speedinghzl commited on
Commit
e3ac4c9
·
verified ·
1 Parent(s): 2d68eec

Upload folder using huggingface_hub

Browse files
vit_l16_s12.8B_bs32k/checkpoints/results.jsonl ADDED
@@ -0,0 +1,138 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"imagenet-zeroshot-val-top1": 0.38514, "imagenet-zeroshot-val-top5": 0.66736}
2
+ {"imagenet-zeroshot-val-top1": 0.51966, "imagenet-zeroshot-val-top5": 0.79978}
3
+ {"imagenet-zeroshot-val-top1": 0.55704, "imagenet-zeroshot-val-top5": 0.8318}
4
+ {"imagenet-zeroshot-val-top1": 0.57938, "imagenet-zeroshot-val-top5": 0.85212}
5
+ {"imagenet-zeroshot-val-top1": 0.59082, "imagenet-zeroshot-val-top5": 0.85864}
6
+ {"imagenet-zeroshot-val-top1": 0.59858, "imagenet-zeroshot-val-top5": 0.85784}
7
+ {"imagenet-zeroshot-val-top1": 0.60132, "imagenet-zeroshot-val-top5": 0.86282}
8
+ {"imagenet-zeroshot-val-top1": 0.61392, "imagenet-zeroshot-val-top5": 0.86964}
9
+ {"imagenet-zeroshot-val-top1": 0.61998, "imagenet-zeroshot-val-top5": 0.8735}
10
+ {"imagenet-zeroshot-val-top1": 0.61822, "imagenet-zeroshot-val-top5": 0.87226}
11
+ {"imagenet-zeroshot-val-top1": 0.6203, "imagenet-zeroshot-val-top5": 0.87188}
12
+ {"imagenet-zeroshot-val-top1": 0.6233, "imagenet-zeroshot-val-top5": 0.8822}
13
+ {"imagenet-zeroshot-val-top1": 0.6238, "imagenet-zeroshot-val-top5": 0.87542}
14
+ {"imagenet-zeroshot-val-top1": 0.62626, "imagenet-zeroshot-val-top5": 0.87722}
15
+ {"imagenet-zeroshot-val-top1": 0.6237, "imagenet-zeroshot-val-top5": 0.87672}
16
+ {"imagenet-zeroshot-val-top1": 0.6258, "imagenet-zeroshot-val-top5": 0.88038}
17
+ {"imagenet-zeroshot-val-top1": 0.63064, "imagenet-zeroshot-val-top5": 0.88096}
18
+ {"imagenet-zeroshot-val-top1": 0.62754, "imagenet-zeroshot-val-top5": 0.8807}
19
+ {"imagenet-zeroshot-val-top1": 0.63408, "imagenet-zeroshot-val-top5": 0.88274}
20
+ {"imagenet-zeroshot-val-top1": 0.63202, "imagenet-zeroshot-val-top5": 0.88204}
21
+ {"imagenet-zeroshot-val-top1": 0.63546, "imagenet-zeroshot-val-top5": 0.8832}
22
+ {"imagenet-zeroshot-val-top1": 0.63568, "imagenet-zeroshot-val-top5": 0.88552}
23
+ {"imagenet-zeroshot-val-top1": 0.6413, "imagenet-zeroshot-val-top5": 0.88568}
24
+ {"imagenet-zeroshot-val-top1": 0.6398, "imagenet-zeroshot-val-top5": 0.88396}
25
+ {"imagenet-zeroshot-val-top1": 0.63846, "imagenet-zeroshot-val-top5": 0.88688}
26
+ {"imagenet-zeroshot-val-top1": 0.63622, "imagenet-zeroshot-val-top5": 0.8853}
27
+ {"imagenet-zeroshot-val-top1": 0.63932, "imagenet-zeroshot-val-top5": 0.88702}
28
+ {"imagenet-zeroshot-val-top1": 0.63362, "imagenet-zeroshot-val-top5": 0.88316}
29
