kevinoli commited on
Commit
35ee158
·
verified ·
1 Parent(s): 26089b5

Training in progress, step 1500, checkpoint

Browse files
checkpoint-1500/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-1500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f71b7d16d35f6949a3cb164eff0be104b449928ac0073f1ff7106afb3c0ca41
3
+ size 1711848436
checkpoint-1500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f9a08584f648ea86293b3bc37047bfe09ecba72061f45edba1dd341099d7d25e
3
+ size 3424043887
checkpoint-1500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7fe9a60035ea5ac5318e14330cf266b3d4132df1d5a1ca3d538e23e0f5b9a935
3
+ size 14503
checkpoint-1500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1afba236d9bd01288d52a656be0f18e43a3e5d3b8d7dcb257421d5609a4e90ff
3
+ size 623
checkpoint-1500/trainer_state.json ADDED
@@ -0,0 +1,78 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.8726764917373657,
3
+ "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l57-l/checkpoint-1500",
4
+ "epoch": 0.2763448784082535,
5
+ "eval_steps": 500,
6
+ "global_step": 1500,
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": 3.0293657779693604,
14
+ "learning_rate": 4.884856300663227e-07,
15
+ "loss": 0.271,
16
+ "step": 500
17
+ },
18
+ {
19
+ "epoch": 0.09211495946941783,
20
+ "eval_loss": 1.069305419921875,
21
+ "eval_runtime": 75.1409,
22
+ "eval_samples_per_second": 16.063,
23
+ "eval_steps_per_second": 2.01,
24
+ "step": 500
25
+ },
26
+ {
27
+ "epoch": 0.18422991893883567,
28
+ "grad_norm": 7.3040852546691895,
29
+ "learning_rate": 4.769712601326456e-07,
30
+ "loss": 0.2493,
31
+ "step": 1000
32
+ },
33
+ {
34
+ "epoch": 0.18422991893883567,
35
+ "eval_loss": 0.9427077770233154,
36
+ "eval_runtime": 74.8926,
37
+ "eval_samples_per_second": 16.116,
38
+ "eval_steps_per_second": 2.016,
39
+ "step": 1000
40
+ },
41
+ {
42
+ "epoch": 0.2763448784082535,
43
+ "grad_norm": 0.020736657083034515,
44
+ "learning_rate": 4.654568901989683e-07,
45
+ "loss": 0.2348,
46
+ "step": 1500
47
+ },
48
+ {
49
+ "epoch": 0.2763448784082535,
50
+ "eval_loss": 0.8726764917373657,
51
+ "eval_runtime": 76.6621,
52
+ "eval_samples_per_second": 15.744,
53
+ "eval_steps_per_second": 1.97,
54
+ "step": 1500
55
+ }
56
+ ],
57
+ "logging_steps": 500,
58
+ "max_steps": 21712,
59
+ "num_input_tokens_seen": 0,
60
+ "num_train_epochs": 4,
61
+ "save_steps": 500,
62
+ "stateful_callbacks": {
63
+ "TrainerControl": {
64
+ "args": {
65
+ "should_epoch_stop": false,
66
+ "should_evaluate": false,
67
+ "should_log": false,
68
+ "should_save": true,
69
+ "should_training_stop": false
70
+ },
71
+ "attributes": {}
72
+ }
73
+ },
74
+ "total_flos": 540357043592520.0,
75
+ "train_batch_size": 2,
76
+ "trial_name": null,
77
+ "trial_params": null
78
+ }
checkpoint-1500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf76416e4a857628175385db7f617e54451cd7ffb62c1f9590ee9f30268f8c5b
3
+ size 4847