Lukas Pfahler
commited on
Commit
•
42896b3
1
Parent(s):
0c2389f
use real transformers lib and have changes locally
Browse files- image_processing_kosmos2_5.py +4 -2
- processing_kosmos2_5.py +5 -3
- special_tokens_map.json +30 -0
image_processing_kosmos2_5.py
CHANGED
@@ -16,7 +16,7 @@
|
|
16 |
|
17 |
import math
|
18 |
from typing import Dict, Optional, Union
|
19 |
-
|
20 |
import numpy as np
|
21 |
|
22 |
from transformers.image_processing_utils import BaseImageProcessor, BatchFeature
|
@@ -338,4 +338,6 @@ class Kosmos2_5ImageProcessor(BaseImageProcessor):
|
|
338 |
tensor_type=return_tensors,
|
339 |
)
|
340 |
|
341 |
-
return encoded_outputs
|
|
|
|
|
|
16 |
|
17 |
import math
|
18 |
from typing import Dict, Optional, Union
|
19 |
+
from transformers import AutoImageProcessor
|
20 |
import numpy as np
|
21 |
|
22 |
from transformers.image_processing_utils import BaseImageProcessor, BatchFeature
|
|
|
338 |
tensor_type=return_tensors,
|
339 |
)
|
340 |
|
341 |
+
return encoded_outputs
|
342 |
+
|
343 |
+
AutoImageProcessor.register("Kosmos2_5ImageProcessor", Kosmos2_5ImageProcessor)
|
processing_kosmos2_5.py
CHANGED
@@ -17,12 +17,13 @@ Processor class for Kosmos2_5.
|
|
17 |
"""
|
18 |
|
19 |
from typing import List, Optional, Union
|
20 |
-
|
21 |
from transformers.image_processing_utils import BatchFeature
|
22 |
from transformers.processing_utils import ProcessorMixin
|
23 |
from transformers.tokenization_utils_base import PaddingStrategy, TextInput, TruncationStrategy
|
24 |
from transformers.utils import TensorType, is_torch_available
|
25 |
-
|
|
|
26 |
|
27 |
if is_torch_available():
|
28 |
import torch
|
@@ -49,7 +50,8 @@ class Kosmos2_5Processor(ProcessorMixin):
|
|
49 |
|
50 |
def __init__(self, image_processor, tokenizer):
|
51 |
tokenizer.return_token_type_ids = False
|
52 |
-
|
|
|
53 |
|
54 |
def __call__(
|
55 |
self,
|
|
|
17 |
"""
|
18 |
|
19 |
from typing import List, Optional, Union
|
20 |
+
import transformers
|
21 |
from transformers.image_processing_utils import BatchFeature
|
22 |
from transformers.processing_utils import ProcessorMixin
|
23 |
from transformers.tokenization_utils_base import PaddingStrategy, TextInput, TruncationStrategy
|
24 |
from transformers.utils import TensorType, is_torch_available
|
25 |
+
from .image_processing_kosmos2_5 import Kosmos2_5ImageProcessor
|
26 |
+
transformers.Kosmos2_5ImageProcessor = Kosmos2_5ImageProcessor
|
27 |
|
28 |
if is_torch_available():
|
29 |
import torch
|
|
|
50 |
|
51 |
def __init__(self, image_processor, tokenizer):
|
52 |
tokenizer.return_token_type_ids = False
|
53 |
+
self.image_processor = image_processor
|
54 |
+
self.tokenizer = tokenizer
|
55 |
|
56 |
def __call__(
|
57 |
self,
|
special_tokens_map.json
ADDED
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": {
|
3 |
+
"content": "<s>",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": false,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"eos_token": {
|
10 |
+
"content": "</s>",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": false,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"pad_token": {
|
17 |
+
"content": "<pad>",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": false,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
},
|
23 |
+
"unk_token": {
|
24 |
+
"content": "<unk>",
|
25 |
+
"lstrip": false,
|
26 |
+
"normalized": false,
|
27 |
+
"rstrip": false,
|
28 |
+
"single_word": false
|
29 |
+
}
|
30 |
+
}
|