+ {"imagenet-zeroshot-val-top1": 0.6357, "imagenet-zeroshot-val-top5": 0.88386}
30
+ {"imagenet-zeroshot-val-top1": 0.64356, "imagenet-zeroshot-val-top5": 0.89072}
31
+ {"imagenet-zeroshot-val-top1": 0.64958, "imagenet-zeroshot-val-top5": 0.89226}
32
+ {"imagenet-zeroshot-val-top1": 0.64664, "imagenet-zeroshot-val-top5": 0.89066}
33
+ {"imagenet-zeroshot-val-top1": 0.65204, "imagenet-zeroshot-val-top5": 0.89364}
34
+ {"imagenet-zeroshot-val-top1": 0.64392, "imagenet-zeroshot-val-top5": 0.89074}
35
+ {"imagenet-zeroshot-val-top1": 0.64668, "imagenet-zeroshot-val-top5": 0.89234}
36
+ {"imagenet-zeroshot-val-top1": 0.64746, "imagenet-zeroshot-val-top5": 0.89482}
37
+ {"imagenet-zeroshot-val-top1": 0.64618, "imagenet-zeroshot-val-top5": 0.88832}
38
+ {"imagenet-zeroshot-val-top1": 0.65402, "imagenet-zeroshot-val-top5": 0.89512}
39
+ {"imagenet-zeroshot-val-top1": 0.65366, "imagenet-zeroshot-val-top5": 0.8948}
40
+ {"imagenet-zeroshot-val-top1": 0.65572, "imagenet-zeroshot-val-top5": 0.89768}
41
+ {"imagenet-zeroshot-val-top1": 0.6533, "imagenet-zeroshot-val-top5": 0.89418}
42
+ {"imagenet-zeroshot-val-top1": 0.65544, "imagenet-zeroshot-val-top5": 0.89522}
43
+ {"imagenet-zeroshot-val-top1": 0.65652, "imagenet-zeroshot-val-top5": 0.89776}
44
+ {"imagenet-zeroshot-val-top1": 0.65968, "imagenet-zeroshot-val-top5": 0.90074}
45
+ {"imagenet-zeroshot-val-top1": 0.6633, "imagenet-zeroshot-val-top5": 0.89684}
46
+ {"imagenet-zeroshot-val-top1": 0.65636, "imagenet-zeroshot-val-top5": 0.89488}
47
+ {"imagenet-zeroshot-val-top1": 0.66358, "imagenet-zeroshot-val-top5": 0.89848}
48
+ {"imagenet-zeroshot-val-top1": 0.66048, "imagenet-zeroshot-val-top5": 0.89618}
49
+ {"imagenet-zeroshot-val-top1": 0.66256, "imagenet-zeroshot-val-top5": 0.9016}
50
+ {"imagenet-zeroshot-val-top1": 0.66094, "imagenet-zeroshot-val-top5": 0.89918}
51
+ {"imagenet-zeroshot-val-top1": 0.66014, "imagenet-zeroshot-val-top5": 0.89954}
52
+ {"imagenet-zeroshot-val-top1": 0.66114, "imagenet-zeroshot-val-top5": 0.9003}
53
+ {"imagenet-zeroshot-val-top1": 0.66052, "imagenet-zeroshot-val-top5": 0.89898}
54
+ {"imagenet-zeroshot-val-top1": 0.6687, "imagenet-zeroshot-val-top5": 0.90222}
55
+ {"imagenet-zeroshot-val-top1": 0.66266, "imagenet-zeroshot-val-top5": 0.9004}
56
+ {"imagenet-zeroshot-val-top1": 0.67262, "imagenet-zeroshot-val-top5": 0.90366}
57
+ {"imagenet-zeroshot-val-top1": 0.67186, "imagenet-zeroshot-val-top5": 0.90428}
58
+ {"imagenet-zeroshot-val-top1": 0.66958, "imagenet-zeroshot-val-top5": 0.904}
59
+ {"imagenet-zeroshot-val-top1": 0.6683, "imagenet-zeroshot-val-top5": 0.90266}
