Hugging Face's logo

ChiefTheLord
/
tinyvlm

Image-to-Text
PyTorch
English
vision
Inference Endpoints
Model card Files Community
ChiefTheLord commited on
Commit
93e243d
路
verified 路
1 Parent(s): 8262e1d

Upload folder using huggingface_hub

Browse files
Files changed (6) hide show
  1. flickr30k_checkpoints/checkpoint-776/adapter.safetensors +3 -0
  2. flickr30k_checkpoints/checkpoint-776/eval_state.json +0 -0
  3. flickr30k_checkpoints/checkpoint-776/optimizer.pt +3 -0
  4. flickr30k_checkpoints/checkpoint-776/rng_state.pth +3 -0
  5. flickr30k_checkpoints/checkpoint-776/scheduler.pt +3 -0
  6. flickr30k_checkpoints/checkpoint-776/trainer_state.json +408 -0
flickr30k_checkpoints/checkpoint-776/adapter.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8078156e4beb27f321d9a3433592e1f72690059504fffd4114c600843a014a0b
3
+ size 17064856
flickr30k_checkpoints/checkpoint-776/eval_state.json ADDED
The diff for this file is too large to render. See raw diff
 
flickr30k_checkpoints/checkpoint-776/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a0cb94ca55182f194642acb66a420b599f3c3a0d4183adff286822010aa1b5da
3
+ size 8714492
flickr30k_checkpoints/checkpoint-776/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d138cfe3a4adf21f048848ee35837c9a757a0a3616ff7adbb45b69aac247435
3
+ size 14244
flickr30k_checkpoints/checkpoint-776/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f9e776ccd1d99653e133bec1ef5fc4f228e94f728245cd5cb500ce34b60792e7
3
+ size 1064
flickr30k_checkpoints/checkpoint-776/trainer_state.json ADDED
@@ -0,0 +1,408 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.0,
5
+ "eval_steps": 500,
6
+ "global_step": 776,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.041237113402061855,
13
+ "grad_norm": 0.09626785665750504,
14
+ "learning_rate": 0.0006666666666666666,
15
+ "loss": 4.9272,
16
+ "step": 16
17
+ },
18
+ {
19
+ "epoch": 0.08247422680412371,
20
+ "grad_norm": 0.09776347130537033,
21
+ "learning_rate": 0.0009997207818651273,
22
+ "loss": 4.8403,
23
+ "step": 32
24
+ },
25
+ {
26
+ "epoch": 0.12371134020618557,
27
+ "grad_norm": 0.18136683106422424,
28
+ "learning_rate": 0.0009974889075442521,
29
+ "loss": 4.9176,
30
+ "step": 48
31
+ },
32
+ {
33
+ "epoch": 0.16494845360824742,
34
+ "grad_norm": 0.10246019065380096,
35
+ "learning_rate": 0.0009930351269950143,
36
+ "loss": 4.8586,
37
+ "step": 64
38
+ },
39
+ {
40
+ "epoch": 0.20618556701030927,
41
+ "grad_norm": 0.26106250286102295,
42
+ "learning_rate": 0.0009863793318825185,
43
+ "loss": 4.8768,
44
+ "step": 80
45
+ },
46
+ {
47
+ "epoch": 0.24742268041237114,
48
+ "grad_norm": 0.13016994297504425,
49
+ "learning_rate": 0.0009775512486034563,
50
+ "loss": 4.9537,
51
+ "step": 96
52
+ },
53
+ {
54
+ "epoch": 0.28865979381443296,
55
+ "grad_norm": 0.14532029628753662,
56
+ "learning_rate": 0.0009665903055208013,
57
+ "loss": 4.9051,
58
+ "step": 112
59
+ },
60
+ {
61
+ "epoch": 0.32989690721649484,
62
