Training in progress, step 8000, checkpoint
Browse files
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 966995080
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e6272cdab34ad742ff5d8928b1a3b73208418c7bfce7d87a0db8c4712d83527d
|
| 3 |
size 966995080
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1925064044
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d7de0ef00f4d8d9f4fa9b0e18f25384399f28c54b520f62c3dd31eca12ff60e2
|
| 3 |
size 1925064044
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5bcd75decc8ec809bdd000c1a023eecd569d9a9775fe640822926fa2ab60021b
|
| 3 |
size 14244
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:64cb8f98c15d4c048d873b0e5f521cf693b6283121ee23f325cb6b6e1b684e58
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": 47.53661784287617,
|
| 3 |
"best_model_checkpoint": "./whisper-small-taiwanese/checkpoint-7000",
|
| 4 |
-
"epoch": 2
|
| 5 |
"eval_steps": 1000,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -2030,6 +2030,295 @@
|
|
| 2030 |
"eval_samples_per_second": 2.27,
|
| 2031 |
"eval_steps_per_second": 0.284,
|
| 2032 |
"step": 7000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2033 |
}
|
| 2034 |
],
|
| 2035 |
"logging_steps": 25,
|
|
@@ -2037,7 +2326,7 @@
|
|
| 2037 |
"num_input_tokens_seen": 0,
|
| 2038 |
"num_train_epochs": 4,
|
| 2039 |
"save_steps": 1000,
|
| 2040 |
-
"total_flos": 3.
|
| 2041 |
"train_batch_size": 8,
|
| 2042 |
"trial_name": null,
|
| 2043 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": 47.53661784287617,
|
| 3 |
"best_model_checkpoint": "./whisper-small-taiwanese/checkpoint-7000",
|
| 4 |
+
"epoch": 3.2,
|
| 5 |
"eval_steps": 1000,
|
| 6 |
+
"global_step": 8000,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 2030 |
"eval_samples_per_second": 2.27,
|
| 2031 |
"eval_steps_per_second": 0.284,
|
| 2032 |
"step": 7000
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 2.81,
|
| 2036 |
+
"grad_norm": 12.68954086303711,
|
| 2037 |
+
"learning_rate": 2.438271604938272e-06,
|
| 2038 |
+
"loss": 0.6009,
|
| 2039 |
+
"step": 7025
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 2.82,
|
| 2043 |
+
"grad_norm": 10.792320251464844,
|
| 2044 |
+
"learning_rate": 2.4074074074074075e-06,
|
| 2045 |
+
"loss": 0.5913,
|
| 2046 |
+
"step": 7050
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 2.83,
|
| 2050 |
+
"grad_norm": 10.7244291305542,
|
| 2051 |
+
"learning_rate": 2.3765432098765435e-06,
|
| 2052 |
+
"loss": 0.5993,
|
| 2053 |
+
"step": 7075
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 2.84,
|
| 2057 |
+
"grad_norm": 10.002340316772461,
|
| 2058 |
+
"learning_rate": 2.345679012345679e-06,
|
| 2059 |
+
"loss": 0.5822,
|
| 2060 |
+
"step": 7100
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 2.85,
|
| 2064 |
+
"grad_norm": 9.782127380371094,
|
| 2065 |
+
"learning_rate": 2.314814814814815e-06,
|
| 2066 |
+
"loss": 0.6265,
|
| 2067 |
+
"step": 7125
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 2.86,
|
| 2071 |
+
"grad_norm": 11.622259140014648,
|
| 2072 |
+
"learning_rate": 2.283950617283951e-06,
|
| 2073 |
+
"loss": 0.6251,
|
| 2074 |
+
"step": 7150
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 2.87,
|
| 2078 |
+
"grad_norm": 9.750394821166992,
|
| 2079 |
+
"learning_rate": 2.2530864197530865e-06,
|
| 2080 |
+
"loss": 0.5412,
|
| 2081 |
+
"step": 7175
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 2.88,
|
| 2085 |
+
"grad_norm": 11.742593765258789,
|
| 2086 |
+
"learning_rate": 2.222222222222222e-06,
|
| 2087 |
+
"loss": 0.6171,
|
| 2088 |
+
"step": 7200
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 2.89,
|
| 2092 |
+
"grad_norm": 11.124135971069336,
|
| 2093 |
+
