baiyu858 commited on
Commit
b7aef8f
·
verified ·
1 Parent(s): dad9de0

Add files using upload-large-folder tool

Browse files
config.json ADDED
@@ -0,0 +1,300 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "action_mode": "ee6d",
3
+ "architectures": [
4
+ "WorldBridgeXVLA"
5
+ ],
6
+ "auto_map": {
7
+ "AutoConfig": "configuration_xvla.XVLAConfig",
8
+ "AutoModel": "modeling_xvla.XVLA"
9
+ },
10
+ "bridge_change_bias_scale": 1.0,
11
+ "bridge_flow_bias_scale": 1.0,
12
+ "bridge_injection_start_layer": null,
13
+ "bridge_lora_alpha": 16.0,
14
+ "bridge_lora_dropout": 0.0,
15
+ "bridge_lora_last_n_blocks": 0,
16
+ "bridge_lora_only_bridge_layers": false,
17
+ "bridge_lora_rank": 0,
18
+ "bridge_lora_target_modules": [
19
+ "attn.q",
20
+ "attn.qkv",
21
+ "attn.kv_self",
22
+ "attn.k_future",
23
+ "attn.v_future",
24
+ "attn.k_change",
25
+ "attn.v_change",
26
+ "attn.k_flow",
27
+ "attn.v_flow",
28
+ "attn.proj",
29
+ "mlp.fc1",
30
+ "mlp.fc2",
31
+ "modulation.mlp.1",
32
+ "modulation.mlp.3"
33
+ ],
34
+ "bridge_modulation_scale": 0.1,
35
+ "bridge_num_injected_layers": 4,
36
+ "bridge_use_change_gate": true,
37
+ "bridge_use_flow_gate": true,
38
+ "bridge_use_future_gate": true,
39
+ "change_distill_cosine_weight": 0.1,
40
+ "change_distill_weight": 0.1,
41
+ "change_map_pool_hw": 4,
42
+ "change_predictor_mlp_ratio": 2.0,
43
+ "change_predictor_num_heads": 8,
44
+ "change_token_count": 16,
45
+ "depth": 24,
46
+ "dim_time": 32,
47
+ "florence_config": {
48
+ "_attn_implementation_autoset": true,
49
+ "bos_token_id": 0,
50
+ "eos_token_id": 2,
51
+ "ignore_index": -100,
52
+ "is_encoder_decoder": true,
53
+ "model_type": "florence2",
54
+ "pad_token_id": 1,
55
+ "projection_dim": 1024,
56
+ "text_config": {
57
+ "_attn_implementation_autoset": true,
58
+ "_name_or_path": "",
59
+ "activation_dropout": 0.1,
60
+ "activation_function": "gelu",
61
+ "add_cross_attention": false,
62
+ "architectures": null,
63
+ "attention_dropout": 0.1,
64
+ "bad_words_ids": null,
65
+ "begin_suppress_tokens": null,
66
+ "bos_token_id": 0,
67
+ "chunk_size_feed_forward": 0,
68
+ "classifier_dropout": 0.0,
69
+ "cross_attention_hidden_size": null,
70
+ "d_model": 1024,
71
+ "decoder_attention_heads": 16,
72
+ "decoder_ffn_dim": 4096,
73
+ "decoder_layerdrop": 0.0,
74
+ "decoder_layers": 12,
75
+ "decoder_start_token_id": 2,
76
+ "diversity_penalty": 0.0,
77
+ "do_sample": false,
78
+ "dropout": 0.1,
79
+ "early_stopping": false,
80
+ "encoder_attention_heads": 16,
81
+ "encoder_ffn_dim": 4096,
82
+ "encoder_layerdrop": 0.0,
83
+ "encoder_layers": 12,
84
+ "encoder_no_repeat_ngram_size": 0,
85
+ "eos_token_id": 2,
86
+ "exponential_decay_length_penalty": null,
87
+ "finetuning_task": null,
88
+ "forced_bos_token_id": null,
89
+ "forced_eos_token_id": 2,
90
+ "id2label": {
91
+ "0": "LABEL_0",
92
+ "1": "LABEL_1",
93
+ "2": "LABEL_2"
94
+ },
95
+ "init_std": 0.02,
96
+ "is_decoder": false,
97
+ "is_encoder_decoder": true,
98
+ "label2id": {
99
+ "LABEL_0": 0,
100
+ "LABEL_1": 1,
101
+ "LABEL_2": 2
102
+ },
103
+ "length_penalty": 1.0,
104
+ "max_length": 20,
105
+ "max_position_embeddings": 4096,
106
+ "min_length": 0,
107
+ "model_type": "florence2_language",
108
+ "no_repeat_ngram_size": 0,
109
+ "num_beam_groups": 1,
110
+ "num_beams": 3,
111
+ "num_hidden_layers": 12,
112
+ "num_return_sequences": 1,
113
+ "output_attentions": false,
114
+ "output_hidden_states": false,
115
+ "output_scores": false,
116
+ "pad_token_id": 1,
117
+ "prefix": null,
118
+ "problem_type": null,
