Hugging Face's logo

yujiepan
/
internal.swin-base-food101-int8-structured38.63

Image Classification
Transformers
PyTorch
OpenVINO
swin
vision
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
yujiepan commited on
Commit
32f0516
•
1 Parent(s): 5e4cd78

upload model

Browse files
Files changed (11) hide show
  1. README.md +93 -0
  2. all_results.json +12 -0
  3. config.json +255 -0
  4. openvino_config.json +86 -0
  5. openvino_model.bin +3 -0
  6. openvino_model.xml +3 -0
  7. preprocessor_config.json +23 -0
  8. pytorch_model.bin +3 -0
  9. structured_sparsity.csv +145 -0
  10. trainer_state.json +0 -0
  11. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,93 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - image-classification
5
+ - vision
6
+ - generated_from_trainer
7
+ datasets:
8
+ - food101
9
+ metrics:
10
+ - accuracy
11
+ model-index:
12
+ - name: swin-food101-jpqd-1to2r1.5-epo7-finetuned-student
13
+ results:
14
+ - task:
15
+ name: Image Classification
16
+ type: image-classification
17
+ dataset:
18
+ name: food101
19
+ type: food101
20
+ config: default
21
+ split: validation
22
+ args: default
23
+ metrics:
24
+ - name: Accuracy
25
+ type: accuracy
26
+ value: 0.9123960396039604
27
+ ---
28
+
29
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
30
+ should probably proofread and complete it, then remove this comment. -->
31
+
32
+ # swin-food101-jpqd-1to2r1.5-epo7-finetuned-student
33
+
34
+ This model is a fine-tuned version of [skylord/swin-finetuned-food101](https://huggingface.co/skylord/swin-finetuned-food101) on the food101 dataset.
35
+ It achieves the following results on the evaluation set:
36
+ - Loss: 0.2658
37
+ - Accuracy: 0.9124
38
+
39
+ ## Model description
40
+
41
+ More information needed
42
+
43
+ ## Intended uses & limitations
44
+
45
+ More information needed
46
+
47
+ ## Training and evaluation data
48
+
49
+ More information needed
50
+
51
+ ## Training procedure
52
+
53
+ ### Training hyperparameters
54
+
55
+ The following hyperparameters were used during training:
56
+ - learning_rate: 5e-05
57
+ - train_batch_size: 16
58
+ - eval_batch_size: 128
59
+ - seed: 42
60
+ - gradient_accumulation_steps: 4
61
+ - total_train_batch_size: 64
62
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
63
+ - lr_scheduler_type: linear
64
+ - num_epochs: 7.0
65
+
66
+ ### Training results
67
+
68
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
69
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|
70
+ | 0.2977 | 0.42 | 500 | 0.1949 | 0.9112 |
71
+ | 0.3183 | 0.84 | 1000 | 0.1867 | 0.9144 |
72
+ | 99.9552 | 1.27 | 1500 | 99.4882 | 0.7577 |
73
+ | 162.4195 | 1.69 | 2000 | 162.7763 | 0.3373 |
74
+ | 1.2272 | 2.11 | 2500 | 0.7333 | 0.8564 |
75
+ | 1.0236 | 2.54 | 3000 | 0.5016 | 0.8823 |
76
+ | 0.6472 | 2.96 | 3500 | 0.4337 | 0.8908 |
77
+ | 0.52 | 3.38 | 4000 | 0.3927 | 0.8974 |
78
+ | 0.6075 | 3.8 | 4500 | 0.3506 | 0.9011 |
79
+ | 0.5348 | 4.23 | 5000 | 0.3425 | 0.9006 |
80
+ | 0.444 | 4.65 | 5500 | 0.3268 | 0.9044 |
81
+ | 0.5787 | 5.07 | 6000 | 0.3020 | 0.9078 |
82
+ | 0.3995 | 5.49 | 6500 | 0.2932 | 0.9095 |
83
+ | 0.414 | 5.92 | 7000 | 0.2806 | 0.9104 |
84
+ | 0.4386 | 6.34 | 7500 | 0.2738 | 0.9112 |
85
+ | 0.452 | 6.76 | 8000 | 0.2673 | 0.9127 |
86
+
87
+
88
+ ### Framework versions
89
+
90
+ - Transformers 4.26.0
91
+ - Pytorch 1.13.1+cu116
92
+ - Datasets 2.8.0
93
+ - Tokenizers 0.13.2
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 7.0,
3
+ "eval_accuracy": 0.9123960396039604,
4
+ "eval_loss": 0.2658494710922241,
5
+ "eval_runtime": 228.1316,
6
+ "eval_samples_per_second": 110.682,
7
+ "eval_steps_per_second": 0.868,
8
+ "train_loss": 18.350530295344083,
9
+ "train_runtime": 37482.3852,
10
+ "train_samples_per_second": 14.147,
11
+ "train_steps_per_second": 0.221
12
+ }
config.json ADDED
@@ -0,0 +1,255 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "skylord/swin-finetuned-food101",
3
+ "architectures": [
4
+ "NNCFNetwork"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.0,
7
+ "depths": [
8
+ 2,
9
+ 2,
10
+ 18,
11
+ 2
12
+ ],
13
+ "drop_path_rate": 0.1,
14
+ "embed_dim": 128,
15
+ "encoder_stride": 32,
16
+ "finetuning_task": "image-classification",
17
+ "hidden_act": "gelu",
18
+ "hidden_dropout_prob": 0.0,
19
+ "hidden_size": 1024,
20
+ "id2label": {
21
+ "0": "apple_pie",
22
+ "1": "baby_back_ribs",
23
+ "10": "bruschetta",
24
+ "100": "waffles",
25
+ "11": "caesar_salad",
26
+ "12": "cannoli",
27
+ "13": "caprese_salad",
28
+ "14": "carrot_cake",
29
+ "15": "ceviche",
30
+ "16": "cheesecake",
31
+ "17": "cheese_plate",
32
+ "18": "chicken_curry",
33
+ "19": "chicken_quesadilla",
34
+ "2": "baklava",
35
+ "20": "chicken_wings",
36
+ "21": "chocolate_cake",
37
+ "22": "chocolate_mousse",
38
+ "23": "churros",
39
+ "24": "clam_chowder",
40
+ "25": "club_sandwich",
41
+ "26": "crab_cakes",
42
+ "27": "creme_brulee",
43
+ "28": "croque_madame",
44
+ "29": "cup_cakes",
45
+ "3": "beef_carpaccio",
46
+ "30": "deviled_eggs",
47
+ "31": "donuts",
48
+ "32": "dumplings",
49
+ "33": "edamame",
50
+ "34": "eggs_benedict",
51
+ "35": "escargots",
52
+ "36": "falafel",
53
+ "37": "filet_mignon",
54
+ "38": "fish_and_chips",
55
+ "39": "foie_gras",
56
+ "4": "beef_tartare",
57
+ "40": "french_fries",
58
+ "41": "french_onion_soup",
59
+ "42": "french_toast",
60
+ "43": "fried_calamari",
61
+ "44": "fried_rice",
62
+ "45": "frozen_yogurt",
63
+ "46": "garlic_bread",
64
+ "47": "gnocchi",
65
+ "48": "greek_salad",
66
+ "49": "grilled_cheese_sandwich",
67
+ "5": "beet_salad",
68
+ "50": "grilled_salmon",
69
+ "51": "guacamole",
70
+ "52": "gyoza",
71
+ "53": "hamburger",
72
+ "54": "hot_and_sour_soup",
73
+ "55": "hot_dog",
74
+ "56": "huevos_rancheros",
75
+ "57": "hummus",
76
+ "58": "ice_cream",
77
+ "59": "lasagna",
78
+ "6": "beignets",
79
+ "60": "lobster_bisque",
80
+ "61": "lobster_roll_sandwich",
81
+ "62": "macaroni_and_cheese",
82
+ "63": "macarons",
83
+ "64": "miso_soup",
84
+ "65": "mussels",
85
+ "66": "nachos",
86
+ "67": "omelette",
87
+ "68": "onion_rings",
88
+ "69": "oysters",
89
+ "7": "bibimbap",
90
+ "70": "pad_thai",
91
+ "71": "paella",
92
+ "72": "pancakes",
93
+ "73": "panna_cotta",
94
+ "74": "peking_duck",
95
+ "75": "pho",
96
+ "76": "pizza",
97
+ "77": "pork_chop",
98
+ "78": "poutine",
99
+ "79": "prime_rib",
100