60
+ {"imagenet-zeroshot-val-top1": 0.67044, "imagenet-zeroshot-val-top5": 0.9049}
61
+ {"imagenet-zeroshot-val-top1": 0.67248, "imagenet-zeroshot-val-top5": 0.90566}
62
+ {"imagenet-zeroshot-val-top1": 0.673, "imagenet-zeroshot-val-top5": 0.90948}
63
+ {"imagenet-zeroshot-val-top1": 0.67308, "imagenet-zeroshot-val-top5": 0.90764}
64
+ {"imagenet-zeroshot-val-top1": 0.67212, "imagenet-zeroshot-val-top5": 0.90884}
65
+ {"imagenet-zeroshot-val-top1": 0.6793, "imagenet-zeroshot-val-top5": 0.90964}
66
+ {"imagenet-zeroshot-val-top1": 0.67714, "imagenet-zeroshot-val-top5": 0.90916}
67
+ {"imagenet-zeroshot-val-top1": 0.6848, "imagenet-zeroshot-val-top5": 0.90964}
68
+ {"imagenet-zeroshot-val-top1": 0.68294, "imagenet-zeroshot-val-top5": 0.90966}
69
+ {"imagenet-zeroshot-val-top1": 0.6829, "imagenet-zeroshot-val-top5": 0.91014}
70
+ {"imagenet-zeroshot-val-top1": 0.68554, "imagenet-zeroshot-val-top5": 0.91246}
71
+ {"imagenet-zeroshot-val-top1": 0.68724, "imagenet-zeroshot-val-top5": 0.91394}
72
+ {"imagenet-zeroshot-val-top1": 0.68794, "imagenet-zeroshot-val-top5": 0.91486}
73
+ {"imagenet-zeroshot-val-top1": 0.6846, "imagenet-zeroshot-val-top5": 0.91428}
74
+ {"imagenet-zeroshot-val-top1": 0.69198, "imagenet-zeroshot-val-top5": 0.91662}
75
+ {"imagenet-zeroshot-val-top1": 0.68996, "imagenet-zeroshot-val-top5": 0.91452}
76
+ {"imagenet-zeroshot-val-top1": 0.69498, "imagenet-zeroshot-val-top5": 0.91594}
77
+ {"imagenet-zeroshot-val-top1": 0.69732, "imagenet-zeroshot-val-top5": 0.91694}
78
+ {"imagenet-zeroshot-val-top1": 0.69622, "imagenet-zeroshot-val-top5": 0.91762}
79
+ {"imagenet-zeroshot-val-top1": 0.69038, "imagenet-zeroshot-val-top5": 0.91544}
80
+ {"imagenet-zeroshot-val-top1": 0.69838, "imagenet-zeroshot-val-top5": 0.9199}
81
+ {"imagenet-zeroshot-val-top1": 0.69918, "imagenet-zeroshot-val-top5": 0.9198}
82
+ {"imagenet-zeroshot-val-top1": 0.7029, "imagenet-zeroshot-val-top5": 0.9234}
83
+ {"imagenet-zeroshot-val-top1": 0.70086, "imagenet-zeroshot-val-top5": 0.9207}
84
+ {"imagenet-zeroshot-val-top1": 0.70252, "imagenet-zeroshot-val-top5": 0.92022}
85
+ {"imagenet-zeroshot-val-top1": 0.70282, "imagenet-zeroshot-val-top5": 0.92042}
86
+ {"imagenet-zeroshot-val-top1": 0.70902, "imagenet-zeroshot-val-top5": 0.9229}
87
+ {"imagenet-zeroshot-val-top1": 0.7072, "imagenet-zeroshot-val-top5": 0.9222}
88
+ {"imagenet-zeroshot-val-top1": 0.7102, "imagenet-zeroshot-val-top5": 0.92576}
89
+ {"imagenet-zeroshot-val-top1": 0.7109, "imagenet-zeroshot-val-top5": 0.92614}
90
+ {"imagenet-zeroshot-val-top1": 0.71462, "imagenet-zeroshot-val-top5": 0.92622}