119
+ "pruned_heads": {},
120
+ "remove_invalid_values": false,
121
+ "repetition_penalty": 1.0,
122
+ "return_dict": true,
123
+ "return_dict_in_generate": false,
124
+ "scale_embedding": false,
125
+ "sep_token_id": null,
126
+ "suppress_tokens": null,
127
+ "task_specific_params": null,
128
+ "temperature": 1.0,
129
+ "tf_legacy_loss": false,
130
+ "tie_encoder_decoder": false,
131
+ "tie_word_embeddings": true,
132
+ "tokenizer_class": null,
133
+ "top_k": 50,
134
+ "top_p": 1.0,
135
+ "torch_dtype": null,
136
+ "torchscript": false,
137
+ "typical_p": 1.0,
138
+ "use_bfloat16": false,
139
+ "use_cache": true,
140
+ "vocab_size": 51289
141
+ },
142
+ "torch_dtype": "float16",
143
+ "vision_config": {
144
+ "_attn_implementation_autoset": false,
145
+ "_name_or_path": "",
146
+ "add_cross_attention": false,
147
+ "architectures": null,
148
+ "bad_words_ids": null,
149
+ "begin_suppress_tokens": null,
150
+ "bos_token_id": null,
151
+ "chunk_size_feed_forward": 0,
152
+ "cross_attention_hidden_size": null,
153
+ "decoder_start_token_id": null,
154
+ "depths": [
155
+ 1,
156
+ 1,
157
+ 9,
158
+ 1
159
+ ],
160
+ "dim_embed": [
161
+ 256,
162
+ 512,
163
+ 1024,
164
+ 2048
165
+ ],
166
+ "diversity_penalty": 0.0,
167
+ "do_sample": false,
168
+ "drop_path_rate": 0.1,
169
+ "early_stopping": false,
170
+ "enable_checkpoint": false,
171
+ "encoder_no_repeat_ngram_size": 0,
172
+ "eos_token_id": null,
173
+ "exponential_decay_length_penalty": null,
174
+ "finetuning_task": null,
175
+ "forced_bos_token_id": null,
176
+ "forced_eos_token_id": null,
177
+ "id2label": {
178
+ "0": "LABEL_0",
179
+ "1": "LABEL_1"
180
+ },
181
+ "image_feature_source": [
182
+ "spatial_avg_pool",
183
+ "temporal_avg_pool"
184
+ ],
185
+ "image_pos_embed": {
186
+ "max_pos_embeddings": 50,
187
+ "type": "learned_abs_2d"
188
+ },
189
+ "is_decoder": false,
190
+ "is_encoder_decoder": false,
191
+ "label2id": {
192
+ "LABEL_0": 0,
193
+ "LABEL_1": 1
194
+ },
195
+ "length_penalty": 1.0,
196
+ "max_length": 20,
197
+ "min_length": 0,
198
+ "model_type": "davit",
199
+ "no_repeat_ngram_size": 0,
200
+ "num_beam_groups": 1,
201
+ "num_beams": 1,
202
+ "num_groups": [
203
+ 8,
204
+ 16,
205
+ 32,
206
+ 64
207
+ ],
208
+ "num_heads": [
209
+ 8,
210
+ 16,
211
+ 32,
212
+ 64
213
+ ],
214
+ "num_return_sequences": 1,
215
+ "output_attentions": false,
216
+ "output_hidden_states": false,
217
+ "output_scores": false,
218
+ "pad_token_id": null,
219
+ "patch_padding": [
220
+ 3,
221
+ 1,
222
+ 1,
223
+ 1
224
+ ],
225
+ "patch_prenorm": [
226
+ false,
227
+ true,
228
+ true,
229
+ true
230
+ ],
231
+ "patch_size": [
232
+ 7,
233
+ 3,
234
+ 3,
235
+ 3
236
+ ],
237
+ "patch_stride": [
238
+ 4,
239
+ 2,
240
+ 2,
241
+ 2
242
+ ],
243
+ "prefix": null,
244
+ "problem_type": null,
245
+ "projection_dim": 1024,
246
+ "pruned_heads": {},
247
+ "remove_invalid_values": false,
248
+ "repetition_penalty": 1.0,
249
+ "return_dict": true,
250
+ "return_dict_in_generate": false,
251
+ "sep_token_id": null,
252
+ "suppress_tokens": null,
253
+ "task_specific_params": null,
254
+ "temperature": 1.0,
255
+ "tf_legacy_loss": false,
256
+ "tie_encoder_decoder": false,
257
+ "tie_word_embeddings": true,
258
+ "tokenizer_class": null,
259
+ "top_k": 50,
260
+ "top_p": 1.0,
261
+ "torch_dtype": null,
262
+ "torchscript": false,
263
+ "typical_p": 1.0,
264
+ "use_bfloat16": false,
265
+ "visual_temporal_embedding": {
266
+ "max_temporal_embeddings": 100,
267
+ "type": "COSINE"
268
+ },
269
+ "window_size": 12
270
