Commit
·
2d97160
1
Parent(s):
d5de465
Upload tiny models for TvltForPreTraining
Browse files- config.json +50 -0
- preprocessor_config.json +31 -0
- pytorch_model.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"TvltForPreTraining"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.0,
|
| 6 |
+
"audio_mask_ratio": 0.15,
|
| 7 |
+
"audio_mask_type": "frame-level",
|
| 8 |
+
"audio_patch_size": [
|
| 9 |
+
2,
|
| 10 |
+
2
|
| 11 |
+
],
|
| 12 |
+
"decoder_hidden_size": 64,
|
| 13 |
+
"decoder_intermediate_size": 128,
|
| 14 |
+
"decoder_num_attention_heads": 4,
|
| 15 |
+
"decoder_num_hidden_layers": 2,
|
| 16 |
+
"frequency_length": 16,
|
| 17 |
+
"hidden_act": "gelu",
|
| 18 |
+
"hidden_dropout_prob": 0.0,
|
| 19 |
+
"hidden_size": 128,
|
| 20 |
+
"id2label": {
|
| 21 |
+
"0": "LABEL_0"
|
| 22 |
+
},
|
| 23 |
+
"image_mask_ratio": 0.75,
|
| 24 |
+
"image_patch_size": [
|
| 25 |
+
2,
|
| 26 |
+
2
|
| 27 |
+
],
|
| 28 |
+
"image_size": 32,
|
| 29 |
+
"initializer_range": 0.02,
|
| 30 |
+
"intermediate_size": 128,
|
| 31 |
+
"label2id": {
|
| 32 |
+
"LABEL_0": 0
|
| 33 |
+
},
|
| 34 |
+
"layer_norm_eps": 1e-12,
|
| 35 |
+
"loss_type": "classification",
|
| 36 |
+
"model_type": "tvlt",
|
| 37 |
+
"num_attention_heads": 4,
|
| 38 |
+
"num_audio_channels": 1,
|
| 39 |
+
"num_frames": 2,
|
| 40 |
+
"num_hidden_layers": 12,
|
| 41 |
+
"num_image_channels": 3,
|
| 42 |
+
"pixel_mask_ratio": 0.75,
|
| 43 |
+
"qkv_bias": true,
|
| 44 |
+
"spectrogram_length": 32,
|
| 45 |
+
"task_mae": true,
|
| 46 |
+
"task_matching": true,
|
| 47 |
+
"torch_dtype": "float32",
|
| 48 |
+
"transformers_version": "4.28.0.dev0",
|
| 49 |
+
"use_mean_pooling": true
|
| 50 |
+
}
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"crop_size": {
|
| 3 |
+
"height": 32,
|
| 4 |
+
"width": 32
|
| 5 |
+
},
|
| 6 |
+
"do_center_crop": true,
|
| 7 |
+
"do_normalize": true,
|
| 8 |
+
"do_rescale": true,
|
| 9 |
+
"do_resize": true,
|
| 10 |
+
"image_mean": [
|
| 11 |
+
0.5,
|
| 12 |
+
0.5,
|
| 13 |
+
0.5
|
| 14 |
+
],
|
| 15 |
+
"image_processor_type": "TvltImageProcessor",
|
| 16 |
+
"image_std": [
|
| 17 |
+
0.5,
|
| 18 |
+
0.5,
|
| 19 |
+
0.5
|
| 20 |
+
],
|
| 21 |
+
"num_frames": 8,
|
| 22 |
+
"patch_size": [
|
| 23 |
+
16,
|
| 24 |
+
16
|
| 25 |
+
],
|
| 26 |
+
"resample": 2,
|
| 27 |
+
"rescale_factor": 0.00392156862745098,
|
| 28 |
+
"size": {
|
| 29 |
+
"shortest_edge": 32
|
| 30 |
+
}
|
| 31 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:09b3bff4818e9fbb56880a323ec093f7de5c0fc5fad40b67689950a50e7df97b
|
| 3 |
+
size 5462921
|