+ "8": "bread_pudding",
101
+ "80": "pulled_pork_sandwich",
102
+ "81": "ramen",
103
+ "82": "ravioli",
104
+ "83": "red_velvet_cake",
105
+ "84": "risotto",
106
+ "85": "samosa",
107
+ "86": "sashimi",
108
+ "87": "scallops",
109
+ "88": "seaweed_salad",
110
+ "89": "shrimp_and_grits",
111
+ "9": "breakfast_burrito",
112
+ "90": "spaghetti_bolognese",
113
+ "91": "spaghetti_carbonara",
114
+ "92": "spring_rolls",
115
+ "93": "steak",
116
+ "94": "strawberry_shortcake",
117
+ "95": "sushi",
118
+ "96": "tacos",
119
+ "97": "takoyaki",
120
+ "98": "tiramisu",
121
+ "99": "tuna_tartare"
122
+ },
123
+ "image_size": 224,
124
+ "initializer_range": 0.02,
125
+ "label2id": {
126
+ "apple_pie": "0",
127
+ "baby_back_ribs": "1",
128
+ "baklava": "2",
129
+ "beef_carpaccio": "3",
130
+ "beef_tartare": "4",
131
+ "beet_salad": "5",
132
+ "beignets": "6",
133
+ "bibimbap": "7",
134
+ "bread_pudding": "8",
135
+ "breakfast_burrito": "9",
136
+ "bruschetta": "10",
137
+ "caesar_salad": "11",
138
+ "cannoli": "12",
139
+ "caprese_salad": "13",
140
+ "carrot_cake": "14",
141
+ "ceviche": "15",
142
+ "cheese_plate": "17",
143
+ "cheesecake": "16",
144
+ "chicken_curry": "18",
145
+ "chicken_quesadilla": "19",
146
+ "chicken_wings": "20",
147
+ "chocolate_cake": "21",
148
+ "chocolate_mousse": "22",
149
+ "churros": "23",
150
+ "clam_chowder": "24",
151
+ "club_sandwich": "25",
152
+ "crab_cakes": "26",
153
+ "creme_brulee": "27",
154
+ "croque_madame": "28",
155
+ "cup_cakes": "29",
156
+ "deviled_eggs": "30",
157
+ "donuts": "31",
158
+ "dumplings": "32",
159
+ "edamame": "33",
160
+ "eggs_benedict": "34",
161
+ "escargots": "35",
162
+ "falafel": "36",
163
+ "filet_mignon": "37",
164
+ "fish_and_chips": "38",
165
+ "foie_gras": "39",
166
+ "french_fries": "40",
167
+ "french_onion_soup": "41",
168
+ "french_toast": "42",
169
+ "fried_calamari": "43",
170
+ "fried_rice": "44",
171
+ "frozen_yogurt": "45",
172
+ "garlic_bread": "46",
173
+ "gnocchi": "47",
174
+ "greek_salad": "48",
175
+ "grilled_cheese_sandwich": "49",
176
+ "grilled_salmon": "50",
177
+ "guacamole": "51",
178
+ "gyoza": "52",
179
+ "hamburger": "53",
180
+ "hot_and_sour_soup": "54",
181
+ "hot_dog": "55",
182
+ "huevos_rancheros": "56",
183
+ "hummus": "57",
184
+ "ice_cream": "58",
185
+ "lasagna": "59",
186
+ "lobster_bisque": "60",
187
+ "lobster_roll_sandwich": "61",
188
+ "macaroni_and_cheese": "62",
189
+ "macarons": "63",
190
+ "miso_soup": "64",
191
+ "mussels": "65",
192
+ "nachos": "66",
193
+ "omelette": "67",
194
+ "onion_rings": "68",
195
+ "oysters": "69",
196
+ "pad_thai": "70",
197
+ "paella": "71",
198
+ "pancakes": "72",
199
+ "panna_cotta": "73",
200
+ "peking_duck": "74",
201
+ "pho": "75",
202
+ "pizza": "76",
203
+ "pork_chop": "77",
204
+ "poutine": "78",
205
+ "prime_rib": "79",
206
+ "pulled_pork_sandwich": "80",
207
+ "ramen": "81",
208
+ "ravioli": "82",
209
+ "red_velvet_cake": "83",
210
+ "risotto": "84",
211
+ "samosa": "85",
212
+ "sashimi": "86",
213
+ "scallops": "87",
214
+ "seaweed_salad": "88",
215
+ "shrimp_and_grits": "89",
216
+ "spaghetti_bolognese": "90",
217
+ "spaghetti_carbonara": "91",
218
+ "spring_rolls": "92",
219
+ "steak": "93",
220
+ "strawberry_shortcake": "94",
221
+ "sushi": "95",
222
+ "tacos": "96",
223
+ "takoyaki": "97",
224
+ "tiramisu": "98",
225
+ "tuna_tartare": "99",
226
+ "waffles": "100"
227
+ },
228
+ "layer_norm_eps": 1e-05,
229
+ "mlp_ratio": 4.0,
230
+ "model_type": "swin",
231
+ "num_channels": 3,
232
+ "num_heads": [
233
+ 4,
234
+ 8,
235
+ 16,
236
+ 32
237
+ ],
238
+ "num_layers": 4,
239
+ "out_features": null,
240
+ "patch_size": 4,
241
+ "path_norm": true,
242
+ "problem_type": "single_label_classification",
243
+ "qkv_bias": true,
244
+ "stage_names": [
245
+ "stem",
246
+ "stage1",
247
+ "stage2",
248
+ "stage3",
249
+ "stage4"
250
+ ],
251
+ "torch_dtype": "float32",
252
+ "transformers_version": "4.26.0",
253
+ "use_absolute_embeddings": false,
254
+ "window_size": 7
255
+ }
openvino_config.json ADDED
@@ -0,0 +1,86 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "compression": [
3
+ {
4
+ "algorithm": "movement_sparsity",
5
+ "ignored_scopes": [
6
+ "{re}.*PatchEmbed.*",
7
+ "{re}.*PatchMerging.*",
8
+ "{re}.*classifier.*",
9
+ "{re}.*LayerNorm.*"
10
+ ],
11
+ "params": {
12
+ "enable_structured_masking": true,
13
+ "importance_regularization_factor": 1.5,
14
+ "warmup_end_epoch": 2,
15
+ "warmup_start_epoch": 1
16
+ },
17
+ "sparse_structure_by_scopes": [
18
+ {
19
+ "mode": "block",
20
+ "sparse_factors": [
21
+ 16,
22
+ 16
23
+ ],
24
+ "target_scopes": "{re}.*SwinAttention.*"
25
+ },
26
+ {
27
+ "axis": 0,
28
+ "mode": "per_dim",
29
+ "target_scopes": "{re}.*SwinIntermediate.*"
30
+ },
31
+ {
32
+ "axis": 1,
33
+ "mode": "per_dim",
34
+ "target_scopes": "{re}.*SwinOutput.*"
35
+ }
36
+ ]
37
+ },
38
+ {
39
+ "algorithm": "quantization",
40
+ "export_to_onnx_standard_ops": false,
41
+ "ignored_scopes": [
42
+ "{re}.*__add___[0-1]",
43
+ "{re}.*layer_norm_0",
44
+ "{re}.*matmul_1",
45
+ "{re}.*__truediv__*"
46
+ ],
47
+ "initializer": {
48
+ "batchnorm_adaptation": {
49
+ "num_bn_adaptation_samples": 200
50
+ },
51
+ "range": {
52
+ "num_init_samples": 32,
53
+ "params": {
54
+ "max_percentile": 99.99,
55
+ "min_percentile": 0.01
56
+ },
57
+ "type": "percentile"
58
+ }
59
+ },
60
+ "overflow_fix": "enable",
61
+ "preset": "mixed",
62
+ "scope_overrides": {
63
+ "activations": {
64
+ "{re}.*matmul_0": {
65
+ "mode": "symmetric"
66
+ }
67
+ }
68
+ }
69
+ }
70
+ ],
71
+ "input_info": [
72
+ {
73
+ "keyword": "pixel_values",
74
+ "sample_size": [
75
+ 16,
76
+ 3,
77
+ 224,
78
+ 224
79
+ ],
80
+ "type": "float"
81
+ }
82
+ ],
83
+ "optimum_version": "1.6.3",
84
+ "save_onnx_model": false,
85
+ "transformers_version": "4.26.0"
86
+ }
openvino_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df7508f98eaf8c3e85d9d18fd584f83785927d4b29bbf4b0c49730af7731ad6e
3
+ size 57467668
openvino_model.xml ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:780408c571b7f04617d51a76d20fccbf84fe7442a219d736898a4bdb4cc85b0b
3
+ size 10499070
preprocessor_config.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "do_rescale": true,
4
+ "do_resize": true,
5
+ "feature_extractor_type": "ViTFeatureExtractor",
6
+ "image_mean": [
7
+ 0.485,
8
+ 0.456,
9
+ 0.406
10
+ ],
11
+ "image_processor_type": "ViTFeatureExtractor",