91
+ {"imagenet-zeroshot-val-top1": 0.71544, "imagenet-zeroshot-val-top5": 0.92566}
92
+ {"imagenet-zeroshot-val-top1": 0.71684, "imagenet-zeroshot-val-top5": 0.92566}
93
+ {"imagenet-zeroshot-val-top1": 0.71872, "imagenet-zeroshot-val-top5": 0.92782}
94
+ {"imagenet-zeroshot-val-top1": 0.72132, "imagenet-zeroshot-val-top5": 0.92918}
95
+ {"imagenet-zeroshot-val-top1": 0.72224, "imagenet-zeroshot-val-top5": 0.92878}
96
+ {"imagenet-zeroshot-val-top1": 0.72402, "imagenet-zeroshot-val-top5": 0.93328}
97
+ {"imagenet-zeroshot-val-top1": 0.72424, "imagenet-zeroshot-val-top5": 0.93142}
98
+ {"imagenet-zeroshot-val-top1": 0.72514, "imagenet-zeroshot-val-top5": 0.93266}
99
+ {"imagenet-zeroshot-val-top1": 0.72674, "imagenet-zeroshot-val-top5": 0.93192}
100
+ {"imagenet-zeroshot-val-top1": 0.72948, "imagenet-zeroshot-val-top5": 0.93256}
101
+ {"imagenet-zeroshot-val-top1": 0.73438, "imagenet-zeroshot-val-top5": 0.93586}
102
+ {"imagenet-zeroshot-val-top1": 0.73058, "imagenet-zeroshot-val-top5": 0.93356}
103
+ {"imagenet-zeroshot-val-top1": 0.72942, "imagenet-zeroshot-val-top5": 0.93368}
104
+ {"imagenet-zeroshot-val-top1": 0.7333, "imagenet-zeroshot-val-top5": 0.93544}
105
+ {"imagenet-zeroshot-val-top1": 0.7348, "imagenet-zeroshot-val-top5": 0.93778}
106
+ {"imagenet-zeroshot-val-top1": 0.73806, "imagenet-zeroshot-val-top5": 0.93568}
107
+ {"imagenet-zeroshot-val-top1": 0.74212, "imagenet-zeroshot-val-top5": 0.9385}
108
+ {"imagenet-zeroshot-val-top1": 0.7456, "imagenet-zeroshot-val-top5": 0.9394}
109
+ {"imagenet-zeroshot-val-top1": 0.74688, "imagenet-zeroshot-val-top5": 0.94096}
110
+ {"imagenet-zeroshot-val-top1": 0.74614, "imagenet-zeroshot-val-top5": 0.94026}
111
+ {"imagenet-zeroshot-val-top1": 0.7497, "imagenet-zeroshot-val-top5": 0.94194}
112
+ {"imagenet-zeroshot-val-top1": 0.74968, "imagenet-zeroshot-val-top5": 0.94184}
113
+ {"imagenet-zeroshot-val-top1": 0.75366, "imagenet-zeroshot-val-top5": 0.94118}
114
+ {"imagenet-zeroshot-val-top1": 0.7537, "imagenet-zeroshot-val-top5": 0.94224}
115
+ {"imagenet-zeroshot-val-top1": 0.75544, "imagenet-zeroshot-val-top5": 0.94446}
116
+ {"imagenet-zeroshot-val-top1": 0.7621, "imagenet-zeroshot-val-top5": 0.94586}
117
+ {"imagenet-zeroshot-val-top1": 0.76256, "imagenet-zeroshot-val-top5": 0.94664}
118
+ {"imagenet-zeroshot-val-top1": 0.76328, "imagenet-zeroshot-val-top5": 0.9471}
119
+ {"imagenet-zeroshot-val-top1": 0.7675, "imagenet-zeroshot-val-top5": 0.94804}
120
+ {"imagenet-zeroshot-val-top1": 0.76786, "imagenet-zeroshot-val-top5": 0.94834}
121
