Add config from convert_d_fine_original_pytorch_checkpoint_to_hf.py
Browse files- config.json +31 -46
config.json
CHANGED
|
@@ -1,10 +1,8 @@
|
|
| 1 |
{
|
|
|
|
| 2 |
"activation_dropout": 0.0,
|
| 3 |
"activation_function": "silu",
|
| 4 |
"anchor_image_size": null,
|
| 5 |
-
"architectures": [
|
| 6 |
-
"DFineForObjectDetection"
|
| 7 |
-
],
|
| 8 |
"attention_dropout": 0.0,
|
| 9 |
"auxiliary_loss": true,
|
| 10 |
"backbone": null,
|
|
@@ -28,47 +26,35 @@
|
|
| 28 |
3,
|
| 29 |
4
|
| 30 |
],
|
| 31 |
-
"
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
|
| 39 |
-
|
| 40 |
-
|
| 41 |
-
|
| 42 |
-
|
| 43 |
-
|
| 44 |
-
|
| 45 |
-
|
| 46 |
-
|
| 47 |
-
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
|
| 51 |
-
|
| 52 |
-
|
| 53 |
-
|
| 54 |
-
|
| 55 |
-
|
| 56 |
-
|
| 57 |
-
|
| 58 |
-
|
| 59 |
-
|
| 60 |
-
3
|
| 61 |
-
],
|
| 62 |
-
[
|
| 63 |
-
512,
|
| 64 |
-
128,
|
| 65 |
-
1024,
|
| 66 |
-
1,
|
| 67 |
-
true,
|
| 68 |
-
true,
|
| 69 |
-
5,
|
| 70 |
-
3
|
| 71 |
-
]
|
| 72 |
],
|
| 73 |
"stem_channels": [
|
| 74 |
3,
|
|
@@ -308,8 +294,7 @@
|
|
| 308 |
"positional_encoding_temperature": 10000,
|
| 309 |
"reg_scale": 4.0,
|
| 310 |
"top_prob_values": 4,
|
| 311 |
-
"
|
| 312 |
-
"transformers_version": "4.49.0.dev0",
|
| 313 |
"use_focal_loss": true,
|
| 314 |
"use_pretrained_backbone": false,
|
| 315 |
"use_timm_backbone": false,
|
|
|
|
| 1 |
{
|
| 2 |
+
"_attn_implementation_autoset": true,
|
| 3 |
"activation_dropout": 0.0,
|
| 4 |
"activation_function": "silu",
|
| 5 |
"anchor_image_size": null,
|
|
|
|
|
|
|
|
|
|
| 6 |
"attention_dropout": 0.0,
|
| 7 |
"auxiliary_loss": true,
|
| 8 |
"backbone": null,
|
|
|
|
| 26 |
3,
|
| 27 |
4
|
| 28 |
],
|
| 29 |
+
"stage_in_channels": [
|
| 30 |
+
16,
|
| 31 |
+
64,
|
| 32 |
+
256,
|
| 33 |
+
512
|
| 34 |
+
],
|
| 35 |
+
"stage_mid_channels": [
|
| 36 |
+
16,
|
| 37 |
+
32,
|
| 38 |
+
64,
|
| 39 |
+
128
|
| 40 |
+
],
|
| 41 |
+
"stage_num_blocks": [
|
| 42 |
+
1,
|
| 43 |
+
1,
|
| 44 |
+
2,
|
| 45 |
+
1
|
| 46 |
+
],
|
| 47 |
+
"stage_numb_of_layers": [
|
| 48 |
+
3,
|
| 49 |
+
3,
|
| 50 |
+
3,
|
| 51 |
+
3
|
| 52 |
+
],
|
| 53 |
+
"stage_out_channels": [
|
| 54 |
+
64,
|
| 55 |
+
256,
|
| 56 |
+
512,
|
| 57 |
+
1024
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 58 |
],
|
| 59 |
"stem_channels": [
|
| 60 |
3,
|
|
|
|
| 294 |
"positional_encoding_temperature": 10000,
|
| 295 |
"reg_scale": 4.0,
|
| 296 |
"top_prob_values": 4,
|
| 297 |
+
"transformers_version": "4.50.0.dev0",
|
|
|
|
| 298 |
"use_focal_loss": true,
|
| 299 |
"use_pretrained_backbone": false,
|
| 300 |
"use_timm_backbone": false,
|