12
+ "image_std": [
13
+ 0.229,
14
+ 0.224,
15
+ 0.225
16
+ ],
17
+ "resample": 3,
18
+ "rescale_factor": 0.00392156862745098,
19
+ "size": {
20
+ "height": 224,
21
+ "width": 224
22
+ }
23
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a595da5f8dc53fd4eb8d06539dc0e72af75f0f8bb038ac0931b4d764d3cca25
3
+ size 685689463
structured_sparsity.csv ADDED
@@ -0,0 +1,145 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ,group_id,type,torch_module,weight_shape,pruned_weight_shape,bias_shape,pruned_bias_shape,head_or_channel_id_to_keep,module_node_name
2
+ 0,0,MHSA,nncf_module.swin.encoder.layers.0.blocks.0.attention.self.query,"(128, 128)","(32, 128)","(128,)","(32,)",[1],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[0]/ModuleList[blocks]/SwinLayer[0]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
3
+ 1,0,MHSA,nncf_module.swin.encoder.layers.0.blocks.0.attention.self.key,"(128, 128)","(32, 128)","(128,)","(32,)",[1],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[0]/ModuleList[blocks]/SwinLayer[0]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
4
+ 2,0,MHSA,nncf_module.swin.encoder.layers.0.blocks.0.attention.self.value,"(128, 128)","(32, 128)","(128,)","(32,)",[1],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[0]/ModuleList[blocks]/SwinLayer[0]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
5
+ 3,0,MHSA,nncf_module.swin.encoder.layers.0.blocks.0.attention.output.dense,"(128, 128)","(128, 32)","(128,)","(128,)",[1],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[0]/ModuleList[blocks]/SwinLayer[0]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
6
+ 4,1,FF,nncf_module.swin.encoder.layers.0.blocks.0.intermediate.dense,"(512, 128)","(309, 128)","(512,)","(309,)",[309 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[0]/ModuleList[blocks]/SwinLayer[0]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
7
+ 5,1,FF,nncf_module.swin.encoder.layers.0.blocks.0.output.dense,"(128, 512)","(128, 309)","(128,)","(128,)",[309 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[0]/ModuleList[blocks]/SwinLayer[0]/SwinOutput[output]/NNCFLinear[dense]/linear_0
8
+ 6,2,MHSA,nncf_module.swin.encoder.layers.0.blocks.1.attention.self.query,"(128, 128)","(64, 128)","(128,)","(64,)","[2, 3]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[0]/ModuleList[blocks]/SwinLayer[1]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
9
+ 7,2,MHSA,nncf_module.swin.encoder.layers.0.blocks.1.attention.self.key,"(128, 128)","(64, 128)","(128,)","(64,)","[2, 3]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[0]/ModuleList[blocks]/SwinLayer[1]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
10
+ 8,2,MHSA,nncf_module.swin.encoder.layers.0.blocks.1.attention.self.value,"(128, 128)","(64, 128)","(128,)","(64,)","[2, 3]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[0]/ModuleList[blocks]/SwinLayer[1]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
11
+ 9,2,MHSA,nncf_module.swin.encoder.layers.0.blocks.1.attention.output.dense,"(128, 128)","(128, 64)","(128,)","(128,)","[2, 3]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[0]/ModuleList[blocks]/SwinLayer[1]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
12
+ 10,3,FF,nncf_module.swin.encoder.layers.0.blocks.1.intermediate.dense,"(512, 128)","(392, 128)","(512,)","(392,)",[392 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[0]/ModuleList[blocks]/SwinLayer[1]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
13
+ 11,3,FF,nncf_module.swin.encoder.layers.0.blocks.1.output.dense,"(128, 512)","(128, 392)","(128,)","(128,)",[392 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[0]/ModuleList[blocks]/SwinLayer[1]/SwinOutput[output]/NNCFLinear[dense]/linear_0
14
+ 12,4,MHSA,nncf_module.swin.encoder.layers.1.blocks.0.attention.self.query,"(256, 256)","(128, 256)","(256,)","(128,)","[1, 3, 5, 7]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[1]/ModuleList[blocks]/SwinLayer[0]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
15
+ 13,4,MHSA,nncf_module.swin.encoder.layers.1.blocks.0.attention.self.key,"(256, 256)","(128, 256)","(256,)","(128,)","[1, 3, 5, 7]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[1]/ModuleList[blocks]/SwinLayer[0]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
16
+ 14,4,MHSA,nncf_module.swin.encoder.layers.1.blocks.0.attention.self.value,"(256, 256)","(128, 256)","(256,)","(128,)","[1, 3, 5, 7]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[1]/ModuleList[blocks]/SwinLayer[0]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
17
+ 15,4,MHSA,nncf_module.swin.encoder.layers.1.blocks.0.attention.output.dense,"(256, 256)","(256, 128)","(256,)","(256,)","[1, 3, 5, 7]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[1]/ModuleList[blocks]/SwinLayer[0]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
18
+ 16,5,FF,nncf_module.swin.encoder.layers.1.blocks.0.intermediate.dense,"(1024, 256)","(794, 256)","(1024,)","(794,)",[794 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[1]/ModuleList[blocks]/SwinLayer[0]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
19
+ 17,5,FF,nncf_module.swin.encoder.layers.1.blocks.0.output.dense,"(256, 1024)","(256, 794)","(256,)","(256,)",[794 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[1]/ModuleList[blocks]/SwinLayer[0]/SwinOutput[output]/NNCFLinear[dense]/linear_0
20
+ 18,6,MHSA,nncf_module.swin.encoder.layers.1.blocks.1.attention.self.query,"(256, 256)","(128, 256)","(256,)","(128,)","[0, 1, 5, 6]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[1]/ModuleList[blocks]/SwinLayer[1]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
21
+ 19,6,MHSA,nncf_module.swin.encoder.layers.1.blocks.1.attention.self.key,"(256, 256)","(128, 256)","(256,)","(128,)","[0, 1, 5, 6]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[1]/ModuleList[blocks]/SwinLayer[1]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
22
+ 20,6,MHSA,nncf_module.swin.encoder.layers.1.blocks.1.attention.self.value,"(256, 256)","(128, 256)","(256,)","(128,)","[0, 1, 5, 6]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[1]/ModuleList[blocks]/SwinLayer[1]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
23