+ {"imagenet-zeroshot-val-top1": 0.77178, "imagenet-zeroshot-val-top5": 0.9487}
122
+ {"imagenet-zeroshot-val-top1": 0.77388, "imagenet-zeroshot-val-top5": 0.95058}
123
+ {"imagenet-zeroshot-val-top1": 0.77292, "imagenet-zeroshot-val-top5": 0.95064}
124
+ {"imagenet-zeroshot-val-top1": 0.7766, "imagenet-zeroshot-val-top5": 0.95132}
125
+ {"imagenet-zeroshot-val-top1": 0.77742, "imagenet-zeroshot-val-top5": 0.9527}
126
+ {"imagenet-zeroshot-val-top1": 0.77804, "imagenet-zeroshot-val-top5": 0.95164}
127
+ {"imagenet-zeroshot-val-top1": 0.78128, "imagenet-zeroshot-val-top5": 0.95372}
128
+ {"imagenet-zeroshot-val-top1": 0.78168, "imagenet-zeroshot-val-top5": 0.9532}
129
+ {"imagenet-zeroshot-val-top1": 0.78332, "imagenet-zeroshot-val-top5": 0.95434}
130
+ {"imagenet-zeroshot-val-top1": 0.78666, "imagenet-zeroshot-val-top5": 0.95398}
131
+ {"imagenet-zeroshot-val-top1": 0.78688, "imagenet-zeroshot-val-top5": 0.9541}
132
+ {"imagenet-zeroshot-val-top1": 0.78752, "imagenet-zeroshot-val-top5": 0.95482}
133
+ {"imagenet-zeroshot-val-top1": 0.78816, "imagenet-zeroshot-val-top5": 0.95506}
134
+ {"imagenet-zeroshot-val-top1": 0.78988, "imagenet-zeroshot-val-top5": 0.95546}
135
+ {"imagenet-zeroshot-val-top1": 0.78998, "imagenet-zeroshot-val-top5": 0.95544}
136
+ {"imagenet-zeroshot-val-top1": 0.78978, "imagenet-zeroshot-val-top5": 0.95522}
137
+ {"imagenet-zeroshot-val-top1": 0.79008, "imagenet-zeroshot-val-top5": 0.95516}
138
+ {"imagenet-zeroshot-val-top1": 0.79008, "imagenet-zeroshot-val-top5": 0.95516}
vit_l16_s12.8B_bs32k/checkpoints/step_390656.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a94be4af60168b9683747c04fb731a6ffec864ff9f69183fca889eef09bb9840
3
+ size 5133443602
vit_l16_s12.8B_bs32k/out.log ADDED
The diff for this file is too large to render. See raw diff
 
vit_l16_s12.8B_bs32k/params.txt ADDED
@@ -0,0 +1,106 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ accum_freq: 1
2
+ aug_cfg: {'scale': [0.4, 1.0], 'color_jitter': [0.32, 0.32, 0.32, 0.08], 'color_jitter_prob': 0.8, 'gray_scale_prob': 0.2}
3
+ batch_size: 512
4
+ beta1: 0.9
5
+ beta2: 0.95
6
+ bin_balanced_sampling_expand: None
7
+ bin_balanced_sampling_nbins: None
8
+ checkpoint_path: ./logs/vit_l16_s12.8B_bs32k/checkpoints
9
+ coca_caption_loss_weight: 2.0
10
+ coca_contrastive_loss_weight: 1.0
11
+ copy_codebase: False
12
+ csv_caption_key: title
13
+ csv_img_key: filepath
14
+ csv_separator:
15
+ dataset_resampled: False
16
+ dataset_type: webdataset
17
+ ddp_static_graph: True
18
+ debug: False
19
+ delete_prev_step_ckpt: True
20
+ delete_previous_checkpoint: False
