kevinoli commited on
Commit
3c1b477
1 Parent(s): 32bc6c3

Training in progress, step 6000, checkpoint

Browse files
checkpoint-6000/config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "openai/clip-vit-large-patch14-336",
3
+ "architectures": [
4
+ "CLIPModel"
5
+ ],
6
+ "initializer_factor": 1.0,
7
+ "logit_scale_init_value": 2.6592,
8
+ "model_type": "clip",
9
+ "projection_dim": 768,
10
+ "text_config": {
11
+ "dropout": 0.0,
12
+ "hidden_size": 768,
13
+ "intermediate_size": 3072,
14
+ "model_type": "clip_text_model",
15
+ "num_attention_heads": 12,
16
+ "projection_dim": 768
17
+ },
18
+ "torch_dtype": "float32",
19
+ "transformers_version": "4.45.0.dev0",
20
+ "vision_config": {
21
+ "dropout": 0.0,
22
+ "hidden_size": 1024,
23
+ "image_size": 336,
24
+ "intermediate_size": 4096,
25
+ "model_type": "clip_vision_model",
26
+ "num_attention_heads": 16,
27
+ "num_hidden_layers": 24,
28
+ "patch_size": 14,
29
+ "projection_dim": 768
30
+ }
31
+ }
checkpoint-6000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bd7ec14e27a16222acf0b78d41404007df36b059dd2bbc769aaafca07bdfc55d
3
+ size 1711848436
checkpoint-6000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85fbdde82096ee138fe1737b43bed7668337af2ba039ee79823014ac67e5589b
3
+ size 3424043887
checkpoint-6000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2540cc73259311480694ca6863e114df4e2c18e837dc0f7839bdb7cbd64a19a0
3
+ size 14503
checkpoint-6000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b414376c5fc3e40d23d8bc86633f1317793cc2e9aa4019b618112c2f042ca79f
3
+ size 623
checkpoint-6000/trainer_state.json ADDED
@@ -0,0 +1,213 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.6995787620544434,
3
+ "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l37-l/checkpoint-6000",
4
+ "epoch": 1.105379513633014,
5
+ "eval_steps": 500,
6
+ "global_step": 6000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.09211495946941783,
13
+ "grad_norm": 8.437943458557129,
14
+ "learning_rate": 2.907885040530582e-07,
15
+ "loss": 0.3017,
16
+ "step": 500
17
+ },
18
+ {
19
+ "epoch": 0.09211495946941783,
20
+ "eval_loss": 1.164457082748413,
21
+ "eval_runtime": 75.2697,
22
+ "eval_samples_per_second": 16.036,
23
+ "eval_steps_per_second": 2.006,
24
+ "step": 500
25
+ },
26
+ {
27
+ "epoch": 0.18422991893883567,
28
+ "grad_norm": 14.740598678588867,
29
+ "learning_rate": 2.815770081061164e-07,
30
+ "loss": 0.2899,
31
+ "step": 1000
32
+ },
33
+ {
34
+ "epoch": 0.18422991893883567,
35
+ "eval_loss": 1.01627779006958,
36
+ "eval_runtime": 76.4934,
37
+ "eval_samples_per_second": 15.779,
38
+ "eval_steps_per_second": 1.974,
39
+ "step": 1000
40
+ },
41
+ {
42
+ "epoch": 0.2763448784082535,
43
+ "grad_norm": 0.17031528055667877,
44
+ "learning_rate": 2.723655121591746e-07,
45
+ "loss": 0.273,
46
+ "step": 1500
47
+ },
48
+ {
49
+ "epoch": 0.2763448784082535,
50
+ "eval_loss": 0.9395213723182678,
51
+ "eval_runtime": 76.6256,
52
+ "eval_samples_per_second": 15.752,
53
+ "eval_steps_per_second": 1.971,
54
+ "step": 1500
55
+ },
56
+ {
57
+ "epoch": 0.36845983787767134,
58
+ "grad_norm": 0.0030106802005320787,
59
+ "learning_rate": 2.6315401621223287e-07,
60
+ "loss": 0.1818,
61
+ "step": 2000
62
+ },
63
+ {
64
+ "epoch": 0.36845983787767134,
65
+ "eval_loss": 0.8942967057228088,
66
+ "eval_runtime": 76.8089,
67
+ "eval_samples_per_second": 15.714,
68
+ "eval_steps_per_second": 1.966,
69
+ "step": 2000
70
+ },
71
+ {
72
+ "epoch": 0.46057479734708917,
73
+ "grad_norm": 474.4410705566406,
74
+ "learning_rate": 2.539425202652911e-07,
75
+ "loss": 0.2094,
76
+ "step": 2500
77
+ },
78
+ {
79
+ "epoch": 0.46057479734708917,
80