+ 21,6,MHSA,nncf_module.swin.encoder.layers.1.blocks.1.attention.output.dense,"(256, 256)","(256, 128)","(256,)","(256,)","[0, 1, 5, 6]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[1]/ModuleList[blocks]/SwinLayer[1]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
24
+ 22,7,FF,nncf_module.swin.encoder.layers.1.blocks.1.intermediate.dense,"(1024, 256)","(786, 256)","(1024,)","(786,)",[786 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[1]/ModuleList[blocks]/SwinLayer[1]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
25
+ 23,7,FF,nncf_module.swin.encoder.layers.1.blocks.1.output.dense,"(256, 1024)","(256, 786)","(256,)","(256,)",[786 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[1]/ModuleList[blocks]/SwinLayer[1]/SwinOutput[output]/NNCFLinear[dense]/linear_0
26
+ 24,8,MHSA,nncf_module.swin.encoder.layers.2.blocks.0.attention.self.query,"(512, 512)","(256, 512)","(512,)","(256,)","[3, 4, 5, 8, 9, 10, 13, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[0]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
27
+ 25,8,MHSA,nncf_module.swin.encoder.layers.2.blocks.0.attention.self.key,"(512, 512)","(256, 512)","(512,)","(256,)","[3, 4, 5, 8, 9, 10, 13, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[0]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
28
+ 26,8,MHSA,nncf_module.swin.encoder.layers.2.blocks.0.attention.self.value,"(512, 512)","(256, 512)","(512,)","(256,)","[3, 4, 5, 8, 9, 10, 13, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[0]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
29
+ 27,8,MHSA,nncf_module.swin.encoder.layers.2.blocks.0.attention.output.dense,"(512, 512)","(512, 256)","(512,)","(512,)","[3, 4, 5, 8, 9, 10, 13, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[0]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
30
+ 28,9,FF,nncf_module.swin.encoder.layers.2.blocks.0.intermediate.dense,"(2048, 512)","(1240, 512)","(2048,)","(1240,)",[1240 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[0]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
31
+ 29,9,FF,nncf_module.swin.encoder.layers.2.blocks.0.output.dense,"(512, 2048)","(512, 1240)","(512,)","(512,)",[1240 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[0]/SwinOutput[output]/NNCFLinear[dense]/linear_0
32
+ 30,10,MHSA,nncf_module.swin.encoder.layers.2.blocks.1.attention.self.query,"(512, 512)","(320, 512)","(512,)","(320,)","[1, 3, 6, 7, 9, 10, 11, 12, 13, 14]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[1]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
33
+ 31,10,MHSA,nncf_module.swin.encoder.layers.2.blocks.1.attention.self.key,"(512, 512)","(320, 512)","(512,)","(320,)","[1, 3, 6, 7, 9, 10, 11, 12, 13, 14]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[1]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
34
+ 32,10,MHSA,nncf_module.swin.encoder.layers.2.blocks.1.attention.self.value,"(512, 512)","(320, 512)","(512,)","(320,)","[1, 3, 6, 7, 9, 10, 11, 12, 13, 14]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[1]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
35
+ 33,10,MHSA,nncf_module.swin.encoder.layers.2.blocks.1.attention.output.dense,"(512, 512)","(512, 320)","(512,)","(512,)","[1, 3, 6, 7, 9, 10, 11, 12, 13, 14]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[1]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
36
+ 34,11,FF,nncf_module.swin.encoder.layers.2.blocks.1.intermediate.dense,"(2048, 512)","(1254, 512)","(2048,)","(1254,)",[1254 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[1]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
37
+ 35,11,FF,nncf_module.swin.encoder.layers.2.blocks.1.output.dense,"(512, 2048)","(512, 1254)","(512,)","(512,)",[1254 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[1]/SwinOutput[output]/NNCFLinear[dense]/linear_0
38
+ 36,12,MHSA,nncf_module.swin.encoder.layers.2.blocks.2.attention.self.query,"(512, 512)","(320, 512)","(512,)","(320,)","[1, 4, 5, 6, 7, 8, 9, 10, 11, 12]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[2]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
39
+ 37,12,MHSA,nncf_module.swin.encoder.layers.2.blocks.2.attention.self.key,"(512, 512)","(320, 512)","(512,)","(320,)","[1, 4, 5, 6, 7, 8, 9, 10, 11, 12]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[2]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
40
+ 38,12,MHSA,nncf_module.swin.encoder.layers.2.blocks.2.attention.self.value,"(512, 512)","(320, 512)","(512,)","(320,)","[1, 4, 5, 6, 7, 8, 9, 10, 11, 12]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[2]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
41
+ 39,12,MHSA,nncf_module.swin.encoder.layers.2.blocks.2.attention.output.dense,"(512, 512)","(512, 320)","(512,)","(512,)","[1, 4, 5, 6, 7, 8, 9, 10, 11, 12]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[2]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
42
+ 40,13,FF,nncf_module.swin.encoder.layers.2.blocks.2.intermediate.dense,"(2048, 512)","(1280, 512)","(2048,)","(1280,)",[1280 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[2]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
43
+ 41,13,FF,nncf_module.swin.encoder.layers.2.blocks.2.output.dense,"(512, 2048)","(512, 1280)","(512,)","(512,)",[1280 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[2]/SwinOutput[output]/NNCFLinear[dense]/linear_0
44
+ 42,14,MHSA,nncf_module.swin.encoder.layers.2.blocks.3.attention.self.query,"(512, 512)","(256, 512)","(512,)","(256,)","[0, 2, 3, 4, 6, 7, 9, 11]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[3]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
45
+ 43,14,MHSA,nncf_module.swin.encoder.layers.2.blocks.3.attention.self.key,"(512, 512)","(256, 512)","(512,)","(256,)","[0, 2, 3, 4, 6, 7, 9, 11]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[3]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
46
+ 44,14,MHSA,nncf_module.swin.encoder.layers.2.blocks.3.attention.self.value,"(512, 512)","(256, 512)","(512,)","(256,)","[0, 2, 3, 4, 6, 7, 9, 11]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[3]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
47
+ 45,14,MHSA,nncf_module.swin.encoder.layers.2.blocks.3.attention.output.dense,"(512, 512)","(512, 256)","(512,)","(512,)","[0, 2, 3, 4, 6, 7, 9, 11]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[3]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
48
