Instructions to use ChayanM/ViT-Bert_Mimic with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ChayanM/ViT-Bert_Mimic with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="ChayanM/ViT-Bert_Mimic")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("ChayanM/ViT-Bert_Mimic")
model = AutoModelForMultimodalLM.from_pretrained("ChayanM/ViT-Bert_Mimic", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ChayanM/ViT-Bert_Mimic with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ChayanM/ViT-Bert_Mimic"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ChayanM/ViT-Bert_Mimic",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ChayanM/ViT-Bert_Mimic

SGLang

How to use ChayanM/ViT-Bert_Mimic with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "ChayanM/ViT-Bert_Mimic" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ChayanM/ViT-Bert_Mimic",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "ChayanM/ViT-Bert_Mimic" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ChayanM/ViT-Bert_Mimic",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ChayanM/ViT-Bert_Mimic with Docker Model Runner:
```
docker model run hf.co/ChayanM/ViT-Bert_Mimic
```

ChayanM commited on Apr 9, 2024

Commit

dfd3089

verified ·

1 Parent(s): 7d53580

Model save

Browse files

Files changed (1) hide show

README.md +10 -22

README.md CHANGED Viewed

@@ -15,12 +15,12 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1692
-- Rouge1: 20.3071
-- Rouge2: 17.0769
-- Rougel: 20.3425
-- Rougelsum: 20.2544
-- Gen Len: 17.62
 ## Model description
@@ -45,27 +45,15 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 15
 ### Training results
 | Training Loss | Epoch | Step | Validation Loss | Rouge1  | Rouge2  | Rougel  | Rougelsum | Gen Len |
 |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
-| No log        | 1.0   | 250  | 0.1710          | 0.0     | 0.0     | 0.0     | 0.0       | 20.0    |
-| 0.3208        | 2.0   | 500  | 0.1424          | 0.0     | 0.0     | 0.0     | 0.0       | 20.0    |
-| 0.3208        | 3.0   | 750  | 0.1396          | 28.6163 | 23.3316 | 28.7485 | 28.5768   | 15.76   |
-| 0.087         | 4.0   | 1000 | 0.1349          | 15.1174 | 12.3333 | 15.1103 | 15.0938   | 18.16   |
-| 0.087         | 5.0   | 1250 | 0.1373          | 1.5     | 0.6667  | 1.5     | 1.5       | 19.8    |
-| 0.0621        | 6.0   | 1500 | 0.1418          | 29.7634 | 24.3915 | 29.9052 | 29.7154   | 16.0    |
-| 0.0621        | 7.0   | 1750 | 0.1453          | 1.0     | 1.0     | 1.0     | 1.0       | 19.92   |
-| 0.0415        | 8.0   | 2000 | 0.1504          | 6.75    | 5.6667  | 6.75    | 6.75      | 19.22   |
-| 0.0415        | 9.0   | 2250 | 0.1547          | 18.9444 | 16.7436 | 19.1202 | 18.9032   | 18.06   |
-| 0.0254        | 10.0  | 2500 | 0.1591          | 19.2342 | 15.6667 | 19.2782 | 19.1123   | 17.44   |
-| 0.0254        | 11.0  | 2750 | 0.1601          | 19.9154 | 17.3654 | 20.011  | 19.8499   | 17.91   |
-| 0.0147        | 12.0  | 3000 | 0.1647          | 16.9972 | 13.0769 | 16.9972 | 16.9154   | 17.96   |
-| 0.0147        | 13.0  | 3250 | 0.1676          | 22.2859 | 16.9093 | 22.4548 | 22.1609   | 16.66   |
-| 0.0094        | 14.0  | 3500 | 0.1684          | 20.2714 | 16.7308 | 20.4544 | 20.2507   | 17.47   |
-| 0.0094        | 15.0  | 3750 | 0.1692          | 20.3071 | 17.0769 | 20.3425 | 20.2544   | 17.62   |
 ### Framework versions

 This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.0924
+- Rouge1: 34.7587
+- Rouge2: 25.3644
+- Rougel: 34.5531
+- Rougelsum: 34.6322
+- Gen Len: 13.584
 ## Model description
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
+- num_epochs: 3
 ### Training results
 | Training Loss | Epoch | Step | Validation Loss | Rouge1  | Rouge2  | Rougel  | Rougelsum | Gen Len |
 |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
+| 0.0952        | 1.0   | 2500 | 0.0960          | 26.9673 | 20.9727 | 26.939  | 26.9315   | 15.552  |
+| 0.0768        | 2.0   | 5000 | 0.0931          | 35.4227 | 25.9227 | 35.1707 | 35.3002   | 13.552  |
+| 0.0662        | 3.0   | 7500 | 0.0924          | 34.7587 | 25.3644 | 34.5531 | 34.6322   | 13.584  |
 ### Framework versions