+ "grad_norm": 0.12766750156879425,
63
+ "learning_rate": 0.0009535454568671704,
64
+ "loss": 4.9145,
65
+ "step": 128
66
+ },
67
+ {
68
+ "epoch": 0.3711340206185567,
69
+ "grad_norm": 0.16211313009262085,
70
+ "learning_rate": 0.0009384749641033358,
71
+ "loss": 4.9382,
72
+ "step": 144
73
+ },
74
+ {
75
+ "epoch": 0.41237113402061853,
76
+ "grad_norm": 0.18606019020080566,
77
+ "learning_rate": 0.0009214461357083985,
78
+ "loss": 4.8932,
79
+ "step": 160
80
+ },
81
+ {
82
+ "epoch": 0.4536082474226804,
83
+ "grad_norm": 0.10332559049129486,
84
+ "learning_rate": 0.0009025350265637815,
85
+ "loss": 4.8723,
86
+ "step": 176
87
+ },
88
+ {
89
+ "epoch": 0.4948453608247423,
90
+ "grad_norm": 0.21772044897079468,
91
+ "learning_rate": 0.0008818260982736661,
92
+ "loss": 4.875,
93
+ "step": 192
94
+ },
95
+ {
96
+ "epoch": 0.5360824742268041,
97
+ "grad_norm": 0.1752215325832367,
98
+ "learning_rate": 0.0008594118419389648,
99
+ "loss": 4.9184,
100
+ "step": 208
101
+ },
102
+ {
103
+ "epoch": 0.5773195876288659,
104
+ "grad_norm": 0.15603622794151306,
105
+ "learning_rate": 0.0008353923650696118,
106
+ "loss": 4.8573,
107
+ "step": 224
108
+ },
109
+ {
110
+ "epoch": 0.6185567010309279,
111
+ "grad_norm": 0.19524474442005157,
112
+ "learning_rate": 0.0008098749444801224,
113
+ "loss": 4.9145,
114
+ "step": 240
115
+ },
116
+ {
117
+ "epoch": 0.6597938144329897,
118
+ "grad_norm": 0.14055490493774414,
119
+ "learning_rate": 0.0007829735471652976,
120
+ "loss": 4.9301,
121
+ "step": 256
122
+ },
123
+ {
124
+ "epoch": 0.7010309278350515,
125
+ "grad_norm": 0.1492496281862259,
126
+ "learning_rate": 0.0007548083212959587,
127
+ "loss": 4.8914,
128
+ "step": 272
129
+ },
130
+ {
131
+ "epoch": 0.7422680412371134,
132
+ "grad_norm": 0.20217546820640564,
133
+ "learning_rate": 0.0007255050596080509,
134
+ "loss": 4.9268,
135
+ "step": 288
136
+ },
137
+ {
138
+ "epoch": 0.7835051546391752,
139
+ "grad_norm": 0.12439503520727158,
140
+ "learning_rate": 0.0006951946375817474,
141
+ "loss": 4.8838,
142
+ "step": 304
143
+ },
144
+ {
145
+ "epoch": 0.8247422680412371,
146
+ "grad_norm": 0.15163038671016693,
147
+ "learning_rate": 0.0006640124289197845,
148
+ "loss": 4.9025,
149
+ "step": 320
150
+ },
151
+ {
152
+ "epoch": 0.865979381443299,
153
+ "grad_norm": 0.15424683690071106,
154
+ "learning_rate": 0.0006320977009356432,
155
+ "loss": 4.867,
156
+ "step": 336
157
+ },
158
+ {
159
+ "epoch": 0.9072164948453608,
160
+ "grad_norm": 0.12109530717134476,
161
+ "learning_rate": 0.000599592992551918,
162
+ "loss": 4.9068,
163
+ "step": 352
164
+ },
165
+ {
166
+ "epoch": 0.9484536082474226,
167
+ "grad_norm": 0.15034383535385132,
168
+ "learning_rate": 0.0005666434776868895,
169
+ "loss": 4.8993,
170
+ "step": 368
171
+ },
172
+ {
173
+ "epoch": 0.9896907216494846,
174
+ "grad_norm": 0.1772756278514862,
175
+ "learning_rate": 0.0005333963168725609,
176
+ "loss": 4.8522,
177
+ "step": 384
178
+ },
179
+ {
180