+ 46,15,FF,nncf_module.swin.encoder.layers.2.blocks.3.intermediate.dense,"(2048, 512)","(1202, 512)","(2048,)","(1202,)",[1202 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[3]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
49
+ 47,15,FF,nncf_module.swin.encoder.layers.2.blocks.3.output.dense,"(512, 2048)","(512, 1202)","(512,)","(512,)",[1202 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[3]/SwinOutput[output]/NNCFLinear[dense]/linear_0
50
+ 48,16,MHSA,nncf_module.swin.encoder.layers.2.blocks.4.attention.self.query,"(512, 512)","(448, 512)","(512,)","(448,)","[0, 1, 2, 3, 4, 5, 6, 7, 9, 10, 11, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[4]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
51
+ 49,16,MHSA,nncf_module.swin.encoder.layers.2.blocks.4.attention.self.key,"(512, 512)","(448, 512)","(512,)","(448,)","[0, 1, 2, 3, 4, 5, 6, 7, 9, 10, 11, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[4]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
52
+ 50,16,MHSA,nncf_module.swin.encoder.layers.2.blocks.4.attention.self.value,"(512, 512)","(448, 512)","(512,)","(448,)","[0, 1, 2, 3, 4, 5, 6, 7, 9, 10, 11, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[4]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
53
+ 51,16,MHSA,nncf_module.swin.encoder.layers.2.blocks.4.attention.output.dense,"(512, 512)","(512, 448)","(512,)","(512,)","[0, 1, 2, 3, 4, 5, 6, 7, 9, 10, 11, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[4]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
54
+ 52,17,FF,nncf_module.swin.encoder.layers.2.blocks.4.intermediate.dense,"(2048, 512)","(1226, 512)","(2048,)","(1226,)",[1226 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[4]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
55
+ 53,17,FF,nncf_module.swin.encoder.layers.2.blocks.4.output.dense,"(512, 2048)","(512, 1226)","(512,)","(512,)",[1226 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[4]/SwinOutput[output]/NNCFLinear[dense]/linear_0
56
+ 54,18,MHSA,nncf_module.swin.encoder.layers.2.blocks.5.attention.self.query,"(512, 512)","(160, 512)","(512,)","(160,)","[0, 1, 5, 13, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[5]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
57
+ 55,18,MHSA,nncf_module.swin.encoder.layers.2.blocks.5.attention.self.key,"(512, 512)","(160, 512)","(512,)","(160,)","[0, 1, 5, 13, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[5]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
58
+ 56,18,MHSA,nncf_module.swin.encoder.layers.2.blocks.5.attention.self.value,"(512, 512)","(160, 512)","(512,)","(160,)","[0, 1, 5, 13, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[5]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
59
+ 57,18,MHSA,nncf_module.swin.encoder.layers.2.blocks.5.attention.output.dense,"(512, 512)","(512, 160)","(512,)","(512,)","[0, 1, 5, 13, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[5]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
60
+ 58,19,FF,nncf_module.swin.encoder.layers.2.blocks.5.intermediate.dense,"(2048, 512)","(1188, 512)","(2048,)","(1188,)",[1188 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[5]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
61
+ 59,19,FF,nncf_module.swin.encoder.layers.2.blocks.5.output.dense,"(512, 2048)","(512, 1188)","(512,)","(512,)",[1188 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[5]/SwinOutput[output]/NNCFLinear[dense]/linear_0
62
+ 60,20,MHSA,nncf_module.swin.encoder.layers.2.blocks.6.attention.self.query,"(512, 512)","(352, 512)","(512,)","(352,)","[0, 2, 3, 6, 7, 8, 9, 12, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[6]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
63
+ 61,20,MHSA,nncf_module.swin.encoder.layers.2.blocks.6.attention.self.key,"(512, 512)","(352, 512)","(512,)","(352,)","[0, 2, 3, 6, 7, 8, 9, 12, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[6]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
64
+ 62,20,MHSA,nncf_module.swin.encoder.layers.2.blocks.6.attention.self.value,"(512, 512)","(352, 512)","(512,)","(352,)","[0, 2, 3, 6, 7, 8, 9, 12, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[6]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
65
+ 63,20,MHSA,nncf_module.swin.encoder.layers.2.blocks.6.attention.output.dense,"(512, 512)","(512, 352)","(512,)","(512,)","[0, 2, 3, 6, 7, 8, 9, 12, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[6]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
66
+ 64,21,FF,nncf_module.swin.encoder.layers.2.blocks.6.intermediate.dense,"(2048, 512)","(1211, 512)","(2048,)","(1211,)",[1211 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[6]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
67
+ 65,21,FF,nncf_module.swin.encoder.layers.2.blocks.6.output.dense,"(512, 2048)","(512, 1211)","(512,)","(512,)",[1211 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[6]/SwinOutput[output]/NNCFLinear[dense]/linear_0
68
+ 66,22,MHSA,nncf_module.swin.encoder.layers.2.blocks.7.attention.self.query,"(512, 512)","(320, 512)","(512,)","(320,)","[0, 1, 2, 4, 5, 6, 11, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[7]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
69
+ 67,22,MHSA,nncf_module.swin.encoder.layers.2.blocks.7.attention.self.key,"(512, 512)","(320, 512)","(512,)","(320,)","[0, 1, 2, 4, 5, 6, 11, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[7]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
70
+ 68,22,MHSA,nncf_module.swin.encoder.layers.2.blocks.7.attention.self.value,"(512, 512)","(320, 512)","(512,)","(320,)","[0, 1, 2, 4, 5, 6, 11, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[7]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
71
+ 69,22,MHSA,nncf_module.swin.encoder.layers.2.blocks.7.attention.output.dense,"(512, 512)","(512, 320)","(512,)","(512,)","[0, 1, 2, 4, 5, 6, 11, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[7]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
72
+ 70,23,FF,nncf_module.swin.encoder.layers.2.blocks.7.intermediate.dense,"(2048, 512)","(1248, 512)","(2048,)","(1248,)",[1248 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[7]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
73
+ 71,23,FF,nncf_module.swin.encoder.layers.2.blocks.7.output.dense,"(512, 2048)","(512, 1248)","(512,)","(512,)",[1248 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[7]/SwinOutput[output]/NNCFLinear[dense]/linear_0
