Hugging Face's logo

flax-community
/
multilingual-vqa-pt-ckpts

TensorBoard
Model card Files Metrics Community
gchhablani commited on
Commit
06a770b
•
1 Parent(s): 37d35c0

Reset Repo

Browse files
Files changed (5) hide show
  1. ckpt-4999/config.json +0 -305
  2. ckpt-4999/flax_model.msgpack +0 -3
  3. ckpt-4999/opt_state.msgpack +0 -3
  4. ckpt-4999/training_state.json +0 -1
  5. events.out.tfevents.1626223943.t1v-n-8499dd91-w-0.592185.3.v2 +0 -3
ckpt-4999/config.json DELETED
@@ -1,305 +0,0 @@
1
- {
2
- "architectures": [
3
- "CLIPVisionBertForMaskedLM"
4
- ],
5
- "bert_config": {
6
- "_name_or_path": "",
7
- "add_cross_attention": false,
8
- "architectures": [
9
- "BertForMaskedLM"
10
- ],
11
- "attention_probs_dropout_prob": 0.1,
12
- "bad_words_ids": null,
13
- "bos_token_id": null,
14
- "chunk_size_feed_forward": 0,
15
- "decoder_start_token_id": null,
16
- "directionality": "bidi",
17
- "diversity_penalty": 0.0,
18
- "do_sample": false,
19
- "early_stopping": false,
20
- "encoder_no_repeat_ngram_size": 0,
21
- "eos_token_id": null,
22
- "finetuning_task": null,
23
- "forced_bos_token_id": null,
24
- "forced_eos_token_id": null,
25
- "gradient_checkpointing": false,
26
- "hidden_act": "gelu",
27
- "hidden_dropout_prob": 0.1,
28
- "hidden_size": 768,
29
- "id2label": {
30
- "0": "LABEL_0",
31
- "1": "LABEL_1"
32
- },
33
- "initializer_range": 0.02,
34
- "intermediate_size": 3072,
35
- "is_decoder": false,
36
- "is_encoder_decoder": false,
37
- "label2id": {
38
- "LABEL_0": 0,
39
- "LABEL_1": 1
40
- },
41
- "layer_norm_eps": 1e-12,
42
- "length_penalty": 1.0,
43
- "max_length": 20,
44
- "max_position_embeddings": 512,
45
- "min_length": 0,
46
- "model_type": "bert",
47
- "no_repeat_ngram_size": 0,
48
- "num_attention_heads": 12,
49
- "num_beam_groups": 1,
50
- "num_beams": 1,
51
- "num_hidden_layers": 12,
52
- "num_return_sequences": 1,
53
- "output_attentions": false,
54
- "output_hidden_states": false,
55
- "output_scores": false,
56
- "pad_token_id": 0,
57
- "pooler_fc_size": 768,
58
- "pooler_num_attention_heads": 12,
59
- "pooler_num_fc_layers": 3,
60
- "pooler_size_per_head": 128,
61
- "pooler_type": "first_token_transform",
62
- "position_embedding_type": "absolute",
63
- "prefix": null,
64
- "problem_type": null,
65
- "pruned_heads": {},
66
- "remove_invalid_values": false,
67
- "repetition_penalty": 1.0,
68
- "return_dict": true,
69
- "return_dict_in_generate": false,
70
- "sep_token_id": null,
71
- "task_specific_params": null,
72
- "temperature": 1.0,
73
- "tie_encoder_decoder": false,
74
- "tie_word_embeddings": true,
75
- "tokenizer_class": null,
76
- "top_k": 50,
77
- "top_p": 1.0,
78
- "torch_dtype": null,
79
- "torchscript": false,
80
- "transformers_version": "4.9.0.dev0",
81
- "type_vocab_size": 2,
82
- "use_bfloat16": false,
83
- "use_cache": true,
84
- "vocab_size": 105879
85
- },
86
- "clip_vision_config": {
87
- "_name_or_path": "",
88
- "add_cross_attention": false,
89
- "architectures": [
90
- "CLIPModel"
91
- ],
92
- "attention_dropout": 0.0,
93
- "bad_words_ids": null,
94
- "bos_token_id": null,
95
- "chunk_size_feed_forward": 0,
96
- "decoder_start_token_id": null,
97
- "diversity_penalty": 0.0,
98
- "do_sample": false,
99
- "dropout": 0.0,
100
- "early_stopping": false,
101
