ayyuce/blip-vqa-rad
Browse files- README.md +17 -9
- config.json +4 -29
- generation_config.json +1 -1
- model.safetensors +1 -1
- training_args.bin +2 -2
README.md
CHANGED
@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
16 |
|
17 |
This model is a fine-tuned version of [Salesforce/blip-vqa-base](https://huggingface.co/Salesforce/blip-vqa-base) on an unknown dataset.
|
18 |
It achieves the following results on the evaluation set:
|
19 |
-
- Loss: 0.
|
20 |
|
21 |
## Model description
|
22 |
|
@@ -39,21 +39,29 @@ The following hyperparameters were used during training:
|
|
39 |
- train_batch_size: 2
|
40 |
- eval_batch_size: 2
|
41 |
- seed: 42
|
42 |
-
- optimizer: Use
|
43 |
- lr_scheduler_type: linear
|
44 |
-
- num_epochs:
|
45 |
|
46 |
### Training results
|
47 |
|
48 |
| Training Loss | Epoch | Step | Validation Loss |
|
49 |
|:-------------:|:-----:|:----:|:---------------:|
|
50 |
-
| 0.
|
51 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
52 |
|
53 |
|
54 |
### Framework versions
|
55 |
|
56 |
-
- Transformers 4.
|
57 |
-
- Pytorch 2.
|
58 |
-
- Datasets 3.
|
59 |
-
- Tokenizers 0.21.
|
|
|
16 |
|
17 |
This model is a fine-tuned version of [Salesforce/blip-vqa-base](https://huggingface.co/Salesforce/blip-vqa-base) on an unknown dataset.
|
18 |
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 0.3083
|
20 |
|
21 |
## Model description
|
22 |
|
|
|
39 |
- train_batch_size: 2
|
40 |
- eval_batch_size: 2
|
41 |
- seed: 42
|
42 |
+
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
43 |
- lr_scheduler_type: linear
|
44 |
+
- num_epochs: 10
|
45 |
|
46 |
### Training results
|
47 |
|
48 |
| Training Loss | Epoch | Step | Validation Loss |
|
49 |
|:-------------:|:-----:|:----:|:---------------:|
|
50 |
+
| 0.3482 | 1.0 | 897 | 0.3006 |
|
51 |
+
| 0.3143 | 2.0 | 1794 | 0.2856 |
|
52 |
+
| 0.2695 | 3.0 | 2691 | 0.2854 |
|
53 |
+
| 0.2053 | 4.0 | 3588 | 0.2910 |
|
54 |
+
| 0.2152 | 5.0 | 4485 | 0.2987 |
|
55 |
+
| 0.2841 | 6.0 | 5382 | 0.2927 |
|
56 |
+
| 0.2106 | 7.0 | 6279 | 0.3007 |
|
57 |
+
| 0.1746 | 8.0 | 7176 | 0.3027 |
|
58 |
+
| 0.2473 | 9.0 | 8073 | 0.3073 |
|
59 |
+
| 0.2259 | 10.0 | 8970 | 0.3083 |
|
60 |
|
61 |
|
62 |
### Framework versions
|
63 |
|
64 |
+
- Transformers 4.47.0
|
65 |
+
- Pytorch 2.5.1+cu121
|
66 |
+
- Datasets 3.3.1
|
67 |
+
- Tokenizers 0.21.0
|
config.json
CHANGED
@@ -1,4 +1,5 @@
|
|
1 |
{
|
|
|
2 |
"architectures": [
|
3 |
"BlipForConditionalGeneration"
|
4 |
],
|
@@ -11,44 +12,18 @@
|
|
11 |
"projection_dim": 512,
|
12 |
"text_config": {
|
13 |
"_attn_implementation_autoset": true,
|
14 |
-
"attention_probs_dropout_prob": 0.0,
|
15 |
-
"encoder_hidden_size": 768,
|
16 |
-
"hidden_act": "gelu",
|
17 |
-
"hidden_dropout_prob": 0.0,
|
18 |
-
"hidden_size": 768,
|
19 |
"initializer_factor": 1.0,
|
20 |
-
"initializer_range": 0.02,
|
21 |
-
"intermediate_size": 3072,
|
22 |
-
"label_smoothing": 0.0,
|
23 |
-
"layer_norm_eps": 1e-12,
|
24 |
-
"max_position_embeddings": 512,
|
25 |
"model_type": "blip_text_model",
|
26 |
-
"num_attention_heads": 12
|
27 |
-
"num_hidden_layers": 12,
|
28 |
-
"projection_dim": 768,
|
29 |
-
"torch_dtype": "float32",
|
30 |
-
"use_cache": true,
|
31 |
-
"vocab_size": 30524
|
32 |
},
|
33 |
"torch_dtype": "float32",
|
34 |
-
"transformers_version": "4.