74
+ 72,24,MHSA,nncf_module.swin.encoder.layers.2.blocks.8.attention.self.query,"(512, 512)","(288, 512)","(512,)","(288,)","[3, 4, 5, 6, 8, 9, 10, 12, 13]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[8]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
75
+ 73,24,MHSA,nncf_module.swin.encoder.layers.2.blocks.8.attention.self.key,"(512, 512)","(288, 512)","(512,)","(288,)","[3, 4, 5, 6, 8, 9, 10, 12, 13]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[8]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
76
+ 74,24,MHSA,nncf_module.swin.encoder.layers.2.blocks.8.attention.self.value,"(512, 512)","(288, 512)","(512,)","(288,)","[3, 4, 5, 6, 8, 9, 10, 12, 13]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[8]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
77
+ 75,24,MHSA,nncf_module.swin.encoder.layers.2.blocks.8.attention.output.dense,"(512, 512)","(512, 288)","(512,)","(512,)","[3, 4, 5, 6, 8, 9, 10, 12, 13]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[8]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
78
+ 76,25,FF,nncf_module.swin.encoder.layers.2.blocks.8.intermediate.dense,"(2048, 512)","(1222, 512)","(2048,)","(1222,)",[1222 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[8]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
79
+ 77,25,FF,nncf_module.swin.encoder.layers.2.blocks.8.output.dense,"(512, 2048)","(512, 1222)","(512,)","(512,)",[1222 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[8]/SwinOutput[output]/NNCFLinear[dense]/linear_0
80
+ 78,26,MHSA,nncf_module.swin.encoder.layers.2.blocks.9.attention.self.query,"(512, 512)","(320, 512)","(512,)","(320,)","[0, 1, 2, 3, 4, 6, 7, 8, 9, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[9]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
81
+ 79,26,MHSA,nncf_module.swin.encoder.layers.2.blocks.9.attention.self.key,"(512, 512)","(320, 512)","(512,)","(320,)","[0, 1, 2, 3, 4, 6, 7, 8, 9, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[9]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
82
+ 80,26,MHSA,nncf_module.swin.encoder.layers.2.blocks.9.attention.self.value,"(512, 512)","(320, 512)","(512,)","(320,)","[0, 1, 2, 3, 4, 6, 7, 8, 9, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[9]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
83
+ 81,26,MHSA,nncf_module.swin.encoder.layers.2.blocks.9.attention.output.dense,"(512, 512)","(512, 320)","(512,)","(512,)","[0, 1, 2, 3, 4, 6, 7, 8, 9, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[9]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
84
+ 82,27,FF,nncf_module.swin.encoder.layers.2.blocks.9.intermediate.dense,"(2048, 512)","(1241, 512)","(2048,)","(1241,)",[1241 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[9]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
85
+ 83,27,FF,nncf_module.swin.encoder.layers.2.blocks.9.output.dense,"(512, 2048)","(512, 1241)","(512,)","(512,)",[1241 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[9]/SwinOutput[output]/NNCFLinear[dense]/linear_0
86
+ 84,28,MHSA,nncf_module.swin.encoder.layers.2.blocks.10.attention.self.query,"(512, 512)","(448, 512)","(512,)","(448,)","[0, 1, 2, 3, 4, 5, 6, 7, 9, 11, 12, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[10]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
87
+ 85,28,MHSA,nncf_module.swin.encoder.layers.2.blocks.10.attention.self.key,"(512, 512)","(448, 512)","(512,)","(448,)","[0, 1, 2, 3, 4, 5, 6, 7, 9, 11, 12, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[10]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
88
+ 86,28,MHSA,nncf_module.swin.encoder.layers.2.blocks.10.attention.self.value,"(512, 512)","(448, 512)","(512,)","(448,)","[0, 1, 2, 3, 4, 5, 6, 7, 9, 11, 12, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[10]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
89
+ 87,28,MHSA,nncf_module.swin.encoder.layers.2.blocks.10.attention.output.dense,"(512, 512)","(512, 448)","(512,)","(512,)","[0, 1, 2, 3, 4, 5, 6, 7, 9, 11, 12, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[10]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
90
+ 88,29,FF,nncf_module.swin.encoder.layers.2.blocks.10.intermediate.dense,"(2048, 512)","(1236, 512)","(2048,)","(1236,)",[1236 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[10]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
91
+ 89,29,FF,nncf_module.swin.encoder.layers.2.blocks.10.output.dense,"(512, 2048)","(512, 1236)","(512,)","(512,)",[1236 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[10]/SwinOutput[output]/NNCFLinear[dense]/linear_0
92
+ 90,30,MHSA,nncf_module.swin.encoder.layers.2.blocks.11.attention.self.query,"(512, 512)","(288, 512)","(512,)","(288,)","[2, 3, 6, 7, 8, 9, 11, 12, 13]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[11]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
93
+ 91,30,MHSA,nncf_module.swin.encoder.layers.2.blocks.11.attention.self.key,"(512, 512)","(288, 512)","(512,)","(288,)","[2, 3, 6, 7, 8, 9, 11, 12, 13]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[11]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
94
+ 92,30,MHSA,nncf_module.swin.encoder.layers.2.blocks.11.attention.self.value,"(512, 512)","(288, 512)","(512,)","(288,)","[2, 3, 6, 7, 8, 9, 11, 12, 13]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[11]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
95
+ 93,30,MHSA,nncf_module.swin.encoder.layers.2.blocks.11.attention.output.dense,"(512, 512)","(512, 288)","(512,)","(512,)","[2, 3, 6, 7, 8, 9, 11, 12, 13]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[11]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
96
+ 94,31,FF,nncf_module.swin.encoder.layers.2.blocks.11.intermediate.dense,"(2048, 512)","(1215, 512)","(2048,)","(1215,)",[1215 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[11]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
97
+ 95,31,FF,nncf_module.swin.encoder.layers.2.blocks.11.output.dense,"(512, 2048)","(512, 1215)","(512,)","(512,)",[1215 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[11]/SwinOutput[output]/NNCFLinear[dense]/linear_0
98