+ "eval_loss": 0.8330363631248474,
81
+ "eval_runtime": 76.9939,
82
+ "eval_samples_per_second": 15.677,
83
+ "eval_steps_per_second": 1.961,
84
+ "step": 2500
85
+ },
86
+ {
87
+ "epoch": 0.552689756816507,
88
+ "grad_norm": 314.3533935546875,
89
+ "learning_rate": 2.447310243183493e-07,
90
+ "loss": 0.1949,
91
+ "step": 3000
92
+ },
93
+ {
94
+ "epoch": 0.552689756816507,
95
+ "eval_loss": 0.7885717749595642,
96
+ "eval_runtime": 76.7534,
97
+ "eval_samples_per_second": 15.726,
98
+ "eval_steps_per_second": 1.967,
99
+ "step": 3000
100
+ },
101
+ {
102
+ "epoch": 0.6448047162859248,
103
+ "grad_norm": 7.678028106689453,
104
+ "learning_rate": 2.3551952837140753e-07,
105
+ "loss": 0.1335,
106
+ "step": 3500
107
+ },
108
+ {
109
+ "epoch": 0.6448047162859248,
110
+ "eval_loss": 0.7657251358032227,
111
+ "eval_runtime": 76.791,
112
+ "eval_samples_per_second": 15.718,
113
+ "eval_steps_per_second": 1.966,
114
+ "step": 3500
115
+ },
116
+ {
117
+ "epoch": 0.7369196757553427,
118
+ "grad_norm": 30.121461868286133,
119
+ "learning_rate": 2.263080324244657e-07,
120
+ "loss": 0.1541,
121
+ "step": 4000
122
+ },
123
+ {
124
+ "epoch": 0.7369196757553427,
125
+ "eval_loss": 0.7473344802856445,
126
+ "eval_runtime": 76.929,
127
+ "eval_samples_per_second": 15.69,
128
+ "eval_steps_per_second": 1.963,
129
+ "step": 4000
130
+ },
131
+ {
132
+ "epoch": 0.8290346352247605,
133
+ "grad_norm": 654.2782592773438,
134
+ "learning_rate": 2.1709653647752394e-07,
135
+ "loss": 0.1267,
136
+ "step": 4500
137
+ },
138
+ {
139
+ "epoch": 0.8290346352247605,
140
+ "eval_loss": 0.7376419305801392,
141
+ "eval_runtime": 75.2838,
142
+ "eval_samples_per_second": 16.033,
143
+ "eval_steps_per_second": 2.006,
144
+ "step": 4500
145
+ },
146
+ {
147
+ "epoch": 0.9211495946941783,
148
+ "grad_norm": 0.00021513362298719585,
149
+ "learning_rate": 2.0788504053058218e-07,
150
+ "loss": 0.1404,
151
+ "step": 5000
152
+ },
153
+ {
154
+ "epoch": 0.9211495946941783,
155
+ "eval_loss": 0.724165678024292,
156
+ "eval_runtime": 76.8393,
157
+ "eval_samples_per_second": 15.708,
158
+ "eval_steps_per_second": 1.965,
159
+ "step": 5000
160
+ },
161
+ {
162
+ "epoch": 1.013264554163596,
163
+ "grad_norm": 0.0016925306990742683,
164
+ "learning_rate": 1.9867354458364036e-07,
165
+ "loss": 0.1197,
166
+ "step": 5500
167
+ },
168
+ {
169
+ "epoch": 1.013264554163596,
170
+ "eval_loss": 0.7046866416931152,
171
+ "eval_runtime": 76.6635,
172
+ "eval_samples_per_second": 15.744,
173
+ "eval_steps_per_second": 1.97,
174
+ "step": 5500
175
+ },
176
+ {
177
+ "epoch": 1.105379513633014,
178
+ "grad_norm": 0.38834691047668457,
179
+ "learning_rate": 1.894620486366986e-07,
180
+ "loss": 0.0834,
181
+ "step": 6000
182
+ },
183
+ {
184
+ "epoch": 1.105379513633014,
185
+ "eval_loss": 0.6995787620544434,
186
+ "eval_runtime": 77.9523,
187
+ "eval_samples_per_second": 15.484,
188
+ "eval_steps_per_second": 1.937,
189
+ "step": 6000
190
+ }
191
+ ],
192
+ "logging_steps": 500,
193
+ "max_steps": 16284,
194
+ "num_input_tokens_seen": 0,
195
+ "num_train_epochs": 3,
196
+ "save_steps": 500,
197
+ "stateful_callbacks": {
198
+ "TrainerControl": {
199
+ "args": {
200
+ "should_epoch_stop": false,
201
+ "should_evaluate": false,
202
+ "should_log": false,
203
+ "should_save": true,
204
+ "should_training_stop": false
205
+ },
206
+ "attributes": {}
207
+ }
208
+ },
209
+ "total_flos": 2159269624262520.0,
210
+ "train_batch_size": 2,
211
+ "trial_name": null,
212
+ "trial_params": null
213
+ }
checkpoint-6000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76f7dfc201ccc0039551250d06ef1282a727ca0a51016f874c2eff3c580dbf8e
3
+ size 4847