"learning_rate": 2.191358024691358e-06,
|
| 2094 |
+
"loss": 0.555,
|
| 2095 |
+
"step": 7225
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 2.9,
|
| 2099 |
+
"grad_norm": 13.9995756149292,
|
| 2100 |
+
"learning_rate": 2.1604938271604937e-06,
|
| 2101 |
+
"loss": 0.63,
|
| 2102 |
+
"step": 7250
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 2.91,
|
| 2106 |
+
"grad_norm": 10.930980682373047,
|
| 2107 |
+
"learning_rate": 2.1296296296296298e-06,
|
| 2108 |
+
"loss": 0.565,
|
| 2109 |
+
"step": 7275
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 2.92,
|
| 2113 |
+
"grad_norm": 10.784317970275879,
|
| 2114 |
+
"learning_rate": 2.0987654320987654e-06,
|
| 2115 |
+
"loss": 0.5634,
|
| 2116 |
+
"step": 7300
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 2.93,
|
| 2120 |
+
"grad_norm": 14.178128242492676,
|
| 2121 |
+
"learning_rate": 2.0679012345679015e-06,
|
| 2122 |
+
"loss": 0.5464,
|
| 2123 |
+
"step": 7325
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 2.94,
|
| 2127 |
+
"grad_norm": 14.49055004119873,
|
| 2128 |
+
"learning_rate": 2.037037037037037e-06,
|
| 2129 |
+
"loss": 0.6182,
|
| 2130 |
+
"step": 7350
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 2.95,
|
| 2134 |
+
"grad_norm": 9.5054349899292,
|
| 2135 |
+
"learning_rate": 2.006172839506173e-06,
|
| 2136 |
+
"loss": 0.6202,
|
| 2137 |
+
"step": 7375
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 2.96,
|
| 2141 |
+
"grad_norm": 11.535719871520996,
|
| 2142 |
+
"learning_rate": 1.9753086419753087e-06,
|
| 2143 |
+
"loss": 0.5973,
|
| 2144 |
+
"step": 7400
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 2.9699999999999998,
|
| 2148 |
+
"grad_norm": 12.069352149963379,
|
| 2149 |
+
"learning_rate": 1.944444444444445e-06,
|
| 2150 |
+
"loss": 0.5828,
|
| 2151 |
+
"step": 7425
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 2.98,
|
| 2155 |
+
"grad_norm": 10.448741912841797,
|
| 2156 |
+
"learning_rate": 1.9135802469135804e-06,
|
| 2157 |
+
"loss": 0.5961,
|
| 2158 |
+
"step": 7450
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 2.99,
|
| 2162 |
+
"grad_norm": 9.852815628051758,
|
| 2163 |
+
"learning_rate": 1.8827160493827162e-06,
|
| 2164 |
+
"loss": 0.6516,
|
| 2165 |
+
"step": 7475
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 3.0,
|
| 2169 |
+
"grad_norm": 15.387414932250977,
|
| 2170 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 2171 |
+
"loss": 0.5692,
|
| 2172 |
+
"step": 7500
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 3.01,
|
| 2176 |
+
"grad_norm": 7.3440070152282715,
|
| 2177 |
+
"learning_rate": 1.820987654320988e-06,
|
| 2178 |
+
"loss": 0.4318,
|
| 2179 |
+
"step": 7525
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 3.02,
|
| 2183 |
+
"grad_norm": 7.697348594665527,
|
| 2184 |
+
"learning_rate": 1.7901234567901235e-06,
|
| 2185 |
+
"loss": 0.457,
|
| 2186 |
+
"step": 7550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 3.03,
|
| 2190 |
+
"grad_norm": 11.595518112182617,
|
| 2191 |
+
"learning_rate": 1.7592592592592594e-06,
|
| 2192 |
+
"loss": 0.4364,
|
| 2193 |
+
"step": 7575
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 3.04,
|
| 2197 |
+
"grad_norm": 10.860459327697754,
|
| 2198 |
+
"learning_rate": 1.7283950617283952e-06,
|
| 2199 |
+
"loss": 0.4025,
|
| 2200 |
+
"step": 7600
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 3.05,
|
| 2204 |
+
"grad_norm": 8.72067642211914,
|
| 2205 |
+
"learning_rate": 1.697530864197531e-06,
|
| 2206 |
+
"loss": 0.4296,
|
| 2207 |
+
"step": 7625
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 3.06,
|
| 2211 |
+
"grad_norm": 9.136699676513672,
|
| 2212 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 2213 |
