Training in progress, epoch 14
Browse files
config.json
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "MCG-NJU/videomae-base-finetuned-kinetics",
|
| 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": "BAND",
|
| 16 |
+
"1": "BOX",
|
| 17 |
+
"2": "DIRECTION",
|
| 18 |
+
"3": "ELECTRIC_CAR",
|
| 19 |
+
"4": "FACILITY",
|
| 20 |
+
"5": "KANBAN",
|
| 21 |
+
"6": "NEUTRAL",
|
| 22 |
+
"7": "PALLET",
|
| 23 |
+
"8": "PARTS",
|
| 24 |
+
"9": "SHELF",
|
| 25 |
+
"10": "WAGON"
|
| 26 |
+
},
|
| 27 |
+
"image_size": 224,
|
| 28 |
+
"initializer_range": 0.02,
|
| 29 |
+
"intermediate_size": 3072,
|
| 30 |
+
"label2id": {
|
| 31 |
+
"BAND": 0,
|
| 32 |
+
"BOX": 1,
|
| 33 |
+
"DIRECTION": 2,
|
| 34 |
+
"ELECTRIC_CAR": 3,
|
| 35 |
+
"FACILITY": 4,
|
| 36 |
+
"KANBAN": 5,
|
| 37 |
+
"NEUTRAL": 6,
|
| 38 |
+
"PALLET": 7,
|
| 39 |
+
"PARTS": 8,
|
| 40 |
+
"SHELF": 9,
|
| 41 |
+
"WAGON": 10
|
| 42 |
+
},
|
| 43 |
+
"layer_norm_eps": 1e-12,
|
| 44 |
+
"model_type": "videomae",
|
| 45 |
+
"norm_pix_loss": false,
|
| 46 |
+
"num_attention_heads": 12,
|
| 47 |
+
"num_channels": 3,
|
| 48 |
+
"num_frames": 16,
|
| 49 |
+
"num_hidden_layers": 12,
|
| 50 |
+
"patch_size": 16,
|
| 51 |
+
"problem_type": "single_label_classification",
|
| 52 |
+
"qkv_bias": true,
|
| 53 |
+
"torch_dtype": "float32",
|
| 54 |
+
"transformers_version": "4.38.1",
|
| 55 |
+
"tubelet_size": 2,
|
| 56 |
+
"use_mean_pooling": true
|
| 57 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:798d17fd21adac399641511a00fe37ba83c1d8a11ac5a35a0e7e878a33e79dc2
|
| 3 |
+
size 344965020
|
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 |
+
}
|
runs/Aug13_15-36-46_WIN-645TP8FEMLV/events.out.tfevents.1723531010.WIN-645TP8FEMLV.22720.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4355e248ab798aafc123ac747661e9894907a8f99c67c07b25e1122d5a942cf9
|
| 3 |
+
size 52138
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:28556506204a847c00bda7437e096310d7491f195b55898586275853eb03b555
|
| 3 |
+
size 4475
|