- "encoder_no_repeat_ngram_size": 0,
102
- "eos_token_id": null,
103
- "finetuning_task": null,
104
- "forced_bos_token_id": null,
105
- "forced_eos_token_id": null,
106
- "gradient_checkpointing": false,
107
- "hidden_act": "quick_gelu",
108
- "hidden_size": 768,
109
- "id2label": {
110
- "0": "LABEL_0",
111
- "1": "LABEL_1"
112
- },
113
- "image_size": 224,
114
- "initializer_factor": 1.0,
115
- "initializer_range": 0.02,
116
- "intermediate_size": 3072,
117
- "is_decoder": false,
118
- "is_encoder_decoder": false,
119
- "label2id": {
120
- "LABEL_0": 0,
121
- "LABEL_1": 1
122
- },
123
- "layer_norm_eps": 1e-05,
124
- "length_penalty": 1.0,
125
- "max_length": 20,
126
- "min_length": 0,
127
- "model_type": "clip_vision_model",
128
- "no_repeat_ngram_size": 0,
129
- "num_attention_heads": 12,
130
- "num_beam_groups": 1,
131
- "num_beams": 1,
132
- "num_hidden_layers": 12,
133
- "num_return_sequences": 1,
134
- "output_attentions": false,
135
- "output_hidden_states": false,
136
- "output_scores": false,
137
- "pad_token_id": null,
138
- "patch_size": 32,
139
- "prefix": null,
140
- "problem_type": null,
141
- "projection_dim": 512,
142
- "pruned_heads": {},
143
- "remove_invalid_values": false,
144
- "repetition_penalty": 1.0,
145
- "return_dict": true,
146
- "return_dict_in_generate": false,
147
- "sep_token_id": null,
148
- "task_specific_params": null,
149
- "temperature": 1.0,
150
- "text_config": {
151
- "_name_or_path": "",
152
- "add_cross_attention": false,
153
- "architectures": null,
154
- "attention_dropout": 0.0,
155
- "bad_words_ids": null,
156
- "bos_token_id": 0,
157
- "chunk_size_feed_forward": 0,
158
- "decoder_start_token_id": null,
159
- "diversity_penalty": 0.0,
160
- "do_sample": false,
161
- "dropout": 0.0,
162
- "early_stopping": false,
163
- "encoder_no_repeat_ngram_size": 0,
164
- "eos_token_id": 2,
165
- "finetuning_task": null,
166
- "forced_bos_token_id": null,
167
- "forced_eos_token_id": null,
168
- "gradient_checkpointing": false,
169
- "hidden_act": "quick_gelu",
170
- "hidden_size": 512,
171
- "id2label": {
172
- "0": "LABEL_0",
173
- "1": "LABEL_1"
174
- },
175
- "initializer_factor": 1.0,
176
- "initializer_range": 0.02,
177
- "intermediate_size": 2048,
178
- "is_decoder": false,
179
- "is_encoder_decoder": false,
180
- "label2id": {
181
- "LABEL_0": 0,
182
- "LABEL_1": 1
183
- },
184
- "layer_norm_eps": 1e-05,
185
- "length_penalty": 1.0,
186
- "max_length": 20,
187
- "max_position_embeddings": 77,
188
- "min_length": 0,
189
- "model_type": "clip_text_model",
190
- "no_repeat_ngram_size": 0,
191
- "num_attention_heads": 8,
192
- "num_beam_groups": 1,
193
- "num_beams": 1,
194
- "num_hidden_layers": 12,
195
- "num_return_sequences": 1,
196
- "output_attentions": false,
197
- "output_hidden_states": false,
198
- "output_scores": false,
199
- "pad_token_id": 1,
200
- "prefix": null,
201
- "problem_type": null,
202
- "pruned_heads": {},
203
- "remove_invalid_values": false,
204
- "repetition_penalty": 1.0,
205
- "return_dict": true,
206
- "return_dict_in_generate": false,
207
- "sep_token_id": null,
208
- "task_specific_params": null,
209
- "temperature": 1.0,
210
- "tie_encoder_decoder": false,
211
- "tie_word_embeddings": true,
212
- "tokenizer_class": null,
213
- "top_k": 50,
214
- "top_p": 1.0,
