Commit
•
e8459ed
1
Parent(s):
1170378
Chest-Xray-report-generation
Browse files- README.md +59 -0
- config.json +184 -0
- generation_config.json +9 -0
- model.safetensors +3 -0
- preprocessor_config.json +22 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
tags:
|
3 |
+
- generated_from_trainer
|
4 |
+
model-index:
|
5 |
+
- name: image-caption-generator
|
6 |
+
results: []
|
7 |
+
---
|
8 |
+
|
9 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
10 |
+
should probably proofread and complete it, then remove this comment. -->
|
11 |
+
|
12 |
+
# image-caption-generator
|
13 |
+
|
14 |
+
This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
|
15 |
+
It achieves the following results on the evaluation set:
|
16 |
+
- Loss: 0.3393
|
17 |
+
|
18 |
+
## Model description
|
19 |
+
|
20 |
+
More information needed
|
21 |
+
|
22 |
+
## Intended uses & limitations
|
23 |
+
|
24 |
+
More information needed
|
25 |
+
|
26 |
+
## Training and evaluation data
|
27 |
+
|
28 |
+
More information needed
|
29 |
+
|
30 |
+
## Training procedure
|
31 |
+
|
32 |
+
### Training hyperparameters
|
33 |
+
|
34 |
+
The following hyperparameters were used during training:
|
35 |
+
- learning_rate: 5e-05
|
36 |
+
- train_batch_size: 32
|
37 |
+
- eval_batch_size: 32
|
38 |
+
- seed: 42
|
39 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
40 |
+
- lr_scheduler_type: linear
|
41 |
+
- num_epochs: 5
|
42 |
+
|
43 |
+
### Training results
|
44 |
+
|
45 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
46 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
47 |
+
| No log | 1.0 | 203 | 0.4971 |
|
48 |
+
| No log | 2.0 | 406 | 0.4143 |
|
49 |
+
| 0.6123 | 3.0 | 609 | 0.3731 |
|
50 |
+
| 0.6123 | 4.0 | 812 | 0.3481 |
|
51 |
+
| 0.4078 | 5.0 | 1015 | 0.3393 |
|
52 |
+
|
53 |
+
|
54 |
+
### Framework versions
|
55 |
+
|
56 |
+
- Transformers 4.36.2
|
57 |
+
- Pytorch 2.0.0
|
58 |
+
- Datasets 2.1.0
|
59 |
+
- Tokenizers 0.15.0
|
config.json
ADDED
@@ -0,0 +1,184 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"VisionEncoderDecoderModel"
|
4 |
+
],
|
5 |
+
"decoder": {
|
6 |
+
"_name_or_path": "gpt2",
|
7 |
+
"activation_function": "gelu_new",
|
8 |
+
"add_cross_attention": true,
|
9 |
+
"architectures": [
|
10 |
+
"GPT2LMHeadModel"
|
11 |
+
],
|
12 |
+
"attn_pdrop": 0.1,
|
13 |
+
"bad_words_ids": null,
|
14 |
+
"begin_suppress_tokens": null,
|
15 |
+
"bos_token_id": 50256,
|
16 |
+
"chunk_size_feed_forward": 0,
|
17 |
+
"cross_attention_hidden_size": null,
|
18 |
+
"decoder_start_token_id": null,
|
19 |
+
"diversity_penalty": 0.0,
|
20 |
+
"do_sample": false,
|
21 |
+
"early_stopping": false,
|
22 |
+
"embd_pdrop": 0.1,
|
23 |
+
"encoder_no_repeat_ngram_size": 0,
|
24 |
+
"eos_token_id": 50256,
|
25 |
+
"exponential_decay_length_penalty": null,
|
26 |
+
"finetuning_task": null,
|
27 |
+
"forced_bos_token_id": null,
|
28 |
+
"forced_eos_token_id": null,
|
29 |
+
"id2label": {
|
30 |
+
"0": "LABEL_0",
|
31 |
+