|
35 |
"vision_config": {
|
36 |
"_attn_implementation_autoset": true,
|
37 |
-
"attention_dropout": 0.0,
|
38 |
"dropout": 0.0,
|
39 |
-
"hidden_act": "gelu",
|
40 |
-
"hidden_size": 768,
|
41 |
-
"image_size": 384,
|
42 |
"initializer_factor": 1.0,
|
43 |
"initializer_range": 0.02,
|
44 |
-
"intermediate_size": 3072,
|
45 |
-
"layer_norm_eps": 1e-05,
|
46 |
"model_type": "blip_vision_model",
|
47 |
-
"
|
48 |
-
"num_channels": 3,
|
49 |
-
"num_hidden_layers": 12,
|
50 |
-
"patch_size": 16,
|
51 |
-
"projection_dim": 512,
|
52 |
-
"torch_dtype": "float32"
|
53 |
}
|
54 |
}
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "Salesforce/blip-vqa-base",
|
3 |
"architectures": [
|
4 |
"BlipForConditionalGeneration"
|
5 |
],
|
|
|
12 |
"projection_dim": 512,
|
13 |
"text_config": {
|
14 |
"_attn_implementation_autoset": true,
|
|
|
|
|
|
|
|
|
|
|
15 |
"initializer_factor": 1.0,
|
|
|
|
|
|
|
|
|
|
|
16 |
"model_type": "blip_text_model",
|
17 |
+
"num_attention_heads": 12
|
|
|
|
|
|
|
|
|
|
|
18 |
},
|
19 |
"torch_dtype": "float32",
|
20 |
+
"transformers_version": "4.47.0",
|
21 |
"vision_config": {
|
22 |
"_attn_implementation_autoset": true,
|
|
|
23 |
"dropout": 0.0,
|
|
|
|
|
|
|
24 |
"initializer_factor": 1.0,
|
25 |
"initializer_range": 0.02,
|
|
|
|
|
26 |
"model_type": "blip_vision_model",
|
27 |
+
"num_channels": 3
|
|
|
|
|
|
|
|
|
|
|
28 |
}
|
29 |
}
|
generation_config.json
CHANGED
@@ -3,5 +3,5 @@
|
|
3 |
"bos_token_id": 30522,
|
4 |
"eos_token_id": 2,
|
5 |
"pad_token_id": 0,
|
6 |
-
"transformers_version": "4.
|
7 |
}
|
|
|
3 |
"bos_token_id": 30522,
|
4 |
"eos_token_id": 2,
|
5 |
"pad_token_id": 0,
|
6 |
+
"transformers_version": "4.47.0"
|
7 |
}
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 989717056
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a4c5a7f2b692b3b08bd9194d5897fa37cbff309d1e36841be618b2dff5e829ef
|
3 |
size 989717056
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:694457f2f177f35a1c7b78ae2c8dcc4a0ac46d9534b37dfd1b78e54ab8af48bd
|
3 |
+
size 5304
|