215
- "torchscript": false,
216
- "transformers_version": "4.7.0.dev0",
217
- "use_bfloat16": false,
218
- "vocab_size": 49408
219
- },
220
- "text_config_dict": null,
221
- "tie_encoder_decoder": false,
222
- "tie_word_embeddings": true,
223
- "tokenizer_class": null,
224
- "top_k": 50,
225
- "top_p": 1.0,
226
- "torch_dtype": null,
227
- "torchscript": false,
228
- "transformers_version": "4.9.0.dev0",
229
- "use_bfloat16": false,
230
- "vision_config": {
231
- "_name_or_path": "",
232
- "add_cross_attention": false,
233
- "architectures": null,
234
- "attention_dropout": 0.0,
235
- "bad_words_ids": null,
236
- "bos_token_id": null,
237
- "chunk_size_feed_forward": 0,
238
- "decoder_start_token_id": null,
239
- "diversity_penalty": 0.0,
240
- "do_sample": false,
241
- "dropout": 0.0,
242
- "early_stopping": false,
243
- "encoder_no_repeat_ngram_size": 0,
244
- "eos_token_id": null,
245
- "finetuning_task": null,
246
- "forced_bos_token_id": null,
247
- "forced_eos_token_id": null,
248
- "gradient_checkpointing": false,
249
- "hidden_act": "quick_gelu",
250
- "hidden_size": 768,
251
- "id2label": {
252
- "0": "LABEL_0",
253
- "1": "LABEL_1"
254
- },
255
- "image_size": 224,
256
- "initializer_factor": 1.0,
257
- "initializer_range": 0.02,
258
- "intermediate_size": 3072,
259
- "is_decoder": false,
260
- "is_encoder_decoder": false,
261
- "label2id": {
262
- "LABEL_0": 0,
263
- "LABEL_1": 1
264
- },
265
- "layer_norm_eps": 1e-05,
266
- "length_penalty": 1.0,
267
- "max_length": 20,
268
- "min_length": 0,
269
- "model_type": "clip_vision_model",
270
- "no_repeat_ngram_size": 0,
271
- "num_attention_heads": 12,
272
- "num_beam_groups": 1,
273
- "num_beams": 1,
274
- "num_hidden_layers": 12,
275
- "num_return_sequences": 1,
276
- "output_attentions": false,
277
- "output_hidden_states": false,
278
- "output_scores": false,
279
- "pad_token_id": null,
280
- "patch_size": 32,
281
- "prefix": null,
282
- "problem_type": null,
283
- "pruned_heads": {},
284
- "remove_invalid_values": false,
285
- "repetition_penalty": 1.0,
286
- "return_dict": true,
287
- "return_dict_in_generate": false,
288
- "sep_token_id": null,
289
- "task_specific_params": null,
290
- "temperature": 1.0,
291
- "tie_encoder_decoder": false,
292
- "tie_word_embeddings": true,
293
- "tokenizer_class": null,
294
- "top_k": 50,
295
- "top_p": 1.0,
296
- "torchscript": false,
297
- "transformers_version": "4.7.0.dev0",
298
- "use_bfloat16": false
299
- },
300
- "vision_config_dict": null
301
- },
302
- "model_type": "clip-vision-bert",
303
- "seed": 42,
304
- "transformers_version": null
305
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ckpt-4999/flax_model.msgpack DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:a29e008f515861c1638338a05f821e69cd284b330905d33ce73a2c57fa4f85d0
3
- size 1023635055
 
 
 
 
ckpt-4999/opt_state.msgpack DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:a5760b5497b54f1473dfe479536dfe4cc4e74e4ba435d0bbce6fea9b1900663b
3
- size 12853588
 
 
 
 
ckpt-4999/training_state.json DELETED
@@ -1 +0,0 @@
1
- {"step": 5000}
 
 
events.out.tfevents.1626223943.t1v-n-8499dd91-w-0.592185.3.v2 DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:facba75fdd9054b7b32d5dfa27cbef46e7df81a6d7da46b96fc1c3bddcd72853
3
- size 735204