Training in progress, step 500
Browse files- config.json +30 -3
- model.safetensors +1 -1
- training_args.bin +1 -1
config.json
CHANGED
|
@@ -1,5 +1,4 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "OFA-Sys/chinese-clip-vit-base-patch16",
|
| 3 |
"architectures": [
|
| 4 |
"ChineseCLIPDualTextModel"
|
| 5 |
],
|
|
@@ -13,23 +12,51 @@
|
|
| 13 |
"architectures": [
|
| 14 |
"ChineseCLIPTextModel"
|
| 15 |
],
|
|
|
|
| 16 |
"bos_token_id": 0,
|
| 17 |
"directionality": "bidi",
|
| 18 |
"eos_token_id": 2,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
"model_type": "chinese_clip_text_model",
|
|
|
|
|
|
|
| 20 |
"output_past": true,
|
| 21 |
"pooler_fc_size": 768,
|
| 22 |
"pooler_num_attention_heads": 12,
|
| 23 |
"pooler_num_fc_layers": 3,
|
| 24 |
"pooler_size_per_head": 128,
|
| 25 |
"pooler_type": "first_token_transform",
|
|
|
|
|
|
|
|
|
|
|
|
|
| 26 |
"vocab_size": 21128
|
| 27 |
},
|
| 28 |
"torch_dtype": "float32",
|
| 29 |
-
"transformers_version": "4.
|
| 30 |
"vision_config": {
|
|
|
|
| 31 |
"dropout": 0.0,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
"model_type": "chinese_clip_vision_model",
|
| 33 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 34 |
}
|
| 35 |
}
|
|
|
|
| 1 |
{
|
|
|
|
| 2 |
"architectures": [
|
| 3 |
"ChineseCLIPDualTextModel"
|
| 4 |
],
|
|
|
|
| 12 |
"architectures": [
|
| 13 |
"ChineseCLIPTextModel"
|
| 14 |
],
|
| 15 |
+
"attention_probs_dropout_prob": 0.1,
|
| 16 |
"bos_token_id": 0,
|
| 17 |
"directionality": "bidi",
|
| 18 |
"eos_token_id": 2,
|
| 19 |
+
"hidden_act": "gelu",
|
| 20 |
+
"hidden_dropout_prob": 0.1,
|
| 21 |
+
"hidden_size": 768,
|
| 22 |
+
"initializer_factor": 1.0,
|
| 23 |
+
"initializer_range": 0.02,
|
| 24 |
+
"intermediate_size": 3072,
|
| 25 |
+
"layer_norm_eps": 1e-12,
|
| 26 |
+
"max_position_embeddings": 512,
|
| 27 |
"model_type": "chinese_clip_text_model",
|
| 28 |
+
"num_attention_heads": 12,
|
| 29 |
+
"num_hidden_layers": 12,
|
| 30 |
"output_past": true,
|
| 31 |
"pooler_fc_size": 768,
|
| 32 |
"pooler_num_attention_heads": 12,
|
| 33 |
"pooler_num_fc_layers": 3,
|
| 34 |
"pooler_size_per_head": 128,
|
| 35 |
"pooler_type": "first_token_transform",
|
| 36 |
+
"position_embedding_type": "absolute",
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"type_vocab_size": 2,
|
| 39 |
+
"use_cache": true,
|
| 40 |
"vocab_size": 21128
|
| 41 |
},
|
| 42 |
"torch_dtype": "float32",
|
| 43 |
+
"transformers_version": "4.50.0",
|
| 44 |
"vision_config": {
|
| 45 |
+
"attention_dropout": 0.0,
|
| 46 |
"dropout": 0.0,
|
| 47 |
+
"hidden_act": "quick_gelu",
|
| 48 |
+
"hidden_size": 768,
|
| 49 |
+
"image_size": 224,
|
| 50 |
+
"initializer_factor": 1.0,
|
| 51 |
+
"initializer_range": 0.02,
|
| 52 |
+
"intermediate_size": 3072,
|
| 53 |
+
"layer_norm_eps": 1e-05,
|
| 54 |
"model_type": "chinese_clip_vision_model",
|
| 55 |
+
"num_attention_heads": 12,
|
| 56 |
+
"num_channels": 3,
|
| 57 |
+
"num_hidden_layers": 12,
|
| 58 |
+
"patch_size": 16,
|
| 59 |
+
"projection_dim": 512,
|
| 60 |
+
"torch_dtype": "float32"
|
| 61 |
}
|
| 62 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 816611916
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:11f909873fdc7838e11fe9c30fbf3ec74b19eb1dfe4298dfe4f3d25f367904cd
|
| 3 |
size 816611916
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5368
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4760b48d8ca9c7295e9f6fad0a7c66c199044ff324849da0a9b7b1b7b12dce6a
|
| 3 |
size 5368
|