+ 96,32,MHSA,nncf_module.swin.encoder.layers.2.blocks.12.attention.self.query,"(512, 512)","(352, 512)","(512,)","(352,)","[1, 2, 3, 4, 5, 6, 7, 9, 10, 12, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[12]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
99
+ 97,32,MHSA,nncf_module.swin.encoder.layers.2.blocks.12.attention.self.key,"(512, 512)","(352, 512)","(512,)","(352,)","[1, 2, 3, 4, 5, 6, 7, 9, 10, 12, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[12]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
100
+ 98,32,MHSA,nncf_module.swin.encoder.layers.2.blocks.12.attention.self.value,"(512, 512)","(352, 512)","(512,)","(352,)","[1, 2, 3, 4, 5, 6, 7, 9, 10, 12, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[12]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
101
+ 99,32,MHSA,nncf_module.swin.encoder.layers.2.blocks.12.attention.output.dense,"(512, 512)","(512, 352)","(512,)","(512,)","[1, 2, 3, 4, 5, 6, 7, 9, 10, 12, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[12]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
102
+ 100,33,FF,nncf_module.swin.encoder.layers.2.blocks.12.intermediate.dense,"(2048, 512)","(1250, 512)","(2048,)","(1250,)",[1250 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[12]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
103
+ 101,33,FF,nncf_module.swin.encoder.layers.2.blocks.12.output.dense,"(512, 2048)","(512, 1250)","(512,)","(512,)",[1250 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[12]/SwinOutput[output]/NNCFLinear[dense]/linear_0
104
+ 102,34,MHSA,nncf_module.swin.encoder.layers.2.blocks.13.attention.self.query,"(512, 512)","(224, 512)","(512,)","(224,)","[2, 3, 4, 6, 8, 11, 12]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[13]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
105
+ 103,34,MHSA,nncf_module.swin.encoder.layers.2.blocks.13.attention.self.key,"(512, 512)","(224, 512)","(512,)","(224,)","[2, 3, 4, 6, 8, 11, 12]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[13]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
106
+ 104,34,MHSA,nncf_module.swin.encoder.layers.2.blocks.13.attention.self.value,"(512, 512)","(224, 512)","(512,)","(224,)","[2, 3, 4, 6, 8, 11, 12]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[13]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
107
+ 105,34,MHSA,nncf_module.swin.encoder.layers.2.blocks.13.attention.output.dense,"(512, 512)","(512, 224)","(512,)","(512,)","[2, 3, 4, 6, 8, 11, 12]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[13]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
108
+ 106,35,FF,nncf_module.swin.encoder.layers.2.blocks.13.intermediate.dense,"(2048, 512)","(1224, 512)","(2048,)","(1224,)",[1224 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[13]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
109
+ 107,35,FF,nncf_module.swin.encoder.layers.2.blocks.13.output.dense,"(512, 2048)","(512, 1224)","(512,)","(512,)",[1224 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[13]/SwinOutput[output]/NNCFLinear[dense]/linear_0
110
+ 108,36,MHSA,nncf_module.swin.encoder.layers.2.blocks.14.attention.self.query,"(512, 512)","(352, 512)","(512,)","(352,)","[0, 2, 3, 4, 5, 9, 10, 12, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[14]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
111
+ 109,36,MHSA,nncf_module.swin.encoder.layers.2.blocks.14.attention.self.key,"(512, 512)","(352, 512)","(512,)","(352,)","[0, 2, 3, 4, 5, 9, 10, 12, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[14]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
112
+ 110,36,MHSA,nncf_module.swin.encoder.layers.2.blocks.14.attention.self.value,"(512, 512)","(352, 512)","(512,)","(352,)","[0, 2, 3, 4, 5, 9, 10, 12, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[14]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
113
+ 111,36,MHSA,nncf_module.swin.encoder.layers.2.blocks.14.attention.output.dense,"(512, 512)","(512, 352)","(512,)","(512,)","[0, 2, 3, 4, 5, 9, 10, 12, 13, 14, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[14]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
114
+ 112,37,FF,nncf_module.swin.encoder.layers.2.blocks.14.intermediate.dense,"(2048, 512)","(1195, 512)","(2048,)","(1195,)",[1195 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[14]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
115
+ 113,37,FF,nncf_module.swin.encoder.layers.2.blocks.14.output.dense,"(512, 2048)","(512, 1195)","(512,)","(512,)",[1195 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[14]/SwinOutput[output]/NNCFLinear[dense]/linear_0
116
+ 114,38,MHSA,nncf_module.swin.encoder.layers.2.blocks.15.attention.self.query,"(512, 512)","(288, 512)","(512,)","(288,)","[1, 2, 3, 4, 6, 9, 10, 13, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[15]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
117
+ 115,38,MHSA,nncf_module.swin.encoder.layers.2.blocks.15.attention.self.key,"(512, 512)","(288, 512)","(512,)","(288,)","[1, 2, 3, 4, 6, 9, 10, 13, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[15]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
118
+ 116,38,MHSA,nncf_module.swin.encoder.layers.2.blocks.15.attention.self.value,"(512, 512)","(288, 512)","(512,)","(288,)","[1, 2, 3, 4, 6, 9, 10, 13, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[15]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
119
+ 117,38,MHSA,nncf_module.swin.encoder.layers.2.blocks.15.attention.output.dense,"(512, 512)","(512, 288)","(512,)","(512,)","[1, 2, 3, 4, 6, 9, 10, 13, 15]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[15]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
120
+ 118,39,FF,nncf_module.swin.encoder.layers.2.blocks.15.intermediate.dense,"(2048, 512)","(1093, 512)","(2048,)","(1093,)",[1093 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[15]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
121
+ 119,39,FF,nncf_module.swin.encoder.layers.2.blocks.15.output.dense,"(512, 2048)","(512, 1093)","(512,)","(512,)",[1093 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[15]/SwinOutput[output]/NNCFLinear[dense]/linear_0
122
+ 120,40,MHSA,nncf_module.swin.encoder.layers.2.blocks.16.attention.self.query,"(512, 512)","(256, 512)","(512,)","(256,)","[0, 2, 3, 6, 7, 8, 9, 14]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[16]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
123
