Training in progress, epoch 49
Browse files
.ipynb_checkpoints/config-checkpoint.json
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "MCG-NJU/videomae-base",
|
| 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": "drinking",
|
| 16 |
+
"1": "eating",
|
| 17 |
+
"2": "fastening_seat_belt",
|
| 18 |
+
"3": "fetching_an_object",
|
| 19 |
+
"4": "interacting_with_phone",
|
| 20 |
+
"5": "opening_bottle",
|
| 21 |
+
"6": "placing_an_object",
|
| 22 |
+
"7": "putting_on_jacket",
|
| 23 |
+
"8": "reading_magazine",
|
| 24 |
+
"9": "reading_newspaper",
|
| 25 |
+
"10": "sitting_still",
|
| 26 |
+
"11": "taking_off_jacket",
|
| 27 |
+
"12": "talking_on_phone",
|
| 28 |
+
"13": "using_multimedia_display",
|
| 29 |
+
"14": "working_on_laptop",
|
| 30 |
+
"15": "writing"
|
| 31 |
+
},
|
| 32 |
+
"image_size": 224,
|
| 33 |
+
"initializer_range": 0.02,
|
| 34 |
+
"intermediate_size": 3072,
|
| 35 |
+
"label2id": {
|
| 36 |
+
"drinking": 0,
|
| 37 |
+
"eating": 1,
|
| 38 |
+
"fastening_seat_belt": 2,
|
| 39 |
+
"fetching_an_object": 3,
|
| 40 |
+
"interacting_with_phone": 4,
|
| 41 |
+
"opening_bottle": 5,
|
| 42 |
+
"placing_an_object": 6,
|
| 43 |
+
"putting_on_jacket": 7,
|
| 44 |
+
"reading_magazine": 8,
|
| 45 |
+
"reading_newspaper": 9,
|
| 46 |
+
"sitting_still": 10,
|
| 47 |
+
"taking_off_jacket": 11,
|
| 48 |
+
"talking_on_phone": 12,
|
| 49 |
+
"using_multimedia_display": 13,
|
| 50 |
+
"working_on_laptop": 14,
|
| 51 |
+
"writing": 15
|
| 52 |
+
},
|
| 53 |
+
"layer_norm_eps": 1e-12,
|
| 54 |
+
"model_type": "videomae",
|
| 55 |
+
"norm_pix_loss": true,
|
| 56 |
+
"num_attention_heads": 12,
|
| 57 |
+
"num_channels": 3,
|
| 58 |
+
"num_frames": 16,
|
| 59 |
+
"num_hidden_layers": 12,
|
| 60 |
+
"patch_size": 16,
|
| 61 |
+
"problem_type": "single_label_classification",
|
| 62 |
+
"qkv_bias": true,
|
| 63 |
+
"torch_dtype": "float32",
|
| 64 |
+
"transformers_version": "4.46.3",
|
| 65 |
+
"tubelet_size": 2,
|
| 66 |
+
"use_mean_pooling": false
|
| 67 |
+
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 344980440
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2223a23f7a41594345a7ceb45744aad1adae328310510f80cebf021118f391e2
|
| 3 |
size 344980440
|