Xenova HF staff commited on
Commit
4e6668f
1 Parent(s): 1563fb6

Upload folder using huggingface_hub

Browse files
config.json CHANGED
@@ -145,8 +145,7 @@
145
  50361,
146
  50362
147
  ],
148
- "torch_dtype": "float32",
149
- "transformers_version": "4.29.2",
150
  "use_cache": true,
151
  "use_weighted_layer_sum": false,
152
  "vocab_size": 51865
 
145
  50361,
146
  50362
147
  ],
148
+ "transformers_version": "4.30.2",
 
149
  "use_cache": true,
150
  "use_weighted_layer_sum": false,
151
  "vocab_size": 51865
generation_config.json CHANGED
@@ -217,5 +217,5 @@
217
  "transcribe": 50359,
218
  "translate": 50358
219
  },
220
- "transformers_version": "4.29.2"
221
  }
 
217
  "transcribe": 50359,
218
  "translate": 50358
219
  },
220
+ "transformers_version": "4.30.2"
221
  }
onnx/decoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:03d1c8736b387f84d94f0ceccdcdabddc0c9c953ff4b22a2f41304f341f96c03
3
- size 1507102
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1c0a21faf8ba940db5fcc29bdc0e66412d05fd80cf8dcfac8c92d23ba166d01
3
+ size 1506542
onnx/decoder_model.onnx_data CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3a57a63ddaabbaf0aab0f2d8ba3487d70962296033347b82c1bf60221b00fd3e
3
- size 7783260160
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d580f467fb53b33bbfdc1010bffc8b46a79e74ff5c4f19f7bf1516ec7767327d
3
+ size 3891630080
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5b1a1b7a899882f3bfdd5f6a0de29fcc1bb220e61f91293058a7c3f398d42aff
3
- size 2762394
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6386da420526f0323712dfe66e02004259ef44377e927624ab1e5d5953fa73c
3
+ size 2761864
onnx/decoder_model_merged.onnx_data CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3a57a63ddaabbaf0aab0f2d8ba3487d70962296033347b82c1bf60221b00fd3e
3
- size 7783260160
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d580f467fb53b33bbfdc1010bffc8b46a79e74ff5c4f19f7bf1516ec7767327d
3
+ size 3891630080
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5af419cb35af4dd563b859d54761c1865edcf8a7b0c5b7cce3b666c734b8c053
3
- size 983867549
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2ca3d44b85d0dd8442ce7e4a168265d0678fa01d8f239d3aeee0ade5c51e88d1
3
+ size 980944149
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b963e09c97852775c27b568ec224fbb41696d4f6271cbbf1c6c371e8ddc67730
3
- size 981928446
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:318d59d78c33a872b843347a7ca309003d6cde6cd972e250407cb502ecbba730
3
+ size 979005046
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d0ff014c984ac4a2689ceaf336eab6212c244cc11fe0783dc813f9fa367411bc
3
- size 1294069
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:564c4dd801433b3ccf607b947ec8dac4c6cb6538edb72b6cc0cf0217141ba289
3
+ size 1293576
onnx/decoder_with_past_model.onnx_data CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f4ec656bc837a275fac9d4f26cc3255b0eb19760f61e4442999a79bce265d8bb
3
- size 6944071680
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8293518c6d66788e5020c694e1a317b744db62a0c918f6ff574bdbed8e23f99e
3
+ size 3472035840
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2f768b48810f303b1750eea573d941b4e8e2816e76288aaaf1e6129594784e44
3
- size 876151256
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:912d66d239bac2b85e77b116aa92d68a1e45a0b9b164b3a2c044c33d35427a56
3
+ size 873637776
onnx/encoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:550015511f42e017e3f38d809ec5c0b6ab6013a203c8cab7fe5535a4e885e921
3
- size 602752
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c34e862d089a0c6a33d9d63620882927f2bf41872a257807e81cdc4ddae5a85c
3
+ size 602071
onnx/encoder_model.onnx_data CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ab1b99d6696bc44ac1a2bb072ed1e73233d522fac325498b374fe4778394b191
3
- size 5094277120
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c869922c76d760a49fbffeff93e93e7a99f016a28af0a854c1454785666070ad
3
+ size 2547138560
onnx/encoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7339d32967d4ce44d856bf98f216eed4a6f3847e70a7228f52094a59cd8eaebb
3
- size 646822577
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5dcb8d36c86031fb95287fdef2e1aa43f46e13f79028577923c5c76652c2597b
3
+ size 644978442
quant_config.json ADDED
@@ -0,0 +1,89 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "per_channel": false,
3
+ "reduce_range": false,
4
+ "per_model_config": {
5
+ "encoder_model": {
6
+ "op_types": [
7
+ "Div",
8
+ "Erf",
9
+ "Gather",
10
+ "MatMul",
11
+ "Conv",
12
+ "ReduceMean",
13
+ "Pow",
14
+ "Softmax",
15
+ "Sub",
16
+ "Constant",
17
+ "Add",
18
+ "Transpose",
19
+ "Mul",
20
+ "Unsqueeze",
21
+ "Shape",
22
+ "Sqrt",
23
+ "Reshape",
24
+ "Concat"
25
+ ],
26
+ "weight_type": "QUInt8"
27
+ },
28
+ "decoder_model": {
29
+ "op_types": [
30
+ "Cast",
31
+ "Squeeze",
32
+ "Slice",
33
+ "Less",
34
+ "Sub",
35
+ "Div",
36
+ "Expand",
37
+ "Gather",
38
+ "Range",
39
+ "ConstantOfShape",
40
+ "Mul",
41
+ "Shape",
42
+ "Sqrt",
43
+ "Reshape",
44
+ "ReduceMean",
45
+ "Pow",
46
+ "Constant",
47
+ "Add",
48
+ "Unsqueeze",
49
+ "Concat",
50
+ "Equal",
51
+ "MatMul",
52
+ "Erf",
53
+ "Softmax",
54
+ "Where",
55
+ "Transpose"
56
+ ],
57
+ "weight_type": "QInt8"
58
+ },
59
+ "decoder_model_merged": {
60
+ "op_types": [
61
+ "If"
62
+ ],
63
+ "weight_type": "QInt8"
64
+ },
65
+ "decoder_with_past_model": {
66
+ "op_types": [
67
+ "Div",
68
+ "Gather",
69
+ "MatMul",
70
+ "Erf",
71
+ "ReduceMean",
72
+ "Pow",
73
+ "Softmax",
74
+ "Slice",
75
+ "Sub",
76
+ "Constant",
77
+ "Add",
78
+ "Transpose",
79
+ "Mul",
80
+ "Unsqueeze",
81
+ "Shape",
82
+ "Sqrt",
83
+ "Reshape",
84
+ "Concat"
85
+ ],
86
+ "weight_type": "QInt8"
87
+ }
88
+ }
89
+ }