21
+ device: cuda:0
22
+ dist_backend: nccl
23
+ dist_url: env://
24
+ distill: False
25
+ distill_model: None
26
+ distill_pretrained: None
27
+ distributed: True
28
+ epochs: 10
29
+ epochs_cooldown: None
30
+ eps: 1e-06
31
+ flash_attn: False
32
+ force_custom_text: False
33
+ force_image_size: 224
34
+ force_patch_dropout: None
35
+ force_quick_gelu: False
36
+ gather_with_grad: True
37
+ global_batch_size: 32768
38
+ grad_checkpointing: True
39
+ grad_clip_norm: None
40
+ horovod: False
41
+ image_mean: None
42
+ image_std: None
43
+ imagenet_v2: None
44
+ imagenet_val: /mnt/bn/zilongdata-us/dataset/ILSVRC/Data/CLS-LOC/val
45
+ local_loss: True
46
+ local_rank: 0
47
+ lock_image: False
48
+ lock_image_freeze_bn_stats: False
49
+ lock_image_unlocked_groups: 0
50
+ lock_text: False
51
+ lock_text_freeze_layer_norm: False
52
+ lock_text_unlocked_layers: 0
53
+ log_every_n_steps: 64
54
+ log_level: 20
55
+ log_local: False
56
+ log_path: ./logs/vit_l16_s12.8B_bs32k/out.log
57
+ logs: ./logs
58
+ lr: 0.001
59
+ lr_cooldown_end: 0.0
60
+ lr_cooldown_power: 1.0
61
+ lr_multiplier_text: None
62
+ lr_scheduler: cosine
63
+ model: ViT-L-16
64
+ name: vit_l16_s12.8B_bs32k
65
+ no_set_device_rank: False
66
+ precision: amp_bfloat16
67
+ pretrained:
68
+ pretrained_image: False
69
+ pretrained_optim_scaler: False
70
+ pretrained_text:
71
+ rank: 0
72
+ remote_sync: None
73
+ remote_sync_frequency: 300
74
+ remote_sync_protocol: s3
75
+ report_to: tensorboard
76
+ resume: None
77
+ save_every_n_steps: 3052
78
+ save_frequency: 1
79
+ save_most_recent: False
80
+ seed: 0
81
+ skip_scheduler: False
82
+ tensorboard: True
83
+ tensorboard_path: ./logs/vit_l16_s12.8B_bs32k/tensorboard
84
+ torchcompile: False
85
+ torchscript: False
86
+ trace: False
87
+ train_data: /mnt/bn/zilongdata-us/dataset/datacomp-1b-webdataset/{000000..140146}.tar
88
+ train_data_upsampling_factors: None
89
+ train_num_samples: 1280000000
90
+ unlock_text_proj: False
91
+ unset_text_grad_checkpointing: False
92
+ use_bn_sync: False
93
+ use_bnb_linear: None
94
+ val_data: None
95
+ val_frequency: 1
96
+ val_num_samples: None
97
+ val_steps: 3052
98
+ wandb: False
99
+ wandb_notes:
100
+ wandb_project_name: open-clip
101
+ warmup: 1563
102
+ wd: 0.2
103
+ workers: 6
104
+ world_size: 64
105
+ zeroshot_frequency: 1
106
+ zeroshot_steps: 3052
vit_l16_s12.8B_bs32k/tensorboard/events.out.tfevents.1713911109.n107-071-092.2941.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:00b658d7140d5c2fcd19ef5020c3d6a8bf17a2fde3a0f1f401f8922090d03e00
3
+ size 2815108