Fixing tokenizer class
Browse files- Dockerfile +1 -0
- requirements.txt +2 -1
Dockerfile
CHANGED
|
@@ -5,6 +5,7 @@ WORKDIR /app
|
|
| 5 |
|
| 6 |
# Install system dependencies
|
| 7 |
RUN apt-get update && apt-get install -y \
|
|
|
|
| 8 |
tesseract-ocr \
|
| 9 |
tesseract-ocr-eng \
|
| 10 |
poppler-utils \
|
|
|
|
| 5 |
|
| 6 |
# Install system dependencies
|
| 7 |
RUN apt-get update && apt-get install -y \
|
| 8 |
+
git \
|
| 9 |
tesseract-ocr \
|
| 10 |
tesseract-ocr-eng \
|
| 11 |
poppler-utils \
|
requirements.txt
CHANGED
|
@@ -1,8 +1,9 @@
|
|
| 1 |
fastapi==0.109.0
|
| 2 |
uvicorn[standard]==0.27.0
|
| 3 |
python-multipart==0.0.6
|
| 4 |
-
transformers
|
| 5 |
accelerate>=0.30
|
|
|
|
| 6 |
torch==2.2.2
|
| 7 |
pypdf==4.0.1
|
| 8 |
pdf2image==1.17.0
|
|
|
|
| 1 |
fastapi==0.109.0
|
| 2 |
uvicorn[standard]==0.27.0
|
| 3 |
python-multipart==0.0.6
|
| 4 |
+
transformers @ git+https://github.com/huggingface/transformers.git@main
|
| 5 |
accelerate>=0.30
|
| 6 |
+
tiktoken
|
| 7 |
torch==2.2.2
|
| 8 |
pypdf==4.0.1
|
| 9 |
pdf2image==1.17.0
|