+
"loss": 0.4182,
|
| 2214 |
+
"step": 7650
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 3.07,
|
| 2218 |
+
"grad_norm": 9.509159088134766,
|
| 2219 |
+
"learning_rate": 1.6358024691358027e-06,
|
| 2220 |
+
"loss": 0.4468,
|
| 2221 |
+
"step": 7675
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 3.08,
|
| 2225 |
+
"grad_norm": 9.242144584655762,
|
| 2226 |
+
"learning_rate": 1.6049382716049383e-06,
|
| 2227 |
+
"loss": 0.4146,
|
| 2228 |
+
"step": 7700
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 3.09,
|
| 2232 |
+
"grad_norm": 9.915699005126953,
|
| 2233 |
+
"learning_rate": 1.5740740740740742e-06,
|
| 2234 |
+
"loss": 0.4493,
|
| 2235 |
+
"step": 7725
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 3.1,
|
| 2239 |
+
"grad_norm": 9.449831008911133,
|
| 2240 |
+
"learning_rate": 1.54320987654321e-06,
|
| 2241 |
+
"loss": 0.4995,
|
| 2242 |
+
"step": 7750
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 3.11,
|
| 2246 |
+
"grad_norm": 10.038348197937012,
|
| 2247 |
+
"learning_rate": 1.5123456790123458e-06,
|
| 2248 |
+
"loss": 0.4491,
|
| 2249 |
+
"step": 7775
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 3.12,
|
| 2253 |
+
"grad_norm": 10.652295112609863,
|
| 2254 |
+
"learning_rate": 1.4814814814814815e-06,
|
| 2255 |
+
"loss": 0.4216,
|
| 2256 |
+
"step": 7800
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 3.13,
|
| 2260 |
+
"grad_norm": 13.189153671264648,
|
| 2261 |
+
"learning_rate": 1.4506172839506175e-06,
|
| 2262 |
+
"loss": 0.4502,
|
| 2263 |
+
"step": 7825
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 3.14,
|
| 2267 |
+
"grad_norm": 9.493179321289062,
|
| 2268 |
+
"learning_rate": 1.4197530864197531e-06,
|
| 2269 |
+
"loss": 0.4084,
|
| 2270 |
+
"step": 7850
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 3.15,
|
| 2274 |
+
"grad_norm": 8.161423683166504,
|
| 2275 |
+
"learning_rate": 1.3888888888888892e-06,
|
| 2276 |
+
"loss": 0.441,
|
| 2277 |
+
"step": 7875
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 3.16,
|
| 2281 |
+
"grad_norm": 8.957440376281738,
|
| 2282 |
+
"learning_rate": 1.3580246913580248e-06,
|
| 2283 |
+
"loss": 0.3998,
|
| 2284 |
+
"step": 7900
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 3.17,
|
| 2288 |
+
"grad_norm": 10.819154739379883,
|
| 2289 |
+
"learning_rate": 1.3271604938271606e-06,
|
| 2290 |
+
"loss": 0.3827,
|
| 2291 |
+
"step": 7925
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 3.18,
|
| 2295 |
+
"grad_norm": 8.5838041305542,
|
| 2296 |
+
"learning_rate": 1.2962962962962962e-06,
|
| 2297 |
+
"loss": 0.4418,
|
| 2298 |
+
"step": 7950
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 3.19,
|
| 2302 |
+
"grad_norm": 9.781057357788086,
|
| 2303 |
+
"learning_rate": 1.2654320987654323e-06,
|
| 2304 |
+
"loss": 0.4375,
|
| 2305 |
+
"step": 7975
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 3.2,
|
| 2309 |
+
"grad_norm": 9.840452194213867,
|
| 2310 |
+
"learning_rate": 1.234567901234568e-06,
|
| 2311 |
+
"loss": 0.3952,
|
| 2312 |
+
"step": 8000
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 3.2,
|
| 2316 |
+
"eval_cer": 48.29247278138952,
|
| 2317 |
+
"eval_loss": 0.9025322794914246,
|
| 2318 |
+
"eval_runtime": 1733.8061,
|
| 2319 |
+
"eval_samples_per_second": 2.27,
|
| 2320 |
+
"eval_steps_per_second": 0.284,
|
| 2321 |
+
"step": 8000
|
| 2322 |
}
|
| 2323 |
],
|
| 2324 |
"logging_steps": 25,
|
|
|
|
| 2326 |
"num_input_tokens_seen": 0,
|
| 2327 |
"num_train_epochs": 4,
|
| 2328 |
"save_steps": 1000,
|
| 2329 |
+
"total_flos": 3.693893124096e+19,
|
| 2330 |
"train_batch_size": 8,
|
| 2331 |
"trial_name": null,
|
| 2332 |
"trial_params": null
|