"1": "LABEL_1"
|
32 |
+
},
|
33 |
+
"initializer_range": 0.02,
|
34 |
+
"is_decoder": true,
|
35 |
+
"is_encoder_decoder": false,
|
36 |
+
"label2id": {
|
37 |
+
"LABEL_0": 0,
|
38 |
+
"LABEL_1": 1
|
39 |
+
},
|
40 |
+
"layer_norm_epsilon": 1e-05,
|
41 |
+
"length_penalty": 1.0,
|
42 |
+
"max_length": 20,
|
43 |
+
"min_length": 0,
|
44 |
+
"model_type": "gpt2",
|
45 |
+
"n_ctx": 1024,
|
46 |
+
"n_embd": 768,
|
47 |
+
"n_head": 12,
|
48 |
+
"n_inner": null,
|
49 |
+
"n_layer": 12,
|
50 |
+
"n_positions": 1024,
|
51 |
+
"no_repeat_ngram_size": 0,
|
52 |
+
"num_beam_groups": 1,
|
53 |
+
"num_beams": 1,
|
54 |
+
"num_return_sequences": 1,
|
55 |
+
"output_attentions": false,
|
56 |
+
"output_hidden_states": false,
|
57 |
+
"output_scores": false,
|
58 |
+
"pad_token_id": null,
|
59 |
+
"prefix": null,
|
60 |
+
"problem_type": null,
|
61 |
+
"pruned_heads": {},
|
62 |
+
"remove_invalid_values": false,
|
63 |
+
"reorder_and_upcast_attn": false,
|
64 |
+
"repetition_penalty": 1.0,
|
65 |
+
"resid_pdrop": 0.1,
|
66 |
+
"return_dict": true,
|
67 |
+
"return_dict_in_generate": false,
|
68 |
+
"scale_attn_by_inverse_layer_idx": false,
|
69 |
+
"scale_attn_weights": true,
|
70 |
+
"sep_token_id": null,
|
71 |
+
"summary_activation": null,
|
72 |
+
"summary_first_dropout": 0.1,
|
73 |
+
"summary_proj_to_labels": true,
|
74 |
+
"summary_type": "cls_index",
|
75 |
+
"summary_use_proj": true,
|
76 |
+
"suppress_tokens": null,
|
77 |
+
"task_specific_params": {
|
78 |
+
"text-generation": {
|
79 |
+
"do_sample": true,
|
80 |
+
"max_length": 50
|
81 |
+
}
|
82 |
+
},
|
83 |
+
"temperature": 1.0,
|
84 |
+
"tf_legacy_loss": false,
|
85 |
+
"tie_encoder_decoder": false,
|
86 |
+
"tie_word_embeddings": true,
|
87 |
+
"tokenizer_class": null,
|
88 |
+
"top_k": 50,
|
89 |
+
"top_p": 1.0,
|
90 |
+
"torch_dtype": null,
|
91 |
+
"torchscript": false,
|
92 |
+
"typical_p": 1.0,
|
93 |
+
"use_bfloat16": false,
|
94 |
+
"use_cache": true,
|
95 |
+
"vocab_size": 50257
|
96 |
+
},
|
97 |
+
"decoder_start_token_id": 50256,
|
98 |
+
"encoder": {
|
99 |
+
"_name_or_path": "google/vit-base-patch16-224-in21k",
|
100 |
+
"add_cross_attention": false,
|
101 |
+
"architectures": [
|
102 |
+
"ViTModel"
|
103 |
+
],
|
104 |
+
"attention_probs_dropout_prob": 0.0,
|
105 |
+
"bad_words_ids": null,
|
106 |
+
"begin_suppress_tokens": null,
|
107 |
+
"bos_token_id": null,
|
108 |
+
"chunk_size_feed_forward": 0,
|
109 |
+
"cross_attention_hidden_size": null,
|
110 |
+
"decoder_start_token_id": null,
|
111 |
+
"diversity_penalty": 0.0,
|
112 |
+
"do_sample": false,
|
113 |
+
"early_stopping": false,
|
114 |
+
"encoder_no_repeat_ngram_size": 0,
|
115 |
+
"encoder_stride": 16,
|
116 |
+
"eos_token_id": null,
|
117 |
+
"exponential_decay_length_penalty": null,
|
118 |
+
"finetuning_task": null,
|
119 |
+
"forced_bos_token_id": null,
|
120 |
+
"forced_eos_token_id": null,
|
121 |
+
"hidden_act": "gelu",
|
122 |
+
"hidden_dropout_prob": 0.0,
|
123 |
+
