Commit
•
eb2ee8f
1
Parent(s):
2acdee1
Training in progress, epoch 0
Browse files- config.json +63 -0
- model.safetensors +3 -0
- preprocessor_config.json +26 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "MCG-NJU/videomae-base-ssv2",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"VideoMAEForVideoClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"decoder_hidden_size": 384,
|
| 8 |
+
"decoder_intermediate_size": 1536,
|
| 9 |
+
"decoder_num_attention_heads": 6,
|
| 10 |
+
"decoder_num_hidden_layers": 4,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.0,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"id2label": {
|
| 15 |
+
"0": "climb",
|
| 16 |
+
"1": "crawl",
|
| 17 |
+
"2": "grasp",
|
| 18 |
+
"3": "hiding",
|
| 19 |
+
"4": "jump",
|
| 20 |
+
"5": "pick_up",
|
| 21 |
+
"6": "pull",
|
| 22 |
+
"7": "push",
|
| 23 |
+
"8": "put_down",
|
| 24 |
+
"9": "roll",
|
| 25 |
+
"10": "running",
|
| 26 |
+
"11": "slide",
|
| 27 |
+
"12": "walking",
|
| 28 |
+
"13": "zibaroon"
|
| 29 |
+
},
|
| 30 |
+
"image_size": 224,
|
| 31 |
+
"initializer_range": 0.02,
|
| 32 |
+
"intermediate_size": 3072,
|
| 33 |
+
"label2id": {
|
| 34 |
+
"climb": 0,
|
| 35 |
+
"crawl": 1,
|
| 36 |
+
"grasp": 2,
|
| 37 |
+
"hiding": 3,
|
| 38 |
+
"jump": 4,
|
| 39 |
+
"pick_up": 5,
|
| 40 |
+
"pull": 6,
|
| 41 |
+
"push": 7,
|
| 42 |
+
"put_down": 8,
|
| 43 |
+
"roll": 9,
|
| 44 |
+
"running": 10,
|
| 45 |
+
"slide": 11,
|
| 46 |
+
"walking": 12,
|
| 47 |
+
"zibaroon": 13
|
| 48 |
+
},
|
| 49 |
+
"layer_norm_eps": 1e-12,
|
| 50 |
+
"model_type": "videomae",
|
| 51 |
+
"norm_pix_loss": true,
|
| 52 |
+
"num_attention_heads": 12,
|
| 53 |
+
"num_channels": 3,
|
| 54 |
+
"num_frames": 16,
|
| 55 |
+
"num_hidden_layers": 12,
|
| 56 |
+
"patch_size": 16,
|
| 57 |
+
"problem_type": "single_label_classification",
|
| 58 |
+
"qkv_bias": true,
|
| 59 |
+
"torch_dtype": "float32",
|
| 60 |
+
"transformers_version": "4.43.3",
|
| 61 |
+
"tubelet_size": 2,
|
| 62 |
+
"use_mean_pooling": false
|
| 63 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3888a2eb018c4c7d6cacf1fdc9066fd517039fc5dd95e8b56e9038dee6c47808
|
| 3 |
+
size 344974288
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"crop_size": {
|
| 3 |
+
"height": 224,
|
| 4 |
+
"width": 224
|
| 5 |
+
},
|
| 6 |
+
"do_center_crop": true,
|
| 7 |
+
"do_normalize": true,
|
| 8 |
+
"do_rescale": true,
|
| 9 |
+
"do_resize": true,
|
| 10 |
+
"image_mean": [
|
| 11 |
+
0.485,
|
| 12 |
+
0.456,
|
| 13 |
+
0.406
|
| 14 |
+
],
|
| 15 |
+
"image_processor_type": "VideoMAEImageProcessor",
|
| 16 |
+
"image_std": [
|
| 17 |
+
0.229,
|
| 18 |
+
0.224,
|
| 19 |
+
0.225
|
| 20 |
+
],
|
| 21 |
+
"resample": 2,
|
| 22 |
+
"rescale_factor": 0.00392156862745098,
|
| 23 |
+
"size": {
|
| 24 |
+
"shortest_edge": 224
|
| 25 |
+
}
|
| 26 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f4902b80382fd912b465dcddf8a981f8bc938b1adefc37d1a20f80f30e799de7
|
| 3 |
+
size 5240
|