ckandemir commited on
Commit
bd08319
1 Parent(s): b51955d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +23 -100
README.md CHANGED
@@ -74,105 +74,6 @@ The dataset was preprocessed to fit the model's input requirements, with special
74
 
75
  <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
76
 
77
- [More Information Needed]
78
-
79
- ## Evaluation
80
-
81
- <!-- This section describes the evaluation protocols and provides the results. -->
82
-
83
- ### Testing Data, Factors & Metrics
84
-
85
- #### Testing Data
86
-
87
- <!-- This should link to a Data Card if possible. -->
88
-
89
- [More Information Needed]
90
-
91
- #### Factors
92
-
93
- <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
94
-
95
- [More Information Needed]
96
-
97
- #### Metrics
98
-
99
- <!-- These are the evaluation metrics being used, ideally with a description of why. -->
100
-
101
- [More Information Needed]
102
-
103
- ### Results
104
-
105
- [More Information Needed]
106
-
107
- #### Summary
108
-
109
-
110
-
111
- ## Model Examination [optional]
112
-
113
- <!-- Relevant interpretability work for the model goes here -->
114
-
115
- [More Information Needed]
116
-
117
- ## Environmental Impact
118
-
119
- <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
120
-
121
- Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
122
-
123
- - **Hardware Type:** [More Information Needed]
124
- - **Hours used:** [More Information Needed]
125
- - **Cloud Provider:** [More Information Needed]
126
- - **Compute Region:** [More Information Needed]
127
- - **Carbon Emitted:** [More Information Needed]
128
-
129
- ## Technical Specifications [optional]
130
-
131
- ### Model Architecture and Objective
132
-
133
- [More Information Needed]
134
-
135
- ### Compute Infrastructure
136
-
137
- [More Information Needed]
138
-
139
- #### Hardware
140
-
141
- [More Information Needed]
142
-
143
- #### Software
144
-
145
- [More Information Needed]
146
-
147
- ## Citation [optional]
148
-
149
- <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
150
-
151
- **BibTeX:**
152
-
153
- [More Information Needed]
154
-
155
- **APA:**
156
-
157
- [More Information Needed]
158
-
159
- ## Glossary [optional]
160
-
161
- <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
162
-
163
- [More Information Needed]
164
-
165
- ## More Information [optional]
166
-
167
- [More Information Needed]
168
-
169
- ## Model Card Authors [optional]
170
-
171
- [More Information Needed]
172
-
173
- ## Model Card Contact
174
-
175
- [More Information Needed]
176
 
177
 
178
  ## Training procedure
@@ -193,4 +94,26 @@ The following `bitsandbytes` quantization config was used during training:
193
  ### Framework versions
194
 
195
 
196
- - PEFT 0.6.2.dev0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
74
 
75
  <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
76
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
77
 
78
 
79
  ## Training procedure
 
94
  ### Framework versions
95
 
96
 
97
+ - PEFT 0.6.2.dev0
98
+
99
+ ### Code
100
+
101
+ ```python
102
+ from peft import PeftModel, PeftConfig
103
+
104
+ peft_model_id='ckandemir/idefics-9b-pokemon-peft'
105
+ quantization_config = BitsAndBytesConfig(
106
+ load_in_4bit=True,
107
+ bnb_4bit_use_double_quant=True,
108
+ bnb_4bit_quant_type="nf4",
109
+ bnb_4bit_compute_dtype=torch.float16,
110
+ llm_int8_skip_modules=["lm_head", "embed_tokens"],
111
+ )
112
+ config = PeftConfig.from_pretrained(peft_model_id)
113
+ # Load the model with the appropriate configuration for inference
114
+ model = IdeficsForVisionText2Text.from_pretrained(config.base_model_name_or_path, quantization_config=quantization_config, device_map="auto")
115
+ processor = AutoProcessor.from_pretrained(config.base_model_name_or_path)
116
+
117
+ model = PeftModel.from_pretrained(model, peft_model_id, device_map={"":0})
118
+
119
+ ```