"hidden_size": 768,
|
124 |
+
"id2label": {
|
125 |
+
"0": "LABEL_0",
|
126 |
+
"1": "LABEL_1"
|
127 |
+
},
|
128 |
+
"image_size": 224,
|
129 |
+
"initializer_range": 0.02,
|
130 |
+
"intermediate_size": 3072,
|
131 |
+
"is_decoder": false,
|
132 |
+
"is_encoder_decoder": false,
|
133 |
+
"label2id": {
|
134 |
+
"LABEL_0": 0,
|
135 |
+
"LABEL_1": 1
|
136 |
+
},
|
137 |
+
"layer_norm_eps": 1e-12,
|
138 |
+
"length_penalty": 1.0,
|
139 |
+
"max_length": 20,
|
140 |
+
"min_length": 0,
|
141 |
+
"model_type": "vit",
|
142 |
+
"no_repeat_ngram_size": 0,
|
143 |
+
"num_attention_heads": 12,
|
144 |
+
"num_beam_groups": 1,
|
145 |
+
"num_beams": 1,
|
146 |
+
"num_channels": 3,
|
147 |
+
"num_hidden_layers": 12,
|
148 |
+
"num_return_sequences": 1,
|
149 |
+
"output_attentions": false,
|
150 |
+
"output_hidden_states": false,
|
151 |
+
"output_scores": false,
|
152 |
+
"pad_token_id": null,
|
153 |
+
"patch_size": 16,
|
154 |
+
"prefix": null,
|
155 |
+
"problem_type": null,
|
156 |
+
"pruned_heads": {},
|
157 |
+
"qkv_bias": true,
|
158 |
+
"remove_invalid_values": false,
|
159 |
+
"repetition_penalty": 1.0,
|
160 |
+
"return_dict": true,
|
161 |
+
"return_dict_in_generate": false,
|
162 |
+
"sep_token_id": null,
|
163 |
+
"suppress_tokens": null,
|
164 |
+
"task_specific_params": null,
|
165 |
+
"temperature": 1.0,
|
166 |
+
"tf_legacy_loss": false,
|
167 |
+
"tie_encoder_decoder": false,
|
168 |
+
"tie_word_embeddings": true,
|
169 |
+
"tokenizer_class": null,
|
170 |
+
"top_k": 50,
|
171 |
+
"top_p": 1.0,
|
172 |
+
"torch_dtype": null,
|
173 |
+
"torchscript": false,
|
174 |
+
"typical_p": 1.0,
|
175 |
+
"use_bfloat16": false
|
176 |
+
},
|
177 |
+
"is_encoder_decoder": true,
|
178 |
+
"model_type": "vision-encoder-decoder",
|
179 |
+
"num_beams": 4,
|
180 |
+
"pad_token_id": 50256,
|
181 |
+
"tie_word_embeddings": false,
|
182 |
+
"torch_dtype": "float32",
|
183 |
+
"transformers_version": "4.36.2"
|
184 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 50256,
|
4 |
+
"decoder_start_token_id": 50256,
|
5 |
+
"eos_token_id": 50256,
|
6 |
+
"num_beams": 4,
|
7 |
+
"pad_token_id": 50256,
|
8 |
+
"transformers_version": "4.36.2"
|
9 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:25d40e3da8ef3017faf6733e519967966ac40ff4cc7278ad7d0bacc29311b49a
|
3 |
+
size 956835520
|
preprocessor_config.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_normalize": true,
|
3 |
+
"do_rescale": true,
|
4 |
+
"do_resize": true,
|
5 |
+
"image_mean": [
|
6 |
+
0.5,
|
7 |
+
0.5,
|
8 |
+
0.5
|
9 |
+
],
|
10 |
+
"image_processor_type": "ViTFeatureExtractor",
|
11 |
+
"image_std": [
|
12 |
+
0.5,
|
13 |
+
0.5,
|
14 |
+
0.5
|
15 |
+
],
|
16 |
+
"resample": 2,
|
17 |
+
"rescale_factor": 0.00392156862745098,
|
18 |
+
"size": {
|
19 |
+
"height": 224,
|
20 |
+
"width": 224
|
21 |
+
}
|
22 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d290cd27a317ef7616d329594409ec5d660657e715c25e131e26634cb3f0d68d
|
3 |
+
size 4411
|