+ "epoch": 1.0,
181
+ "eval_bleu": 0.08197595877772341,
182
+ "eval_cap_loss": 1.7311131333567433,
183
+ "eval_con_loss": 2.0061442756775727,
184
+ "eval_loss": 3.737257410570518,
185
+ "step": 388
186
+ },
187
+ {
188
+ "epoch": 1.0,
189
+ "eval_bleu": 0.08197595877772341,
190
+ "eval_cap_loss": 1.7311131333567433,
191
+ "eval_con_loss": 2.0061442756775727,
192
+ "eval_loss": 3.737257410570518,
193
+ "eval_runtime": 142.0128,
194
+ "eval_samples_per_second": 21.836,
195
+ "eval_steps_per_second": 2.732,
196
+ "step": 388
197
+ },
198
+ {
199
+ "epoch": 1.0309278350515463,
200
+ "grad_norm": 0.1598144769668579,
201
+ "learning_rate": 0.0005,
202
+ "loss": 4.8689,
203
+ "step": 400
204
+ },
205
+ {
206
+ "epoch": 1.0721649484536082,
207
+ "grad_norm": 0.15655860304832458,
208
+ "learning_rate": 0.0004666036831274392,
209
+ "loss": 4.8865,
210
+ "step": 416
211
+ },
212
+ {
213
+ "epoch": 1.1134020618556701,
214
+ "grad_norm": 0.15307271480560303,
215
+ "learning_rate": 0.00043335652231311076,
216
+ "loss": 4.8548,
217
+ "step": 432
218
+ },
219
+ {
220
+ "epoch": 1.1546391752577319,
221
+ "grad_norm": 0.07996341586112976,
222
+ "learning_rate": 0.00040040700744808206,
223
+ "loss": 4.8571,
224
+ "step": 448
225
+ },
226
+ {
227
+ "epoch": 1.1958762886597938,
228
+ "grad_norm": 0.14137883484363556,
229
+ "learning_rate": 0.000367902299064357,
230
+ "loss": 4.9429,
231
+ "step": 464
232
+ },
233
+ {
234
+ "epoch": 1.2371134020618557,
235
+ "grad_norm": 0.14231054484844208,
236
+ "learning_rate": 0.00033598757108021545,
237
+ "loss": 4.8607,
238
+ "step": 480
239
+ },
240
+ {
241
+ "epoch": 1.2783505154639174,
242
+ "grad_norm": 0.1808304786682129,
243
+ "learning_rate": 0.00030480536241825264,
244
+ "loss": 4.8901,
245
+ "step": 496
246
+ },
247
+ {
248
+ "epoch": 1.3195876288659794,
249
+ "grad_norm": 0.14347229897975922,
250
+ "learning_rate": 0.000274494940391949,
251
+ "loss": 4.839,
252
+ "step": 512
253
+ },
254
+ {
255
+ "epoch": 1.3608247422680413,
256
+ "grad_norm": 0.12875933945178986,
257
+ "learning_rate": 0.00024519167870404123,
258
+ "loss": 4.8809,
259
+ "step": 528
260
+ },
261
+ {
262
+ "epoch": 1.402061855670103,
263
+ "grad_norm": 0.17960529029369354,
264
+ "learning_rate": 0.00021702645283470234,
265
+ "loss": 4.848,
266
+ "step": 544
267
+ },
268
+ {
269
+ "epoch": 1.443298969072165,
270
+ "grad_norm": 0.12456343322992325,
271
+ "learning_rate": 0.00019012505551987762,
272
+ "loss": 4.8828,
273
+ "step": 560
274
+ },
275
+ {
276
+ "epoch": 1.4845360824742269,
277
+ "grad_norm": 0.1851247102022171,
278
+ "learning_rate": 0.0001646076349303884,
279
+ "loss": 4.8708,
280
+ "step": 576
281
+ },
282
+ {
283
+ "epoch": 1.5257731958762886,
284
+ "grad_norm": 0.1590530276298523,
285
+ "learning_rate": 0.00014058815806103543,
286
+ "loss": 4.8611,
287
+ "step": 592
288
+ },
289
+ {
290
+ "epoch": 1.5670103092783505,
291
+ "grad_norm": 0.13291551172733307,
292
+ "learning_rate": 0.00011817390172633402,