+ 121,40,MHSA,nncf_module.swin.encoder.layers.2.blocks.16.attention.self.key,"(512, 512)","(256, 512)","(512,)","(256,)","[0, 2, 3, 6, 7, 8, 9, 14]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[16]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
124
+ 122,40,MHSA,nncf_module.swin.encoder.layers.2.blocks.16.attention.self.value,"(512, 512)","(256, 512)","(512,)","(256,)","[0, 2, 3, 6, 7, 8, 9, 14]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[16]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
125
+ 123,40,MHSA,nncf_module.swin.encoder.layers.2.blocks.16.attention.output.dense,"(512, 512)","(512, 256)","(512,)","(512,)","[0, 2, 3, 6, 7, 8, 9, 14]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[16]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
126
+ 124,41,FF,nncf_module.swin.encoder.layers.2.blocks.16.intermediate.dense,"(2048, 512)","(1026, 512)","(2048,)","(1026,)",[1026 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[16]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
127
+ 125,41,FF,nncf_module.swin.encoder.layers.2.blocks.16.output.dense,"(512, 2048)","(512, 1026)","(512,)","(512,)",[1026 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[16]/SwinOutput[output]/NNCFLinear[dense]/linear_0
128
+ 126,42,MHSA,nncf_module.swin.encoder.layers.2.blocks.17.attention.self.query,"(512, 512)","(224, 512)","(512,)","(224,)","[2, 3, 4, 5, 10, 11, 12]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[17]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
129
+ 127,42,MHSA,nncf_module.swin.encoder.layers.2.blocks.17.attention.self.key,"(512, 512)","(224, 512)","(512,)","(224,)","[2, 3, 4, 5, 10, 11, 12]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[17]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
130
+ 128,42,MHSA,nncf_module.swin.encoder.layers.2.blocks.17.attention.self.value,"(512, 512)","(224, 512)","(512,)","(224,)","[2, 3, 4, 5, 10, 11, 12]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[17]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
131
+ 129,42,MHSA,nncf_module.swin.encoder.layers.2.blocks.17.attention.output.dense,"(512, 512)","(512, 224)","(512,)","(512,)","[2, 3, 4, 5, 10, 11, 12]",SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[17]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
132
+ 130,43,FF,nncf_module.swin.encoder.layers.2.blocks.17.intermediate.dense,"(2048, 512)","(1078, 512)","(2048,)","(1078,)",[1078 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[17]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
133
+ 131,43,FF,nncf_module.swin.encoder.layers.2.blocks.17.output.dense,"(512, 2048)","(512, 1078)","(512,)","(512,)",[1078 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[2]/ModuleList[blocks]/SwinLayer[17]/SwinOutput[output]/NNCFLinear[dense]/linear_0
134
+ 132,44,MHSA,nncf_module.swin.encoder.layers.3.blocks.0.attention.self.query,"(1024, 1024)","(1024, 1024)","(1024,)","(1024,)",[32 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[3]/ModuleList[blocks]/SwinLayer[0]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
135
+ 133,44,MHSA,nncf_module.swin.encoder.layers.3.blocks.0.attention.self.key,"(1024, 1024)","(1024, 1024)","(1024,)","(1024,)",[32 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[3]/ModuleList[blocks]/SwinLayer[0]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
136
+ 134,44,MHSA,nncf_module.swin.encoder.layers.3.blocks.0.attention.self.value,"(1024, 1024)","(1024, 1024)","(1024,)","(1024,)",[32 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[3]/ModuleList[blocks]/SwinLayer[0]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
137
+ 135,44,MHSA,nncf_module.swin.encoder.layers.3.blocks.0.attention.output.dense,"(1024, 1024)","(1024, 1024)","(1024,)","(1024,)",[32 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[3]/ModuleList[blocks]/SwinLayer[0]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
138
+ 136,45,FF,nncf_module.swin.encoder.layers.3.blocks.0.intermediate.dense,"(4096, 1024)","(2153, 1024)","(4096,)","(2153,)",[2153 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[3]/ModuleList[blocks]/SwinLayer[0]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
139
+ 137,45,FF,nncf_module.swin.encoder.layers.3.blocks.0.output.dense,"(1024, 4096)","(1024, 2153)","(1024,)","(1024,)",[2153 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[3]/ModuleList[blocks]/SwinLayer[0]/SwinOutput[output]/NNCFLinear[dense]/linear_0
140
+ 138,46,MHSA,nncf_module.swin.encoder.layers.3.blocks.1.attention.self.query,"(1024, 1024)","(1024, 1024)","(1024,)","(1024,)",[32 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[3]/ModuleList[blocks]/SwinLayer[1]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[query]/linear_0
141
+ 139,46,MHSA,nncf_module.swin.encoder.layers.3.blocks.1.attention.self.key,"(1024, 1024)","(1024, 1024)","(1024,)","(1024,)",[32 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[3]/ModuleList[blocks]/SwinLayer[1]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[key]/linear_0
142
+ 140,46,MHSA,nncf_module.swin.encoder.layers.3.blocks.1.attention.self.value,"(1024, 1024)","(1024, 1024)","(1024,)","(1024,)",[32 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[3]/ModuleList[blocks]/SwinLayer[1]/SwinAttention[attention]/SwinSelfAttention[self]/NNCFLinear[value]/linear_0
143
+ 141,46,MHSA,nncf_module.swin.encoder.layers.3.blocks.1.attention.output.dense,"(1024, 1024)","(1024, 1024)","(1024,)","(1024,)",[32 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[3]/ModuleList[blocks]/SwinLayer[1]/SwinAttention[attention]/SwinSelfOutput[output]/NNCFLinear[dense]/linear_0
144
+ 142,47,FF,nncf_module.swin.encoder.layers.3.blocks.1.intermediate.dense,"(4096, 1024)","(1925, 1024)","(4096,)","(1925,)",[1925 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[3]/ModuleList[blocks]/SwinLayer[1]/SwinIntermediate[intermediate]/NNCFLinear[dense]/linear_0
145
+ 143,47,FF,nncf_module.swin.encoder.layers.3.blocks.1.output.dense,"(1024, 4096)","(1024, 1925)","(1024,)","(1024,)",[1925 items],SwinForImageClassification/SwinModel[swin]/SwinEncoder[encoder]/ModuleList[layers]/SwinStage[3]/ModuleList[blocks]/SwinLayer[1]/SwinOutput[output]/NNCFLinear[dense]/linear_0
trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c8a807963eda0954f61bcb8b8b6ae91adba3f21a701e7907725f28ab2cc1ea57
3
+ size 3771