+ },
271
+ "vocab_size": 51289
272
+ },
273
+ "flow_distill_cosine_weight": 0.1,
274
+ "flow_distill_weight": 0.1,
275
+ "flow_map_pool_hw": 4,
276
+ "flow_predictor_mlp_ratio": 2.0,
277
+ "flow_predictor_num_heads": 8,
278
+ "flow_token_count": 16,
279
+ "future_distill_cosine_weight": 0.1,
280
+ "future_distill_weight": 0.1,
281
+ "future_predictor_mlp_ratio": 2.0,
282
+ "future_predictor_num_heads": 8,
283
+ "future_token_count": 4,
284
+ "future_token_pool_hw": 2,
285
+ "hidden_size": 1024,
286
+ "len_soft_prompts": 32,
287
+ "max_action_dim": 20,
288
+ "max_len_seq": 512,
289
+ "mlp_ratio": 4.0,
290
+ "model_type": "world_bridge_xvla",
291
+ "num_actions": 30,
292
+ "num_domains": 30,
293
+ "num_heads": 16,
294
+ "real_action_dim": 20,
295
+ "soft_prompt_length": 32,
296
+ "torch_dtype": "float32",
297
+ "transformers_version": "4.51.3",
298
+ "use_hetero_proj": false,
299
+ "use_proprio": true
300
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c29f4d1b7149b8031e43f69165ecbfe5f317d8784550832239f731421c9afd6
3
+ size 4023608024
preprocessor_config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_map": {
3
+ "AutoProcessor": "processing_xvla.XVLAProcessor"
4
+ },
5
+ "crop_size": {
6
+ "height": 224,
7
+ "width": 224
8
+ },
9
+ "do_center_crop": false,
10
+ "do_convert_rgb": null,
11
+ "do_normalize": true,
12
+ "do_rescale": true,
13
+ "do_resize": true,
14
+ "image_mean": [
15
+ 0.485,
16
+ 0.456,
17
+ 0.406
18
+ ],
19
+ "image_processor_type": "CLIPImageProcessor",
20
+ "image_std": [
21
+ 0.229,
22
+ 0.224,
23
+ 0.225
24
+ ],
25
+ "processor_class": "XVLAProcessor",
26
+ "resample": 3,
27
+ "rescale_factor": 0.00392156862745098,
28
+ "size": {
29
+ "height": 224,
30
+ "width": 224
31
+ }
32
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": true,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": true,
26
+ "normalized": true,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": true,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "</s>",
39
+ "lstrip": false,
40
+ "normalized": true,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "<unk>",
46
+ "lstrip": false,
47
+ "normalized": true,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
state.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"global_step": 20000}
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "<s>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "<pad>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "2": {
21
+ "content": "</s>",
22
+ "lstrip": false,
23
+ "normalized": true,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "3": {
29
+ "content": "<unk>",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "50264": {
37
+ "content": "<mask>",
38
+ "lstrip": true,
39
+ "normalized": true,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ }
44
+ },
45
+ "bos_token": "<s>",
46
+ "clean_up_tokenization_spaces": false,
47
+ "cls_token": "<s>",
48
+ "eos_token": "</s>",
49
+ "errors": "replace",
50
+ "extra_special_tokens": {},
51
+ "mask_token": "<mask>",
52
+ "max_length": 50,
53
+ "model_max_length": 1024,
54
+ "pad_to_multiple_of": null,
55
+ "pad_token": "<pad>",
56
+ "pad_token_type_id": 0,
57
+ "padding_side": "right",
58
+ "processor_class": "XVLAProcessor",
59
+ "sep_token": "</s>",
60
+ "stride": 0,
61
+ "tokenizer_class": "BartTokenizer",
62
+ "trim_offsets": true,
63
+ "truncation_side": "right",
64
+ "truncation_strategy": "longest_first",
65
+ "unk_token": "<unk>"
66
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff