDecoderModel"
5
  ],
 
1
  {
2
+ "_name_or_path": "/Volumes/SSD/vit-base-patch16-224-distilgpt2",
3
  "architectures": [
4
  "VisionEncoderDecoderModel"
5
  ],
generation_config.json CHANGED
@@ -1,13 +1,12 @@
1
  {
2
  "bos_token_id": 50256,
3
- "do_sample": false,
4
  "eos_token_id": 50256,
5
- "max_length": 40,
6
  "no_repeat_ngram_size": 2,
 
7
  "pad_token_id": 50256,
8
  "repetition_penalty": 1.2,
9
- "transformers_version": "4.33.2",
10
  "seed": 12,
11
- "temperature": 0.7,
12
- "use_cache": true
13
- }
 
1
  {
2
  "bos_token_id": 50256,
3
+ "early_stopping": true,
4
  "eos_token_id": 50256,
5
+ "max_length": 50,
6
  "no_repeat_ngram_size": 2,
7
+ "num_beams": 2,
8
  "pad_token_id": 50256,
9
  "repetition_penalty": 1.2,
 
10
  "seed": 12,
11
+ "transformers_version": "4.33.2"
12
+ }
 
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f025f5c7e479b92e67934fdad516b6b8d6295af04a95f84cba26ce54b27426bd
3
- size 732347930
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:74574a184acd40eb36ffb93d03ea3780bbe8f8f5d019aabbdf7554304e704a05
3
+ size 729979160
onnx/decoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e311a0c129981b659cd30cd641d8d759306a4b92a5459616266386431d7f0544
3
  size 385864797
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e3704718aee4c999920574a3fb45b959ee59a9eb687bc5ff39c1ec6c3b3d32fc
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:d603904cb811150731ec5d2d4680e7b3a79bd728b6161a9688563ca92fa403c5
3
  size 387342586
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dfc0cac17407a66735b4b4dd5ce62900967a015a618f24c6036b628e1c600328
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 sha256:0954635de52a206823c20847ff4793622a7b9c0ef1e5adaaedb906dba02ce466
3
  size 99759579
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:083cc1aa2c10ee51436601f996091d4633ff2ad6ab933f1197e23c14893fc89e
3
  size 99759579
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:02ecd9eedc6985c1849cc94aa1039bd9e82220a1ad2e57e9219086a36b5baa24
3
  size 98065763
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:230dd40ee4ba5bd9ec09ad783b45d98ca8acd7b37ff1f73ddbfa6a63c63163ba
3
  size 98065763
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6a607d1e4630646433316bbd5d27de80baa0292d01c9f962f97f783de64f4996
3
  size 385864377
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aae44e11c82e59d84ac7c9f654942dfbcbb2c0d2a03bc8ec49876e81dc3393a1
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 sha256:5c5a4a2f6ecdbee840feb854a400b1b1856a896a29d15993458dc7fd1a128719
3
  size 98063170
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1823eff0a8dc274eabb3df67551bc7330d4e22a22c23676ff978eb33c643896
3
  size 98063170
onnx/encoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4fe7a39c03fc5775d35fc35228d1e6d232165f11a5c00e178fc000d989c1cfa4
3
  size 343440632
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca2910ada0c94d6f7d72c80e7e4f40916c0bfbd1cff763ad86d96e02fd527e90
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:bc44a3d2a121920cca50922f6e05e42283b80c98b806a870c625fa94432711e9
3
  size 87038173
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b9930375f2182d3cb829d78750939633587891b18edb0a66eeb569317ccc0776
3
  size 87038173
quantize_config.json CHANGED
@@ -2,122 +2,122 @@
2
  "per_channel": false,
3
  "reduce_range": false,
4
  "per_model_config": {
5
- "decoder_with_past_model": {
6
  "op_types": [
7
- "Sqrt",
8
- "Transpose",
9
- "Reshape",
10
- "Gather",
11
- "Constant",
12
  "Pow",
 
13
  "Sub",
14
- "Cast",
15
  "Mul",
16
- "Range",
17
- "Where",
18
- "Slice",
19
  "Div",
20
- "Split",
21
- "MatMul",
22
  "Unsqueeze",
23
- "Concat",
 
24
  "Softmax",
25
- "Squeeze",
26
- "Add",
27
  "Shape",
28
- "ConstantOfShape",
29
- "Gemm",
30
- "Tanh",
31
- "ReduceMean"
 
32
  ],
33
- "weight_type": "QInt8"
34
  },
35
  "decoder_model": {
36
  "op_types": [
37
- "Sqrt",
38
- "Transpose",
39
- "Reshape",
40
- "Gather",
41
- "Constant",
42
  "Pow",
43
  "Sub",
44
- "Cast",
45
  "Mul",
46
- "Range",
47
- "Where",
48
- "Slice",
49
  "Div",
50
- "Split",
51
- "MatMul",
52
  "Unsqueeze",
53
- "Concat",
54
  "Softmax",
 
 
55
  "Squeeze",
56
- "Add",
 
 
 