293
+ "loss": 4.9032,
294
+ "step": 608
295
+ },
296
+ {
297
+ "epoch": 1.6082474226804124,
298
+ "grad_norm": 0.20336174964904785,
299
+ "learning_rate": 9.746497343621857e-05,
300
+ "loss": 4.8976,
301
+ "step": 624
302
+ },
303
+ {
304
+ "epoch": 1.6494845360824741,
305
+ "grad_norm": 0.13562831282615662,
306
+ "learning_rate": 7.85538642916015e-05,
307
+ "loss": 4.8911,
308
+ "step": 640
309
+ },
310
+ {
311
+ "epoch": 1.690721649484536,
312
+ "grad_norm": 0.13604836165905,
313
+ "learning_rate": 6.152503589666425e-05,
314
+ "loss": 4.8916,
315
+ "step": 656
316
+ },
317
+ {
318
+ "epoch": 1.731958762886598,
319
+ "grad_norm": 0.15518011152744293,
320
+ "learning_rate": 4.645454313282965e-05,
321
+ "loss": 4.8618,
322
+ "step": 672
323
+ },
324
+ {
325
+ "epoch": 1.7731958762886597,
326
+ "grad_norm": 0.11967703700065613,
327
+ "learning_rate": 3.340969447919873e-05,
328
+ "loss": 4.9006,
329
+ "step": 688
330
+ },
331
+ {
332
+ "epoch": 1.8144329896907216,
333
+ "grad_norm": 0.13278520107269287,
334
+ "learning_rate": 2.2448751396543787e-05,
335
+ "loss": 4.9169,
336
+ "step": 704
337
+ },
338
+ {
339
+ "epoch": 1.8556701030927836,
340
+ "grad_norm": 0.15354037284851074,
341
+ "learning_rate": 1.3620668117481471e-05,
342
+ "loss": 4.8556,
343
+ "step": 720
344
+ },
345
+ {
346
+ "epoch": 1.8969072164948453,
347
+ "grad_norm": 0.14049023389816284,
348
+ "learning_rate": 6.964873004985717e-06,
349
+ "loss": 4.8451,
350
+ "step": 736
351
+ },
352
+ {
353
+ "epoch": 1.9381443298969072,
354
+ "grad_norm": 0.13459183275699615,
355
+ "learning_rate": 2.511092455747932e-06,
356
+ "loss": 4.9148,
357
+ "step": 752
358
+ },
359
+ {
360
+ "epoch": 1.9793814432989691,
361
+ "grad_norm": 0.1388990879058838,
362
+ "learning_rate": 2.792181348726941e-07,
363
+ "loss": 4.8364,
364
+ "step": 768
365
+ },
366
+ {
367
+ "epoch": 2.0,
368
+ "eval_bleu": 0.083526303042587,
369
+ "eval_cap_loss": 1.723467393327005,
370
+ "eval_con_loss": 1.9934953146988583,
371
+ "eval_loss": 3.7169627135561916,
372
+ "step": 776
373
+ },
374
+ {
375
+ "epoch": 2.0,
376
+ "eval_bleu": 0.083526303042587,
377
+ "eval_cap_loss": 1.723467393327005,
378
+ "eval_con_loss": 1.9934953146988583,
379
+ "eval_loss": 3.7169627135561916,
380
+ "eval_runtime": 142.3121,
381
+ "eval_samples_per_second": 21.79,
382
+ "eval_steps_per_second": 2.726,
383
+ "step": 776
384
+ }
385
+ ],
386
+ "logging_steps": 16,
387
+ "max_steps": 776,
388
+ "num_input_tokens_seen": 0,
389
+ "num_train_epochs": 2,
390
+ "save_steps": 500,
391
+ "stateful_callbacks": {
392
+ "TrainerControl": {
393
+ "args": {
394
+ "should_epoch_stop": false,
395
+ "should_evaluate": false,
396
+ "should_log": false,
397
+ "should_save": true,
398
+ "should_training_stop": true
399
+ },
400
+ "attributes": {}
401
+ }
402
+ },
403
+ "total_flos": 0.0,
404
+ "train_batch_size": 32,
405
+ "trial_name": null,
406
+ "trial_params": null,
407
+ "tau_value": 2.5431
408
+ }