57
  "Shape",
58
- "ConstantOfShape",
59
- "Gemm",
60
- "Tanh",
61
- "ReduceMean"
62
  ],
63
  "weight_type": "QInt8"
64
  },
65
- "encoder_model": {
66
  "op_types": [
67
- "Sqrt",
68
- "Transpose",
69
- "Reshape",
70
- "Gather",
71
- "Constant",
72
  "Pow",
73
  "Sub",
74
  "Mul",
75
- "Expand",
76
- "Where",
77
- "Slice",
78
  "Div",
79
- "MatMul",
 
80
  "Unsqueeze",
81
- "Concat",
82
  "Softmax",
83
- "Add",
 
 
 
 
 
 
84
  "Shape",
85
- "Erf",
86
- "ConstantOfShape",
87
  "ReduceMean",
88
- "Equal",
89
- "Conv"
 
90
  ],
91
- "weight_type": "QUInt8"
92
  },
93
  "decoder_model_merged": {
94
  "op_types": [
95
- "Sqrt",
96
- "Transpose",
97
- "Reshape",
98
- "Gather",
99
- "Constant",
100
  "Pow",
101
  "Sub",
102
- "Cast",
103
  "Mul",
104
- "If",
105
- "Range",
106
- "Where",
107
- "Slice",
108
  "Div",
109
- "Split",
110
- "MatMul",
111
  "Unsqueeze",
112
- "Concat",
113
  "Softmax",
 
114
  "Squeeze",
115
- "Add",
 
 
 
 
116
  "Shape",
117
- "ConstantOfShape",
118
- "Gemm",
119
- "Tanh",
120
- "ReduceMean"
121
  ],
122
  "weight_type": "QInt8"
123
  }
 
2
  "per_channel": false,
3
  "reduce_range": false,
4
  "per_model_config": {
5
+ "encoder_model": {
6
  "op_types": [
7
+ "MatMul",
8
+ "Add",
9
+ "Equal",
10
+ "ConstantOfShape",
 
11
  "Pow",
12
+ "Erf",
13
  "Sub",
 
14
  "Mul",
15
+ "Reshape",
 
 
16
  "Div",
17
+ "Where",
 
18
  "Unsqueeze",
19
+ "Expand",
20
+ "Gather",
21
  "Softmax",
22
+ "Transpose",
23
+ "Constant",
24
  "Shape",
25
+ "ReduceMean",
26
+ "Concat",
27
+ "Sqrt",
28
+ "Slice",
29
+ "Conv"
30
  ],
31
+ "weight_type": "QUInt8"
32
  },
33
  "decoder_model": {
34
  "op_types": [
35
+ "MatMul",
36
+ "Add",
37
+ "ConstantOfShape",
 
 
38
  "Pow",
39
  "Sub",
 
40
  "Mul",
41
+ "Reshape",
 
 
42
  "Div",
43
+ "Where",
44
+ "Tanh",
45
  "Unsqueeze",
46
+ "Gather",
47
  "Softmax",
48
+ "Split",
49
+ "Slice",
50
  "Squeeze",
51
+ "Range",
52
+ "Transpose",
53
+ "Cast",
54
+ "Constant",
55
  "Shape",
56
+ "ReduceMean",
57
+ "Concat",
58
+ "Sqrt",
59
+ "Gemm"
60
  ],
61
  "weight_type": "QInt8"
62
  },
63
+ "decoder_with_past_model": {
64
  "op_types": [
65
+ "MatMul",
66
+ "Add",
67
+ "ConstantOfShape",
 
 
68
  "Pow",
69
  "Sub",
70
  "Mul",
71
+ "Reshape",
 
 
72
  "Div",
73
+ "Where",
74
+ "Tanh",
75
  "Unsqueeze",
76
+ "Gather",
77
  "Softmax",
78
+ "Split",
79
+ "Slice",
80
+ "Squeeze",
81
+ "Range",
82
+ "Transpose",
83
+ "Cast",
84
+ "Constant",
85
  "Shape",
 
 
86
  "ReduceMean",
87
+ "Concat",
88
+ "Sqrt",
89
+ "Gemm"
90
  ],
91
+ "weight_type": "QInt8"
92
  },
93
  "decoder_model_merged": {
94
  "op_types": [
95
+ "MatMul",
96
+ "Add",
97
+ "ConstantOfShape",
 
 
98
  "Pow",
99
  "Sub",
 
100
  "Mul",
101
+ "Gemm",
102
+ "Reshape",
 
 
103
  "Div",
104
+ "Where",
105
+ "Tanh",
106
  "Unsqueeze",
107
+ "Gather",
108
  "Softmax",
109
+ "Split",
110
  "Squeeze",
111
+ "Range",
112
+ "Transpose",
113
+ "Cast",
114
+ "Constant",
115
+ "If",
116
  "Shape",
117
+ "ReduceMean",
118
+ "Concat",
119
+ "Sqrt",
120
+ "Slice"
121
  ],
122
  "weight_type": "QInt8"
123
  }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:86336c5fe5b6aa0da13c2f330e26acdb193817ccf3625ddc613a8eb4bbcfe3e7
3
  size 5176
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac92771c5176fd55689bb44c37e3467ce9f70d1a9d642ce3f6e6e4e449de33cd
3
  size 5176